Apple’s much-loved MacBook Air gets even more power, a much better webcam and an unexpected price cut for 2025, making one of the very best consumer laptops even more tempting.

The Guardian’s journalism is independent. We will earn a commission if you buy something through an affiliate link. Learn more.

The company’s thinnest and lightest laptop currently starts at £999 (€1,199/$999/A$1,699) – £100 less than last year’s model – and has Apple’s top M4 chip with a minimum of 16GB of memory, making the cheapest model much more capable.

Nothing has changed since 2022 on the outside other than a fetching new light blue colour that replaces the longstanding “space grey”. The aluminium body is just as well made, thin and light. The keyboard is excellent, the Touch ID fingerprint reader in the power button is fast and the generous trackpad is best in class.

The 13.6in LCD screen is crisp and good, although limited to 60Hz, unlike the faster and smoother screen on the MacBook Pro line and some rivals. New for this year is an upgraded webcam, which is now 12 megapixels and sports Apple’s Centre Stage tech for automatic panning and scanning, plus support for Desk View for demonstrating things remotely, similar to the M4 MacBook Pro.

View image in fullscreen The Centre Stage camera at the top of the screen is a significant upgrade for those who spend time on video calls. Photograph: Samuel Gibbs/The Guardian

Two meaningful changes have been made on the inside. The new M4 chip makes the Air essentially as fast as the MacBook Pro, which is to say very fast and snappy indeed. In tests it is only marginally slower than the Pro despite not having a fan to keep it cool for prolonged periods, and faster than the iPad Pro with M4 chip. There is little the Air will not be able to manage outside workstation-class workloads. Note the cheapest model has two fewer graphics cores on its chip, which will make it slightly less powerful for certain apps and games.

All Airs now also come with at least 16GB of memory, which is the recommended minimum amount for modern laptops. The previous model started with only 8GB of memory, which cost £200 to double.

Along with powerful chips, the Air’s main selling point – very long battery life – remains. For general light workloads using browsers, notes apps, word processing, image editing, chats and emails, the Air regularly lasts more than 16 hours between charges, which is roughly two full work days on battery. Pushed a bit harder, such as developing and editing photos, the Air still managed more than 10 hours’ work on battery.

View image in fullscreen A full charge using a 70W or greater power adaptor and the included MagSafe cable took about 105 minutes, hitting 50% in 28 minutes. Photograph: Samuel Gibbs/The Guardian

Specifications

Screen: 13.6in LCD (2560x1600; 224 ppi) True Tone

Processor: Apple M4 with eight or 10-core GPU

RAM: 16, 24 or 32GB

Storage: 256GB, 512GB, 1TB or 2TB SSD

Operating system: macOS 15 Sequoia

Camera: 12MP Centre Stage

Connectivity: wifi 6E, Bluetooth 5.3, 2x Thunderbolt/USB 4, headphones

Dimensions: 215 x 304.1 x 11.3mm

Weight: 1.24kg

Sustainability

View image in fullscreen The recycled aluminium lid of the MacBook Air in sky blue looks grey in some lights and blue in others. Photograph: Samuel Gibbs/The Guardian

The MacBook Air is made with 55% recycled materials including aluminium, cobalt, copper, glass, gold, lithium, plastic, rare-earth elements, steel and tin. Apple breaks down the computer’s environmental impact in its report.

The computer is generally repairable, and the battery can be replaced for £159 by Apple. The repair specialists iFixit gave the machine five out of 10 for repairability. Apple offers trade-in and free recycling schemes, including for non-Apple products.

macOS Sequoia 15.4

View image in fullscreen MacOS continues to be one of the best operating systems for laptops available, with plenty of customisation options and very useful new window tiling tools. Photograph: Samuel Gibbs/The Guardian

The Air runs the same macOS Sequoia software as the rest of the Mac lineup, recently updated to version 15.4, which brings with it a few more Apple Intelligence features.

The Mac Mail app has been redesigned to more closely resemble the app on an iPhone with AI email categorisation, such as promotional emails, receipts and others. AI notification summaries are useful, as is iPhone mirroring if you also use an Apple smartphone. The new snapping and window tiling tools are very handy, allowing you to quickly arrange your windows with either mouse or keyboard shortcuts, something that previously required third-party tools.

Price

The Apple 13in MacBook Air starts at £999 (€1,199/$999/A$1,699) with an 8-core GPU, 16GB of RAM and 256GB of storage. 15in versions start at £1,199 (€1,499/$1,199/A$2,099).

For comparison, the M4 MacBook Pro starts at £1,599, the Samsung Galaxy Book 4 Edge costs £1,399 and the Microsoft Surface Laptop 7 starts at £1,049.

Verdict

The MacBook Air is the consumer laptop to beat unless you specifically need Windows. It simply offers a better combination of performance, battery life, speakers, keyboard and trackpad than rivals.

The M4 chip upgrade keeps it ahead of the pack while being fanless and therefore silent at all times. The new Centre Stage webcam is a huge improvement, too, for anyone who lives on video calls for work. The screen is still great, although this is one area where better displays can be found on competitors at higher cost. It only has two USB-C ports but at least it can be charged via the MagSafe port to keep them available for connections.

A price cut to about £1,000 or equivalent for the 13in model is very welcome, as is the minimum of 16GB of memory, which makes the 2025 MacBook Air surprisingly good value for a premium notebook.

Pros: super-fast M4 chip, silent and cool running, extremely long battery life, good 13.6in screen, great keyboard, best-in-class trackpad, MagSafe, good speakers, Centre Stage webcam, Touch ID. Cons: only two USB-C ports and no USB-A or SD card slot, no Face ID, RAM and SSD upgrades are expensive and cannot be changed after purchase, no wifi 7.

Chip, memory and webcam upgrades are joined by welcome price cut for the top premium notebook

Apple MacBook Air M4 review: the laptop to beat, now cheaper

Black Mirror is more than science fiction – its stories about modernity have become akin to science folklore, shaping our collective view of technology and the future.

Each new innovation gets an allegory: smartphones as tools for a new age caste system, robot dogs as overzealous human hunters, drones as a murderous swarm, artificial intelligence as new age necromancy, virtual reality and brain chips as seizure-inducing nightmares, to name a few. Episodes most often channel our collective anxieties about the future – or foment new ones through masterly writing, directing, casting and acting. It is a must-watch, but must we take it so seriously?

Black Mirror fails to consistently explore the duality of technology and our reactions to it. It is a critical deficit. The show mimics the folly of Icarus and Daedalus – the original tech bros – and the hubris of Jurassic Park’s Dr Hammond. Missing are the lessons of the Prometheus myth, which shows fire as a boon for humanity, not doom, though its democratization angered benevolent gods. Absent is the plot twist of Pandora’s box that made it philosophically useful: the box also contained hope and opportunity that new knowledge brings. While Black Mirror explores how humans react to technology, it too often does so in service of a dystopian narrative, ignoring Isaac Asimov’s observation: that humans are prone to irrationally fear or resist technology.

Black Mirror is more pessimism porn than Plato’s parable, imparting to its audience a tacit lesson: fear the future more than the past. Fear too much technological change, not too little. It is an inherently populist narrative – one that appeals to nostalgia: intellectually we understand the present is better than the past in large part due to scientific and technological change, yet emotionally and instinctually we can’t help but feel this time in history is different, that the future can only get worse.

The heavy consequences of technological fearmongering

This kind of reductive dystopianism – a hallmark of post-1960s science fiction – clouds our thinking about the future because it “cannot imagine a better future, and it doesn’t ask anyone to bother to make one”, as New Yorker writer Jill Lepore noted in 2017.

We run from the speculative risks of the future, towards the proven dangers of the past, a dynamic I call the Frankenstein fallacy. This pessimistic archetype has bipartisan allure because “it requires so little by way of literary, political, or moral imagination”, according to Lepore; consequently, it is politically useful but unconstructive – as populism tends to be.

Technological pessimism will insure against a more dystopian tomorrow. Fear of genetically modified organisms – which kicked off around the release of Jurassic Park – has seen countries run from GMO food aid toward famine, from vitamin-enriched GMO “golden rice” towards malnutrition leading to millions of avoidable deaths. Countries such as the United States and Germany ran from a future of nuclear energy, towards coal and oil. In the Philippines, a nuclear plant built in the 1970s sits unused – never turned on – while its population deals with sky-high energy prices. All three countries are now trying to reverse course, realizing that the certainty of stasis and stagnation is its own form of dystopia. In contrast, France ran from the past towards the future, overcoming public fears of nuclear disasters, now getting 70% of its electricity from nuclear power.

Countries such as India, Brazil, Mexico and Thailand have run from vapes – outlawing them, while permitting traditional tobacco cigarettes for 1.8 billion of their citizens. Better unsafe than sorry.

In the US, Robert F Kennedy Jr runs from vaccines towards natural herd immunity – although he might be having second thoughts now that the risks have become less abstract. Bipartisan efforts have sought to remove online anonymity to protect children, forgetting that as adults they’ll lose the protection that anonymity brings in the context of free speech. Attacks on environmentally friendly lab-grown beef from Republicans have drawn support by Democrats such as John Fetterman.

In the UK, encryption is under siege, a modern-day promethean protection that angers the powerful, in the name of keeping society from runaway technology. Meanwhile, Adolescence is the latest dystopian Netflix show to shape public policy conversations about technology and the future.

Artificial intelligence has been touted as an existential threat to humanity while it accelerates cancer treatments, reduces sepsis deaths and produces new antibiotics to treat stubborn superbugs.

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

The folly of treating the miracles of science and technology as inevitable curses became unavoidable in the global coronavirus pandemic. The risk of inaction and stasis was too real to ignore, the absence of technology became the threat. Tellingly, it was in this period – between 2020 and 2022 – that Black Mirror went on hiatus. Charlie Brooker said this was because people didn’t want to consume dystopian fiction when everything felt so bleak. In a moment when screens kept us connected, protected and employed, the reductiveness of dystopian science fiction felt silly. Biotechnology like GMOs and mRNA offered existential hope, rather than risk.

Suddenly it became deeply uncool and unintellectual to fear technology – as 5G masts burned and Black Mirror-esque conspiracy theories of computer chips being injected through vaccines spread, dystopian fiction lost its allure, though its vestiges still lurked: Chipotle offered free burritos to the vaccinated, burritos it proudly markets as GMO-free. King Charles, who once warned that genetically modified organisms would cause the biggest environmental disaster of all time, would congratulate Oxford researchers for their GMO-based Covid vaccine. Publications like Scientific American would deplore pandemic conspiracy theories, when less than a year earlier it would amplify unfounded alarmism about 5G.

New parables

A new progressivism – one that embraces construction over obstruction, of pragmatism over precaution – must find new allegories to think about technology and the future. Stories that challenge a mindset the British prime minister, Keir Starmer, criticized for leading us to miss massive opportunities “because of the fears of small risk”.

We must move away from binary tales of catastrophe, not towards naive utopianism that ignores problems and risks that comes with change, but hopeful solutionism that reminds us we can solve and mitigate them. Stories that don’t make us forget that brain chips can liberate paraplegics, robot dogs can protect us from landmines, AI can prevent super bugs and VR can connect us rather than cut us off from reality – even if their vibes are “a bit Black Mirror”.

A new progressivism, one that embraces construction over obstruction, must find new allegories to think about technology and the future

Black Mirror’s pessimism porn won’t lead us to a better future | Louis Anslow

The global rush to AI technology will require almost as much energy by the end of this decade as Japan uses today, but only about half of the demand is likely to be met from renewable sources.

Processing data, mainly for AI, will consume more electricity in the US alone by 2030 than manufacturing steel, cement, chemicals and all other energy-intensive goods combined, according to a report from the International Energy Agency (IEA).

Global electricity demand from datacentres will more than double by 2030, according to the report. AI will be the main driver of that increase, with demand from dedicated AI datacentres alone forecast to more than quadruple.

One datacentre today consumes as much electricity as 100,000 households, but some of those currently under construction will require 20 times more.

But fears that the rapid adoption of AI will destroy hopes of tackling the climate crisis have been “overstated”, according to the report, which was published on Thursday. That is because harnessing AI to make energy use and other activities more efficient could result in savings that reduce greenhouse gas emissions overall.

Fatih Birol, the executive director of the IEA, said: “With the rise of AI, the energy sector is at the forefront of one of the most important technological revolutions of our time. AI is a tool, potentially an incredibly powerful one, but it is up to us – our societies, governments and companies – how we use it.”

Using AI could make it easier to design electricity grids to take more renewable energy. Most grids were designed for centralised fossil fuel power stations that produce reliable levels of electricity, some of which can be turned off and on relatively quickly. They have to be redesigned to balance demand when more of the supply comes from intermittent and sometimes unpredictable sources, such as wind and solar power.

Finding efficiencies within energy systems, and in industrial processes, could also become easier with AI. At present, huge opportunities to increase efficiency are missed, because it is harder for companies to change their processes than to carry on with wasteful practices.

AI could also assist with new technologies such as driverless vehicles or detecting threats to vital infrastructure. The technology could also be used to plan public transport to optimise for people’s journeys, or to design cities or traffic systems. Mining companies could use AI to discover and exploit reserves of critical minerals, which are crucial to modern renewable energy components such as solar panels, wind turbines and electric vehicles.

These uses could offset some of the massive demands that AI will place on the world’s energy systems. But that is likely to require greater direction from governments, the IEA report found. Left alone, the rapid growth of AI could prove a severe problem for energy systems and the environment.

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

AI has the potential to reverse all the gains made in recent years in advanced economies to reduce their energy use, mainly through efficiencies. The rapid increase in AI also means companies will seek the most readily available energy – which could come from gas plants, which were on their way out in many developed countries. In the US, the demand could even be met by coal-fired power stations being given a new lease on life, aided by Donald Trump’s enthusiasm for them.

Done badly, AI could also suck water from some of the world’s driest areas, an investigation by the Guardian revealed, as many AI datacentres use vast quantities of fresh water for cooling their computers.

Claude Turmes, a former Green MEP and energy minister for Luxembourg, said the disadvantages of AI were more likely to materialise than the optimistic projections of the IEA, and governments needed much more help to avoid the pitfalls.

He accused the IEA of painting too rosy a picture and failing to spell out harsh truths to policymakers. He said: “Instead of making practical recommendations to governments on how to regulate and thus minimise the huge negative impact of AI and new mega datacentres on the energy system, the IEA and its [chief] Fatih Birol are making a welcome gift to the new Trump administration and the tech companies which sponsored this new US government.”

The IEA forecast indicates a sharp rise in the requirements of AI, but said threat to the climate was ‘overstated’

Energy demands from AI datacentres to quadruple by 2030, says report

A $2bn (£1.6bn) British startup that uses artificial intelligence to generate realistic avatars has struck a licensing deal with the stock footage firm Shutterstock to help develop its technology.

Synthesia will pay the US-based Shutterstock an undisclosed sum to use its library of corporate video footage to train its latest AI model. It expects that incorporating the clips into its model will produce even more realistic expressions, vocal tones and body language from the avatars.

View image in fullscreen Synthesia licences actors’ likenesses for three years, having paid the performers in cash for up to six hours of work for the shoot. Illustration: Synthesia.io

“Thanks to this partnership with Shutterstock, we hope to try out new approaches that will … increase the realism and expressiveness of our AI generated avatars, bringing them closer to human-like performances,” said Synthesia.

The deal has been announced against a backdrop of tension between AI firms and the creative industries over use of copyright-protected material to train models without seeking permission from the content’s owner first.

The UK government is proposing to relax copyright law in order to allow tech companies to use that content without the owner’s consent – unless they declare an opt-out that blocks the AI firm from doing so – in a move that has drawn the ire of creative professionals.

Synthesia uses human actors to generate digital avatars of people, which are then deployed by companies in corporate videos in a range of scenarios such as advising on cybersecurity, calculating water bills and how to communicate better at work. Its UK clients include Lloyds Bank and British Gas. It is also used by non-corporate bodies including the NHS, the European Commission and the United Nations.

Last month Synthesia announced it was offering company stock to the actors it uses for its most popular avatars, in a move it said recognised the actors’ roles as the faces of the company. Synthesia licences actors’ likenesses for three years, having paid the performers in cash for up to six hours’ work for the shoot to create their avatar.

skip past newsletter promotion Sign up to Business Today Free daily newsletter Get set for the working day – we'll point you to all the business news and analysis you need every morning Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

View image in fullscreen Synthesia does not permit use of its stock avatars for political or news purposes. Illustration: Synthesia.io

Synthesia, which does not permit use of its stock avatars for political or news purposes, will not be using the people in the Shutterstock footage as avatars. Instead, it will use the footage to help its model better understand body language and workplace interactions and environments, such as modelling how people talk when they are sitting behind a desk or in front of a whiteboard.

London-based Synthesia, founded in 2017 by two Danish entrepreneurs and two academics, secured a valuation of $2.1bn this year in a funding round that raised $180m.

Beeban Kidron, a cross-bench peer and prominent critic of the government’s copyright plans, said the deal to licence Shutterstock’s content was proof that the government’s position on copyright was “wrong”.

“I hope this announcement helps the UK government come to a more sophisticated understanding about where the UK’s interests lie,” she said.

The government has said the copyright regime for copyright and AI is holding back the creative industries, media and tech sector from realising their full potential and “cannot continue”.

British startup will license content to train latest model for more ‘human-like performances’ from its avatars

AI avatar generator Synthesia does video footage deal with Shutterstock

The UK communications regulator has announced its first investigation under the new digital safety laws with an inquiry into an online suicide forum.

Ofcom is investigating whether the site breached the Online Safety Act by failing to put in place adequate measures to shield its users from illegal content.

The law requires tech platforms to tackle illegal material – such as encouraging suicide – or face the threat of fines of up to £18m or 10% of global revenue. In extreme cases, Ofcom also has the power to block access to a site or app in the UK.

Ofcom, which is not naming the forum under investigation, said it was focusing on whether the site had put appropriate measures in place to protect UK users, whether it had failed to complete an assessment of the harms the site could cause, as required under the legislation, and whether it had responded adequately to a request for information.

“This is the first investigation opened into an individual online service provider under these new laws,” said Ofcom.

The BBC reported in 2023 that the forum, easily accessible to anyone on the open web, had been connected to at least 50 deaths in the UK and had tens of thousands of members with discussions including methods of suicide.

Last month, obligations came into force under the act requiring the 100,000 services under its scope – from small sites to big platforms such as X, Facebook and Google – to implement safeguards that take action against illegal harms. The legislation lists 130 “priority offences”, or illegal content, that tech companies must tackle as a priority by ensuring their moderation systems are set up to deal with such material.

“We’ve been clear … that failure to comply with the new online safety duties or adequately respond to our information requests may result in enforcement action, and that we would not hesitate to take swift action where we suspect there may be serious breaches,” said Ofcom.

The Molly Rose Foundation, an internet safety charity, called on Ofcom to act “quickly and decisively” and shut down the site. MRF was established by the family of Molly Russell, a British teenager who took her own life after viewing online content related to suicide, depression and self-harm.

skip past newsletter promotion Sign up to First Thing Free daily newsletter Our US morning briefing breaks down the key stories of the day, telling you what’s happening and why it matters Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

“We welcome Ofcom taking the first steps to clamp down on this horrendous site, which exists only to help vulnerable people end their lives,” said Andy Burrows, the hief executive of MRF. “Every day it remains online more vulnerable people are at risk.”

Ofcom’s first investigation to look into whether site took adequate measures to shield users from illegal content

Online suicide forum investigated under new UK digital safety laws

The Australian government spent nearly $3m of taxpayer dollars advertising on Twitter/X in the first year after the world’s richest man, Elon Musk, took over the platform, despite warnings of brand reputation damage that caused the government to initially pause ads.

Data obtained by Guardian Australia, after a protracted freedom of information battle with the federal finance department, revealed $2.7m was spent between November 2022 and November 2023. Musk finalised his purchase of the platform on 28 October 2022.

The data covers all ad spending on government department campaigns including health advisories and messaging around budget initiatives. In 2022-2023 this included advertising around the voice referendum and vaccine advertisements, among others.

The spending comes despite the government pausing ad spend on what was then called Twitter on 29 September 2022, following reports of a number of brands’ advertisements appearing next to inappropriate content on Twitter, a report from the Australian National Audit Office revealed last week.

A spokesperson for the finance department said it was suspended “to evaluate brand safety measures” and restored shortly after.

“Following further review, the government’s master media agency that manages advertising placement advised that sufficient brand safety measures were in place and advertising could return to the platform on 5 October 2022 with the exception of placements on Twitter profiles,” the spokesperson said.

The following year’s data – accounting for the shift when Musk became a full-throated supporter of Donald Trump – is being withheld until January 2026. The department said in its decision that releasing the data now would weaken the government’s ability to negotiate with media on advertising.

Sources have told Guardian Australia the total ad spend for the year is similar to 2022-23.

Axel Bruns, professor in the Digital Media Research Centre at Queensland University of Technology, said the spend could reflect the uncertainty the government had around whether the platform would continue to be popular under Musk’s leadership.

“With hindsight, there now seems little value in government advertising on X – the platform is overrun with disinformation, abuse, hate speech and spam bots, and most sensible users will have left by now,” he said.

“This might not have been as clear at the time when this advertising was purchased, though.”

Bruns said any continued government advertising on X is “unlikely to make its desired impact on audiences” but could be seen to be soft diplomacy to win favour with the Trump administration, but said it would not be a “useful investment” given the reputation damage to continue to advertise on X in its current form.

The continued ad spend comes despite Musk describing the Albanese government as “fascist” over its now-abandoned misinformation legislation, and engaging in several legal battles with the Australian online safety regulator over content moderation on X, one of which was heard in the administrative review tribunal last week.

Musk has become inexorably entwined with Trump and subsequently the new US administration in the past few months. Prior to running the US government’s department of government efficiency (Doge) tasked with slashing government programs and spending, Musk became Trump’s biggest campaign donor in the 2024 presidential election, and invested millions in his failed attempt to have a conservative judge win the Wisconsin supreme court’s election last week.

Musk’s personal popularity has tanked this year, as a result, with a recent poll finding more than half of Americans believed Musk and Doge were harming the US.

The finance department, which collects data from every government agency on ad spending, does not typically break down spending on individual platforms. Total digital ad spending for 2022-2023 was $56.3m. Guardian Australia filed a freedom of information request with the finance department for data specifically on X spending last year, but was initially refused.

After seeking a review of the decision by the Australian information commissioner, the finance department agreed to release the first year of data this week.

X was contacted for comment.

Exclusive: Spending came after the Albanese government paused ads for a week amid reports ads were appearing next to inappropriate content

Australian government gave $2.7m to Elon Musk’s X for advertisements in billionaire’s first year as owner

The EU has revealed details of a €20bn (£17bn) plan to create new sites equipped with vast supercomputers in Europe to develop the next generation of artificial intelligence models, while opening the door to amending its landmark law that regulates the technology.

Publishing a strategy to turn Europe into an “AI continent”, the European Commission vice-president Henna Virkkunen said the technology was at the heart of making Europe more competitive, secure and technologically sovereign, adding: “The global race for AI is far from over.”

The EU is attempting to catch up with the US and China, which have taken a lead in pioneering the technology that increasingly powers shopping websites and self-driving cars, generates text, and is predicted to play a transformative role in healthcare, security, defence and advanced manufacturing, among other sectors.

The US has a commanding lead in AI, far ahead of China. A report from Stanford University this week said 40 “notable AI models” – meaning influential – were produced by institutions in the US in 2024, compared with 15 in China and three in Europe (all French).

In a separate 2024 report, Stanford found that no EU country made the top five for “vibrancy” in AI, a metric that considered private investment, patents and research. It ranked the UK in third place behind the US and China, with France in sixth place and Germany eighth.

The EU has already embarked on a plan to build 13 AI factories – sites with supercomputers and datacentres, where researchers develop and test AI models.

The new AI “gigafactories” would be much larger, targeting what the commission called “moonshots”: significant innovations in healthcare, biotech, industry, robotics and scientific discovery.

While the best-performing AI factories have supercomputers equipped with up to 25,000 advanced AI processors, a gigafactory would exceed 100,000 AI processors, the strategy document said.

These power-hungry facilities, which can require huge amounts of water for cooling, should run “as much as possible” on a green energy supply, an EU official said, with plans for “recycling” water if it was used.

Last year 47% of electricity generated in the EU came from renewable sources, but campaigners fear that power-hungry datacentres could undermine Europe’s climate ambitions.

“Europe has made some strides towards fully renewable electricity, and still has a long way to go, but this could all be undone by ballooning energy consumption,” said John Hyland, a Greenpeace spokesperson. “Every power-guzzling datacentre risks being a lifeline to polluting power plants, well illustrated in Ireland where they consume over a fifth of the electricity.”

skip past newsletter promotion Sign up to Headlines Europe Free newsletter A digest of the morning's main headlines from the Europe edition emailed direct to you every week day Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

EU officials envisage three to five AI gigafactories in the EU at a cost of around €3-5bn each, compared with €600m for the biggest AI factory. The commission is seeking to raise €20bn, using public money to incentivise private investors, via a facility provided by the European Investment Bank. The bloc is also seeking to develop European-made AI semiconductors, advanced chips that are vital for AI factories.

But the commission faced criticism for indicating an openness to rethink its AI Act. The regulation, which aims to control the risks posed by AI, was approved by the European parliament only last year and will not be fully applicable until August 2027. On Wednesday, the commission said it was launching a consultation on “possible simplification” of the AI Act, as part of a wider drive to cut redtape amid concerns about Europe’s lagging economic growth.

The European Consumer Organisation (BEUC) said the commission needed to focus on the implementation and enforcement of the AI Act. “Much of the AI Act does not even apply yet but the commission is already signalling that it is open to cutting some of its requirements. When we know that consumers worry about how AI could be used to harm them, the commission should instead be trying to reassure consumers,” said Frederico Oliveira da Silva at BEUC.

Virkkunen said the commission was “very committed” to the risk-based approach and the main principles of the AI Act, but was looking at whether there were some administrative burdens or reporting obligations that could be cut to make life easier for businesses.

Up to five sites with power-hungry supercomputers and datacentres planned to drive AI ‘moonshots’

EU to build AI gigafactories in €20bn push to catch up with US and China

Tariffs that took effect in the US at midnight on Wednesday are expected to make buying new smartphones more expensive. But opting to fix an old or damaged device as a way to save money may not avoid a higher bill.

“We do expect that we’ll have to increase our own parts prices some, too, unfortunately,” said Elizabeth Chamberlain, the director of sustainability at the device repair company iFixit. “We’re actively looking for sources of harvested parts from within the US. Even with higher prices on parts, however, repairs will still be much more economical than buying new.”

Donald Trump’s tariffs are likely to drive up the cost of smartphone repairs because of the globalized supply chain that produces the devices’ components. The vast majority of parts for popular phones from Apple and Samsung are produced outside of the US. iPhones are primarily manufactured and assembled in China, where businesses that export to the US are bracing for a tariff of over 100%. India, where Apple has shifted some of its production and where Google in part manufactures its Pixel line of phones, is facing a 26% tariff. Samsung’s supply chain is largely based in South Korea, which may see a 25% tariff if the government isn’t able to reach a deal with Trump’s administration. The companies did not respond to requests for comment on potential price increases.

Tariffs may increase demand for phone repairs because a single part, even at a higher price, will remain cheaper than a complete device. The tariffs could drive up the final price of the latest iPhone by nearly $300, according to the Wall Street Journal, which makes repairing the devices people already own a more viable alternative, Chamberlain said.

“It’s a bit too soon to say whether the tariff news is driving more repair demand, but fixing what we’ve got makes more sense than ever, now that buying new will be more expensive,” she said. “We expect the tariffs might also drive demand for refurbishing and parts harvesting. The repair world is full of business models that don’t rely on importing stuff from overseas.”

Increasing price of parts

Repair shops big and small are preparing for the prices of imported parts to go up. A manager of a Brooklyn location of a national repair chain, who asked not to be named because he was not authorized to speak to the press, said that his business is anticipating a 20% increase for many of the parts required to repair devices.

Dan Fernando, the owner and operator of Technesia, an independent repair shop run out of his Philadelphia home, said he is already seeing some price fluctuations for parts like a particular hard drive he often uses to repair computers. Within the past few days, Fernando has seen the price of the one-terabyte drive rise by $10, an increase that will be passed down to his customers, he said. The electronics repairman charges a flat service fee for each repair, and customers are expected to pay for the cost of a part if they need to get a replacement, a common business model in electronics repair. Fernando sources the parts from a phone repair supplier called MobileSentrix, which imports parts to the US from China and other countries.

“For a screen replacement on phones, I charge $50 flat for the service plus the cost of the screen,” he said. “The customer can expect to pay $80 to $90 for a screen replacement. With the new tariff increase, it will maybe cost 50% more.”

Fernando is also concerned customers will turn to shops that offer cheaper repairs that may begin to use lower grade parts to maintain their profit margins, leaving his business in the lurch. “Some people get it from eBay or Amazon. I don’t use it because I don’t know their quality,” he said.

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

Appeal of secondhand devices

With both new devices and repairs becoming pricier, secondhand device marketplaces like Swappa present a third option to concerned customers. The biggest sellers on sites for pre-owned electronics are phones two or three generations behind the latest. At the moment, the best selling devices on Swappa are the iPhones 13 and 14, according to the site’s founder and CEO, Ben Edwards.

“My expectation is that, if these tariffs do remain in place and we do see even further rise in cost of new devices, I think that’ll really drive a lot of new buyers to Swappa and similar sites,” said Edwards. “I’m not sure there’s enough technological advantage with the newest generation.”

But even Swappa may see prices on refurbished devices increase. The site allows sellers – which include both individuals and companies – to set their own prices and provides them with data on previous sales and current demand. When the cost of the newest smartphones go up, sellers often raise the prices of these used devices in response, Edwards said.

“Essentially, demand is ultimately going to kind of dictate the price,” Edwards said. “On Swappa, because of the nature of our marketplace, where we’ve got a lot of different sellers who can each set their price independently, it’s really just market dynamics that are going to dictate the price. If somebody comes in and prices aggressively, they’re going to sell more quickly and that’ll reflect itself in the data.”

In other words, if a consumer is in the market to buy a used device, Edwards advised: “I wouldn’t wait.”

Business owners warn taxes may increase parts prices, making electronics repairs more costly than before

Why you can’t just repair your phone in the US to avoid Trump tariffs

A record number of British drivers are passing their driving test first time while committing zero faults.

Official figures released by the Driver and Vehicle Standards Agency (DVSA) show more than 20,000 drivers received a perfect score when taking their practical test last year – the highest number since figures were first published in 2007-08.

At that time, one in 200 people left their driving instructor surprised by their parallel parking and blind spot awareness. But last year, one in 43 passed with flying colours.

The figures come as the DVSA tries to reduce a record backlog in driving tests caused by the pandemic, which means learner drivers are currently waiting an average of five months to take a test.

The figures show a sharp rise in flawless first tests since the pandemic. The zero-flaw rate increased from 1.4% to 1.9% in 2020-21, and has since risen to 2.3% of first tests in 2023-24.

Several test centres have seen a huge increase in zero fault first attempts. The driving centre in the port town of Barrow-in-Furness saw just under 12% of people pass for the first time with no faults last year – up from just 1% in 2019. Arbroath and Montrose test centres in the north-east of Scotland saw 15% and 12% of first test attempts pass with no faults respectively – up from 7% and 5% in 2019-20.

As well as a rise in flawless first attempts, the figures show a steady increase in the number of people taking tests for automatic cars. Almost a quarter of practical tests – 455,276 – were for an automatic car last year – more than double the number in 2019-20, and almost four times (376%) higher than a decade ago.

The rise in automatic cars doesn’t appear to be linked to a rise in better motorists. Overall figures show just 43% of automatic drivers passed their test in 2023-24, compared with 48% of all drivers (manual or automatic). The government does not publish first test or zero fault figures for automatic cars.

Driving tests were suspended during the Covid-19 lockdowns, resulting in a backlog that is yet to be dealt with. Figures obtained by the AA found that the number of test centres showing the maximum wait of 24 weeks doubled from 94 in February 2024 to 183 this year – with the average driver now waiting 20 weeks for a test.

In December, the Public and Commercial Services Union – which represents the vast majority of driving examiners – reported that its members were being “bullied” by senior DVSA employees to inflate pass rates.

In a submission to the House of Commons transport select committee, they wrote that examiners felt pressured to take out unroadworthy vehicles, and were being encouraged to scope out which test routes have higher pass rates.

The DVSA has been approached for comment.

More than 20,000 drivers received perfect score when taking practical test last year

Record number of UK drivers pass test first time with zero faults, figures show

Andrew Vincent makes a good point in that, very often, artists are already expected to behave like artificial intelligence (Letters, 6 April). But of course creativity is not simply a matter of training on the work of others. Innovative artists make decisions towards low-probability outcomes; imitation, meanwhile, seeks high-probability outcomes.

As things stand, generative AI models are imitation engines – and they do not celebrate their sources, they conceal them. Writers carry forward ideas and techniques, yes, but an immeasurable part of human creativity comes from the certain knowledge that we will one day die. AI does not have that gift. For all it consumes, it does not choose what to remember or believe or feel. Authors are as much up in arms about the extreme-capitalist assumption that we’re simply machines, regurgitating content, as we are about the systemic theft of our work. Human authors also tend to worry about the difference between inspiration and plagiarism. AI has not yet been programmed to care, and no one’s holding their breath on that count.

Timothy X Atack

Bristol

Andrew Vincent says creativity has always “trained” on the work of others. He is right. Reading widely helps my writing, and art influences art. But I’m not objecting to Meta using my work for AI training; I’m objecting to Meta having done so without permission or payment, for commercial gain and outside the scope of fair use. I buy the books I need to read, or borrow them from a proper library, so that the authors get their fair library payment and their intellectual property is respected. Meta has stolen far more books than any single author could read – around 7 million. Please let’s consider properly the long-term consequences of allowing big, wealthy tech to flout legal protections and steal intellectual property from artists.

Abie Longstaff

Chair of the Children’s Writers and Illustrators Group, Society of Authors

Letters: Timothy X Atack thinks AI models are imitation engines – and they do not celebrate their sources, they conceal them. Abie Longstaff says Meta has stolen far more books than any author could read

AI doesn’t care about authors, but Meta should

Brandon Williams, President Trump’s pick to become the keeper of the nation’s nuclear arsenal, testified on Tuesday that he would not recommend that Mr. Trump restart explosive testing of the deadly weapons.

His statement, during a confirmation hearing before the Senate Armed Services Committee, was unexpected. Other advisers to the administration had proposed that the president resume the test detonations for the sake of national security. The last such explosion in the United States was in 1992.

Mr. Trump nominated Mr. Williams, a former Navy officer and one-term congressman from upstate New York, in January to serve as administrator of the National Nuclear Security Administration, which runs the nation’s atomic weapons complex. It is a semiautonomous agency under the umbrella of the Energy Department.

The most prominent ally of Mr. Trump to call for a testing restart is Robert C. O’Brien, who served as his national security adviser from 2019 to 2021. Last summer in Foreign Affairs magazine, he urged that Mr. Trump, if elected to a new term, resume the detonations, arguing that such a move would help the United States “maintain technical and numerical superiority to the combined Chinese and Russian nuclear stockpiles.”

Brandon Williams, the nominee to lead the National Nuclear Security Administration, said he would recommend reliance on “scientific information” rather than a restart of explosive testing.

Nuclear Testing Not Advised, Trump’s Nominee Says in Senate Hearing

Medieval scribes filled volumes called bestiaries with illustrations and descriptions of fantastic creatures. The manuscripts containing representations of these animals also depended on a menagerie of beasts: The covers of these and other volumes were fashioned from the skins of calves, goats, sheep, deer, pigs and, in some macabre instances, humans.

Most of these hides were shorn before they were turned into book bindings. But one set of medieval manuscripts from northeastern France has a peculiar finish: Its weathered covers are covered in clumps of hair.

“These books are too rough and far too hairy to be calfskin,” said Matthew Collins, a bioarchaeologist at the University of Copenhagen and Cambridge University and an author of the new study. But identifying the source of the shaggy leather has proved difficult.

While these furry tomes would seem at home in Hogwarts library, they were originally made in the scriptorium of Clairvaux Abbey, a hub for an order of Catholic monks, the Cistercians. The abbey, founded in 1115 in the Champagne-Ardenne region of France, was home to one of the largest monastic libraries in medieval Europe.

The material on the covers of books from a French abbey was too hairy to have come from calves or other local mammals. Researchers identified its more distant origin.

The Skin on Mysterious Medieval Books Concealed a Shaggy Surprise

So sharp are partisan divisions these days that it can seem as if people are experiencing entirely different realities. Maybe they actually are, according to Leor Zmigrod, a neuroscientist and political psychologist at Cambridge University. In a new book, “The Ideological Brain: The Radical Science of Flexible Thinking,” Dr. Zmigrod explores the emerging evidence that brain physiology and biology help explain not just why people are prone to ideology but how they perceive and share information.

This conversation has been edited for clarity and brevity.

What is ideology?

It’s a narrative about how the world works and how it should work. This potentially could be the social world or the natural world. But it’s not just a story: It has really rigid prescriptions for how we should think, how we should act, how we should interact with other people. An ideology condemns any deviation from its prescribed rules.

You write that rigid thinking can be tempting. Why is that?

Ideologies satisfy the need to try to understand the world, to explain it. And they satisfy our need for connection, for community, for just a sense that we belong to something.

There’s also a resource question. Exploring the world is really cognitively expensive, and just exploiting known patterns and rules can seem to be the most efficient strategy. Also, many people argue — and many ideologies will try to tell you — that adhering to rules is the only good way to live and to live morally.

In her new book, “The Ideological Brain,” the neuroscientist Leor Zmigrod outlines what makes some people prone to rigid thinking.

Ideology May Not Be What You Think but How You’re Wired

Meta is expanding its safety measures for teenagers on Instagram with a block on livestreaming, as the social media company extends its under-18 safeguards to the Facebook and Messenger platforms.

Under-16s will be barred from using Instagram’s Live feature unless they have parental permission. They will also require parental permission to turn off a feature that blurs images containing suspected nudity in their direct messages.

The changes were announced alongside the extension of Instagram’s teen accounts system to Facebook and Messenger. Teen accounts were introduced last year and placed under-18s by default into a setting that includes giving parents the ability to set daily time limits for using the app, to block teenagers from using Instagram at certain times and to see the accounts with which their child is exchanging messages.

Facebook and Messenger teen accounts will be rolled out initially in the US, UK, Australia and Canada. As with the Instagram accounts, users under the age of 16 will need parental permission to change the settings, while 16 and 17-year-olds defaulted into the new features will be able to change them independently.

Meta said the Instagram teen accounts were used by 54 million under-18s around the world, with more than 90% of 13- to 15-year-olds keeping on their default restrictions.

The NSPCC, a leading child protection charity, said it welcomed extending the measures to Facebook and Messenger, but said Meta had to do more work to prevent harmful material appearing on its platforms.

“For these changes to be truly effective, they must be combined with proactive measures so dangerous content doesn’t proliferate on Instagram, Facebook and Messenger in the first place,” said Matthew Sowemimo, the associate head of policy for child safety online at the NSPCC.

The announcement was made as the UK implements the Online Safety Act. Since March, every site and app within the scope of the legislation, which covers more than 100,000 services from Facebook, Google and X to Reddit and OnlyFans, is required to take steps to stop the appearance of illegal content such as child sexual abuse, fraud and terrorism material, or to take it down if it goes online.

The act also contains provisions for protecting children from harm and requires tech platforms to shield under-18s from damaging material such as suicide and self-harm-related content. Reports last week that the act could be watered down as part of a UK-US trade deal were met with protests from child safety groups, which said any compromise would be an “appalling sellout” that would be rejected by voters.

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

Speaking at the time the Instagram restrictions were launched, Meta’s then president of global affairs, Nick Clegg, said the aim was to “shift the balance in favour of parents” when it came to using parental controls. The announcement had come days after Clegg said parents tended not to use child safety measures.

Under-16s will be barred from using the app’s Live feature unless they have parental permission

Meta blocks livestreaming by teenagers on Instagram

The former boss of Google, Eric Schmidt, has bought a Holland Park mansion for nearly £42m, in the latest in a string of big transactions in London’s prime real estate market.

Schmidt, who was the chief executive of Google from 2001 to 2011, bought the double-fronted stuccoed mansion in west London, last May, according to the Financial Times.

The Grade II-listed building also comes with mews houses at the back and was last bought in 2022 for £36.2m, according to official records at the Land Registry. Its sale is the latest in a series of high-profile deals in London’s prime real estate market, after the recent £139m sale of a 40-bedroom mansion inside Regent’s Park.

The ex-Google chief executive, who with his wife owns a string of properties, plans to rent out the property, the FT reported. He is one of many wealthy Americans investing in top-end London property. Last year, Americans overtook Chinese buyers in the capital, accounting for 9.3% of overseas buyers, compared with 5.6% in 2019, according to research by the real estate agent Knight Frank.

The number of Americans applying for UK citizenship also hit a record high last year, after Donald Trump’s bid for the presidency and the abolition of the non-dom tax status in the UK. More than 6,100 US citizens applied for UK citizenship, marking a 26% increase from 2023 and the highest number since data collection began in 2004.

While there have recently been a few large deals at the most expensive end of the market in London, activity overall has been slowing. Last year, the number of London homes selling for more than £5m stood at 443, a drop of 16% against the 525 recorded in 2023, according to research by the estate agent Savills. However, it still stands well above pre-pandemic levels of 308 in 2019.

Overall price growth in the capital has been relatively low. London recorded the lowest annual price growth across the UK in February, according to Nationwide building society, at 1.9%, compared with a national rate of 3.9%. That was led by 13.5% in Northern Ireland, the highest in the region since 2021. Scotland posted a 3.9% rise, with Wales at 3.6%. However, London remained the most expensive place to buy a home, at an average price of £529,369, Nationwide said.

skip past newsletter promotion Sign up to Business Today Free daily newsletter Get set for the working day – we'll point you to all the business news and analysis you need every morning Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

A spokesperson for Schmidt said: “Eric invests in high-end real estate properties around the world.”

Deal to buy Holland Park home, which ex-CEO reportedly plans to rent out, is one of string by Americans in the capital

Ex-Google boss Eric Schmidt buys £42m London mansion

John Hinkley says artificial intelligence “is a tool – like a piano is to a composer” (Letters, 31 March). I don’t find this a realistic comparison. No matter how many times a pianist uses her piano to compose or to play her own or others’ works on it, the piano will never appear alone on stage, playing its own compositions and taking the pianist’s place. It seems likely that this does not hold for AI within the creative industries.

Jill Wallis

Aston Clinton, Buckinghamshire

A general leaflet from Reform UK soliciting my vote at next month’s county council elections smuggled itself into my abode recently. It grumbles about inadequacies of bin collection. I, sadly, am one of those nerdish types who knows that the county council doesn’t do the bin collection. Maybe if Reform is really serious about running local services then it should recruit more nerds? (This is not a job application.)

Michael Loftus

Kidderminster, Worcestershire

Much as I like Polly Hudson’s arguments for the “French exit” (7 April), it amuses me that the equivalent expression here is filer à l’anglaise (to leave in the English way), which, much to my chagrin, is considered most impolite.

David Green

Paris, France

More nominative determinism in the Guardian: the Harris’s hawk that harassed Flamstead was caught by Steve Harris (Report, 3 April). Thus, Harris’s hawk is a Harris’s hawk.

Noel Privett

Whitchurch, Hampshire

F is for Farage in the cost of living ABC (Letters, 6 April).

Peter Gray

Chesterfield, Derbyshire

Brief letters: Tech in creative industries | Reform rubbish | The English exit | Harris’s hawk | The F word

AI piano analogy does not play well for me

Three young children huddle in front of a camera, cross-legged and cupping their hands. “Please support me. We are very poor,” says a boy, staring down the lens.

They appear to be in a mud-brick hut in Afghanistan, living in extreme poverty. But their live stream is reaching viewers in the UK and worldwide – via TikTok Live.

For hours, they beg for virtual “gifts” that can later be exchanged for money. When they get one, they clap politely. On another live stream, a girl jumps up and shouts: “Thank you, we love you!” after receiving a digital rose from a woman in the US, who bought it from TikTok for about 1p. By the time it’s cashed out it could be worth less than a third of a penny.

TikTok says it bans child begging and other forms of begging it considers exploitative, and says it has strict policies on users who go live.

But an Observer investigation has found the practice widespread. Begging live streams are actively promoted by the algorithm and TikTok profits from the content, taking fees and commission of up to 70%.

Olivier de Schutter, the UN special rapporteur on extreme poverty and human rights, called the trend a “shocking development” and accused TikTok and middlemen of “profiting from people’s misery”. “Taking a cut of people’s suffering is nothing short of digital predation. I urge TikTok to take immediate action and enforce its own policies on exploitative begging and seriously question the ‘commission’ it is taking from the world’s most vulnerable people,” he said.

Jeffrey DeMarco, digital harm expert at Save the Children, said: “The documented practices represent significant abuses and immediate action must be taken to ensure platforms no longer allow, or benefit directly or indirectly, from content such as this.”

Analysis conducted between January and April 2025 found evidence of live-begging and related behaviour in countries including Indonesia, Pakistan, Afghanistan, Syria, Egypt and Kenya.

Many of the live streams showed families begging in domestic settings. But some streams appeared to bear the hallmarks of organised begging.

One account, which broadcasts almost every day, regularly features different children. During a broadcast in February, it showed seven young boys begging for digital gifts. The next day, different boys appeared in the same spot, sitting alongside the same two adults. After being alerted to the account by the Observer, TikTok suspended it within two hours. A previous report through the in-app reporting tool had not resulted in any action.

Another account with 5,300 followers featured an elderly man in a wheelchair. The account appears to be run by a third party but there is no information about their identity. The account description says “poor people” followed by crying face emojis.

Beyond begging, live streams have featured people doing degrading and sometimes dangerous stunts in exchange for virtual gifts, such as hitting themselves, staying awake for long periods, covering themselves in mud, or sleeping on camera. One Indonesian live stream showed two young girls lying on a tiled floor in a small, windowless, white-walled studio. On Friday, the same account was broadcasting a live stream of a man asleep on the same tiled floor.

Another broadcast from Pakistan showed three men sitting in the dark with plastic buckets and party hats on their heads. They appeared to fall asleep on camera, only waking when they received a gift to perform a dance.

Launched in August 2020, TikTok Live is a section of the app where creators can broadcast to viewers in real time. According to TikTok, 100 million people went live in 2024, reaching “billions of users”, who can engage with the creator in real time by sending comments or gifts. Streams are promoted to TikTok users through a dedicated live feed and can reach millions of people.

Novita Anggraeni, an Indonesia-based social inclusion expert at the charity Care International, said many creators used TikTok Live to “show their talent” or “connect with people”. It is commonly used by people who film themselves singing, dancing, playing video games, painting or cooking. But, she said, it was increasingly being misused by those who saw it as a “shortcut to making money”, and that “harmful” videos showing begging and related behaviour had “very rapidly grown”. “It’s a whole new form of exploitation,” she added.

She said the streams were driven by poverty and poor digital literacy and that it was not always easy to identify whether people were being coerced. She said apparent victims often denied they were being exploited and said it was “a collaboration to earn money”. But she said there was a “high risk” of abuse by organised begging networks which control access to earnings and organise filming of multiple families.

0:30 Livestreams show children begging on TikTok – video

Marwa Fatafta, from the digital rights organisation Access Now, said the way TikTok Live was designed incentivised begging and risky behaviour because engagement boosts rewards. “At the same time, TikTok is not making enough effort to ensure that this unintended outcome is being addressed,” Fatafta said.

She warned against a blanket approach to moderating begging content, which could penalise people needing humanitarian support in “crisis or conflict-stricken areas”, and is concerned that in some regions, online begging has led to a repressive response from governments.

But she said more needed to be done to “tackle organised begging and exploitation”, and urged TikTok to “hire investigative teams”. “At this point, exploitation for organised begging is not only a foreseeable risk, it’s an existing one. So the platform needs to take care of it,” she said.

This weekend, TikTok said it had taken firm action, including removing accounts flagged by the Observer. It said its policies banned exploitative begging during live streams, including begging that “exploits children or vulnerable people”. “Any live content which features children begging for gifts is not allowed on TikTok,” a spokesperson said.

In cases where it had evidence of exploitative begging or child exploitation, it said it took “firm action”, including closing live streams and issuing permanent bans. It said that through “proactive detection by dedicated teams and technology” it stopped more than 4m live streams each month “to keep our platform safe”. People hosting live streams must have 1,000 followers and be over 18, though children can appear if accompanied by an adult.

It said the amount of commission and fees it took varied. Live streamers do not receive the total value of the gift but are awarded “diamonds” based on gifts and the popularity and duration of their stream. After commission and fees, they are typically left with about half the original gift value, but sometimes as little as 30%. TikTok did not deny the figures but said about 30% of gifts revenue went on “app store fees and payment provider costs”. It offers more than 100 digital gifts that viewers can buy, ranging from a rose costing one coin (about 1p) to a TikTok universe (44,999 coins, or about £450).

The company has previously faced criticism for profiting from exploitative content on TikTok Live, including livestreamed sexual abuse. In 2022, the BBC found families in Syrian refugee camps begging for money. TikTok said it was “deeply concerned” by the findings and was strengthening its global policies.

In 2023, Al Jazeera reported that orphanages in Indonesia were soliciting donations via live streams showing children sleeping in the background. TikTok said it banned begging in demeaning contexts but that the orphanage live stream did not violate its guidelines.

Despite the risks associated with begging on TikTok, charities say there are also potential positive uses. Begging on TikTok has helped those affected by war, poverty and illness to get help from strangers. In one case, a family in the Philippines successfully crowdsourced money for an operation to separate their conjoined twins. In 2023, a man went viral in Indonesia for performing a chicken-pecking dance on TikTok Live. He told local media that he had now recruited others in his village to take part, who are reported to receive a portion of any money generated, which has helped them buy food when rice supplies are low.

But in many cases, it is not clear who stands to profit. Catherine Turner, advocacy director at Anti-Slavery International, said that while some children and adults begged to survive, others were threatened with “physical or psychological coercion” by third parties, who controlled access to their earnings.

TikTok requires government-issued ID from live streamers wanting to earn money from rewards, but for viewers, it is often impossible to know whether those appearing in the content benefit from it themselves. Accounts hosting live streams are often anonymous. While rights groups warn against real-name policies on social media, they say stricter age-assurance processes and moderation systems are needed.

Maya Lahav, researcher in the Centre for Criminology at Oxford, who previously advised tech companies on detecting exploitative cyber-begging, said moderating live streams was resource-heavy and a “huge challenge”. There were ethical issues around whether to remove content or not. “It’s not like other types of influencers aren’t using their children for profit all the time,” she said.

But she said there were serious concerns about human rights abuses, particularly where third parties were involved or those featured were too young, or ill, to consent. “The crux of it is, when does it become exploitation? That’s the balance they need to tread.”

Exploitation fears as people in extreme poverty perform stunts and beg for virtual gifts

‘Profiting from misery’: how TikTok makes money from child begging livestreams

Authors say they are angry that Meta has used their material to train its artificial intelligence (Authors call for UK government to hold Meta accountable for copyright infringement, 31 March). But hasn’t that been going on for thousands of years? Isn’t all human thought an iteration of what has gone before? Artists and scientists have been mining the work of others for generations; that’s how human thought evolves.

Ian McEwan was influenced by LP Hartley’s The Go-Between. George Orwell’s Nighteen Eighty-Four was inspired by Yevgeny Zamyatin’s We. Did Richard Osman invent the genre of cosy crime? The publishing industry as a whole is guilty of putting out bandwagon books, which ape the style, themes and tropes of a hit.

The chief executive of the Society of Authors, Anna Ganley, says writers are “up in arms”. Did she coin that phrase? Creativity has always “trained” on the work of others.

Andrew Vincent

Westmancote, Worcestershire

Letters: Creativity has always ‘trained’ on the work of others, says Andrew Vincent

Mining of authors’ work is nothing new – AI is just doing what creative humans do

Congratulations are in order for Mommy, a Galápagos tortoise and a longtime resident of the Philadelphia Zoo, who recently became a first-time mother at the estimated age of 100.

Mommy, who has lived at the zoo since 1932, laid 16 eggs in November. Four of them have since hatched — the first successful hatching for her species at the zoo, which opened in 1874.

She had help, of course — from Abrazzo, a male tortoise who is also estimated to be about a century old.

Mommy and Abrazzo, both members of the Western Santa Cruz subspecies, are the oldest animals at the Philadelphia Zoo. But Galápagos tortoises can live as long as 200 years, the zoo said, putting them squarely in middle age.

Mommy, a Western Santa Cruz tortoise, recently welcomed four hatchlings at the Philadelphia Zoo, where she has lived since 1932.

An Endangered Galápagos Tortoise Is a First-Time Mother at 100

The smash hit Netflix show Adolescence, which explores a teenage murder fuelled by social media and toxic masculinity, has renewed calls for social media bans in some countries. One of the show’s stars this week said the UK should follow Australia’s lead in banning children aged under 16 from social media platforms.

The ban has been praised in the US and UK, and is described as “world-leading” by the Australian government. Time magazine this week praised the prime minister, Anthony Albanese, for a “remarkable” policy that was “politically uncontroversial” on the basis that both major parties supported it.

Left unsaid was all the criticism raised by mental health groups, LGBTQ+ groups and other campaigners during the rushed process to pass the bill in parliament last year. The committee reviewing the bill only reviewed the legislation for a single day, despite over 15,000 submissions being received.

Author Jonathan Haidt, who reportedly lobbied politicians in Australia to push the policy before it was adopted and privately dismissed critics of his approach, told the New York Times this week that “it’s going to work. It doesn’t have to be perfect at first, but within a few years it will be very good”. If it worked in Australia, it was going to go global “very quickly”, he said.

But nine months out from the policy coming into effect, Australians are still in the dark about how our ban – which was passed by the parliament in November 2024 – will work. And that’s likely to remain the case up to the federal election on 3 May.

A trial of age assurance technology is under way, with schoolchildren still being recruited to participate just weeks before the first report is due.

The under-16s social media ban is due to come into effect in December, but the government faces a number of hurdles before then, including figuring out what tech to use, and whether the platforms – emboldened by the apparent backing of Donald Trump – will comply.

The Age Check Certification Scheme (ACCS), a UK-based company recruited by the Albanese government to conduct an assessment of the technology used to determine whether people are the age they say they are when accessing social media, is due to provide a preliminary report to the government by the end of April.

While the report is said to be on track to be delivered this month, Guardian Australia has confirmed this preliminary report will not be released publicly by the company. A spokesperson for the communications department said the report was never intended for public release and is designed “to afford procedural fairness to trial participants” on any changes that need to be made.

The final report is due just two months after the preliminary report in June, before the communications minister – whoever it will be after the federal election – will get to decide which platforms it applies to, and what technology is appropriate.

ACCS has begun recruitment of school-age children to test out the various technologies, but there is an education and consent process still under way.

From there, schoolchildren will test out age estimation (where tech estimates how old a user is), assurance (where a parent or guardian confirms an age) and verification (using some sort of identity document verification) technologies.

The children will act as “mystery shoppers” and attempt to access a purpose-built online platform through the various age assurance methods, documents released by ACCS state.

This process leaves just weeks to get the trials conducted, analysed, and a final report prepared for government.

The ban is not a major focus of the federal election campaign – it had bipartisan support, after the Coalition pushed for it for months until the Labor government relented.

However, there are still major concerns over how the ban will work, and who is included. TikTok and Meta, for example, are angry over the carve out YouTube received. The government’s messaging on why this exemption was allowed has been mixed.

The communications minister, Michelle Rowland, said last year that YouTube would be included in a range of services exempt from the ban on health and education grounds. But in the draft wording of the document that sets out what services are exempt, YouTube is granted an exemption on its own, while health and education services are another carve out.

In response to questions in Senate estimates from the Greens senator Sarah Hanson-Young last month, the department said the exclusion was “consistent with broad community sentiment, which highlights the value of YouTube as a tool for education and learning”.

Evidence shows most children under 13 accessing social media are accessing YouTube.

A report from the eSafety commissioner last month stated more than 80% of children aged between eight and 12 are accessing social media, despite the current minimum age requirement being 13.

However, this figure was largely skewed by children accessing YouTube, either by watching while logged out, or using a parent or carer’s account. When YouTube is excluded, the figure is closer to 44%.

TikTok and Snapchat are second and third behind YouTube (68%) on 31% and 19%, respectively.

It is also worth noting that the stakeholder advisory board overseeing the trial has some members who have long called for bans or restrictions on online pornography, and have called for online censorship. But missing from the board are digital rights and privacy groups.

Those organisations have subsequently been invited to apply to join the stakeholder advisory board, after inquiries from Guardian Australia, but as of yet do not appear to be included.

Whether all this results in a report that the government can rely upon and implement before the end of this year remains to be seen, as does whether the social media companies will be willing players.

The inconsistent treatment of some platforms over others might lead to companies such as Meta – which has already approached the Trump administration over their treatment by the Australian government – to seek the US government to push back on the ban before it comes into effect.

But this week, Albanese, and the opposition leader, Peter Dutton, said the ban was not up for negotiation.

A trial is under way but the government faces many hurdles – including whether Trump-emboldened tech companies will comply

Australia’s social media ban is attracting global praise – but we’re no closer to knowing how it would work

Meta and Pinterest have reportedly made significant donations to the Molly Rose Foundation, a charity set up to campaign for internet safety.

The foundation was set up in the name of 14-year-old Molly Russell, who took her own life in 2017 after viewing harmful material linked to suicide and self-harm on social media platforms.

The foundation’s latest annual report refers to grants received from donors who wish to remain anonymous, with trustees agreeing to respect their request.

According to the BBC, these payments are believed to have been made by Meta and Pinterest since 2024 and are expected to continue over a number of years. Details of the amounts reportedly donated have not been disclosed. However, it is understood the Russell family have not received any money from the donation.

“Following the coroner’s inquest into Molly’s death, we have decided that we will pursue the aims we share with Meta and Pinterest through the Molly Rose Foundation to help ensure young people have a positive experience online, instead of pursuing legal action,” the Russell family said in a statement. “We, Molly’s family, have always made clear that we would never accept compensation consequent upon Molly’s death.”

The reported donations come as social media companies face increased criticism over the impact of online platforms on children’s mental health.

In January, as part of sweeping changes to its policies, Meta’s boss, Mark Zuckerberg, said the company would scrap factcheckers in order to boost free speech and reduce “censorship”, and would rely on users to report content.

The Molly Rose Foundation previously warned that Meta’s changes could place young people at greater risk of encountering harmful content online. It has launched campaigns urging the government to strengthen the Online Safety Act and place more responsibility on technology companies for content promoted through algorithms.

According to the BBC, the charity has recruited a chief executive officer, two public policy managers, a head of communications and a fundraising manager within the past nine months.

Ian Russell, Molly’s father, remains an unpaid trustee of the foundation and continues to be a prominent internet safety campaigner.

Meta and Pinterest have been approached for comment by the Guardian.

Anonymous donations made to Molly Rose Foundation, set up after Molly Russell, 14, took her own life after viewing harmful material on social media

Meta and Pinterest believed to have donated to Molly Russell charity

Four private astronauts ended a mission in a SpaceX vehicle on Friday by splashing down in waters near Oceanside, Calif.

It was the first time the company had brought people back to Earth in the Pacific Ocean, after six years of its Dragon capsule splashing down off Florida in the Atlantic Ocean or the Gulf of Mexico.

The Fram2 mission had circled the Earth for four days in a north-south orbit. The journey was the first time people have been able to look down directly at the North and South Poles from orbit.

SpaceX moved its operations to the Pacific to eliminate the problem of Dragon debris falling in random parts of Earth. The Pacific is the biggest pool of water on the planet, and the weather along the West Coast of the United States tends to be pretty nice, too, which provides more days favorable for the return of astronauts.

After years of NASA and private crews returning to Earth near Florida, the company shifted its landing zone to the West Coast for the private Fram2 mission.

SpaceX Astronauts Splash Down Off California Coast for the First Time

When territorial animals are confronted by intruders, they instinctively protect their turf — no matter how small.

For warty birch caterpillars, that means patrolling one of the tiniest territories on Earth: the tips of birch leaves. Scientists observed the caterpillars warding off intruders with loud vibrations that advertise they are in command of a domain that stretches a few millimeters across.

“It’s like rap battles,” said Jayne Yack, a professor of neuroethology at Carleton University in Ottawa and an author of the study, which was published on Tuesday in The Journal of Experimental Biology.

Dr. Yack’s team is the first to observe an insect defending a leaf tip, a discovery that hints at a hidden world of territorial disputes playing out on small scales. These caterpillars are kings of the tiniest castles ever identified.

As they cling to leaf tips, newborn warty birch caterpillars produce vibrations that can ward off invaders approaching their millimeter-size domain.

The Very Territorial Caterpillar

Once again, the future of TikTok in the US is at stake. After a years-long tussle over whether or not to ban the app in the country, the deadline for the company to divest or sell its assets to a non-Chinese owner is up again on 5 April. Donald Trump has said his administration is “very close” to a deal for the app.

A handful of potential buyers have said they’re interested in the tremendously popular social media app and various news reports have floated other types of deals, including an investment from the Trump-friendly venture capital firm Andreessen Horowitz or a bid from Amazon. The president signed an executive order in January to postpone a ban-or-divest deadline until April; earlier this week he said he would “like to see TikTok remain alive”. But the path forward for TikTok, and its 170 million US users, remains murky.

After announcing sweeping tariffs on dozens of countries including China, Trump hinted on Thursday aboard Air Force One that he might lessen the trade penalties if ByteDance, the Chinese company that owns TikTok, were to approve a sale.

“We have a situation with TikTok where China will probably say we’ll approve a deal, but will you do something on the tariffs. The tariffs give us great power to negotiate,” he said.

ByteDance has said it has no plans to sell the app and in court filings said that divestiture “is simply not possible: not commercially, not technologically, not legally”. ByteDance and TikTok did not return requests for comment.

The idea of banning TikTok originated with Trump in 2020, who said the Chinese-owned app posed a danger to national security. It quickly became a bipartisan issue and Congress overwhelmingly voted to ban the app last year. In January, the US supreme court sided with Congress and unanimously upheld a federal law requiring TikTok divest or be banned. The deadline was initially set for 19 January.

The night before the deadline, TikTok shuttered the app with a message that read: “Sorry, TikTok isn’t available right now.” Apple and Google also removed it from their app stores, because under the federal law they would be penalized for distributing it. In its message, the social media company said: “We are fortunate that President Trump has indicated that he will work with us on a solution to reinstate TikTok once he takes office. Please stay tuned!”

On 20 January, Trump’s first day in office, he issued the executive order that extended the ban-or-divest deadline by 75 days. Now that cutoff date is looming.

While initially proposing to ban TikTok, Trump made an about-face last year while campaigning for president, having joined the app and amassed millions of followers. In September, he posted to his Truth Social account “FOR ALL THOSE THAT WANT TO SAVE TIK TOK IN AMERICA, VOTE TRUMP!” Since then, he’s been working to make good on that pledge.

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

On Tuesday, CBS reported that Trump has been considering final proposals for TikTok. Those include plans from a long list of investors in the private equity, venture capital and tech industries. Among those investors are asset manager Blackstone, business software company Oracle, e-commerce giant Amazon, a crypto foundation and the founder of OnlyFans. Oracle is reportedly leading a coalition bid with several investors, including Andreessen Horowitz, according to the FT.

Oracle, which was co-founded by Trump ally Larry Ellison, has been looking to buy a lucrative stake in TikTok for years. The software company already houses all of TikTok’s US user data on its cloud infrastructure platform, a deal that came about in 2022 to address security concerns.

“It is highly unlikely that TikTok will go dark again. All signs point to a deal or another extension,” said Kelsey Chickering, a principal analyst for Forrester. “If TikTok divests in the US, the real question is whether its algorithm comes with the sale. TikTok without its algorithm is like Harry Potter without his wand – it’s simply not as powerful.”

Deadline for company to divest from Chinese ownership nears as Trump mulls over potential bids from US firms

TikTok ban deadline looms in US amid last-minute takeover bids

Contrary to its name, the common coot is no ordinary bird. Contrary to its name, the common coot is no ordinary bird. These plump, black-and-white birds have made themselves at home in the canals of central Amsterdam. There, they have become expert recyclers, building their nests out of human trash. These plump, black-and-white birds have made themselves at home in the canals of central Amsterdam. There, they have become expert recyclers, building their nests out of human trash. Recently, Dutch researchers deconstructed some of their nests layer by layer — and discovered human refuse dating back more than three decades. Recently, Dutch researchers deconstructed some of their nests layer by layer — and discovered human refuse dating back more than three decades. The nests were time capsules, animal-created archives of our human-dominated age. The nests were time capsules, animal-created archives of our human-dominated age.

One Bird Nest, 30 Years of Human Trash A coot’s nest reveals that what humans throw away doesn’t really go away.

Auke-Florian Hiemstra, a doctoral student at the Naturalis Biodiversity Center in the Netherlands, has spent years studying how birds use human materials in their nests.

Much of his research involves collecting abandoned coot nests and painstakingly logging their contents: drug paraphernalia, earbuds, windshield wipers. “Everything that ends up in the canals in Amsterdam most likely finds its way to a coot nest,” he said.

Melissa Schriek for The New York Times

But in late 2021, he found a piece of trash, buried at the bottom of a large coot nest, that stopped him short: a wrapper from a Mars bar promoting the 1994 World Cup. “That really gave me goose bumps,” he said. “Suddenly we had this big realization. Like, these deeper layers are actually older layers.”

Fittingly, the nest had been built just outside an archaeological museum, atop a metal pipe jutting above the canal’s surface. Mr. Hiemstra wondered if, like an archaeologist, he could peer back in time by dating the artifacts he found in each layer of the nest.

All told, the nest contained 635 artificial items, including foil from cigarette packages and a ticket to Amsterdam’s National Maritime Museum. Roughly one-third of the items were related to food. Mr. Hiemstra carefully examined each object for an expiration date that might indicate roughly when it had been added to the nest.

Auke-Florian Hiemstra

These were estimates. Highly processed foods could have expiration dates that extended for months or years after when they were eaten. And any given piece of trash might have lingered for a while before a coot added it to its nest.

For a recent paper, published in the journal Ecology, Mr. Hiemstra and his colleagues documented every roughly datable item they found in the nest, opening a small window onto the history of the city’s avian and human residents.

Auke-Florian Hiemstra

Near the top of the nest was a bounty of candy bar wrappers (including one from a Bounty chocolate bar) bearing 2021 expiration dates, as well as packaging from a protein bar set to expire in 2022.

There was also a thick layer of surgical face masks, which was likely to have been added to the nest sometime after the Covid-19 pandemic began in 2020.

“Face masks, of course, are like a little mattress,” Mr. Hiemstra said. “They’re soft and they may be very warm.” But they can also be dangerous, he added, if the birds become entangled in the masks’ elastic loops.

Auke-Florian Hiemstra

Much of the food packaging came from items with long shelf lives. But the scientists also found a carton of milk that must have been purchased — and, one hopes, consumed — close to its expiration date of May 2013.

Mr. Hiemstra also found an abundance of trash emblazoned with the logo of a single brand: McDonald’s. The sauce containers were easy to date, as they bore expiration dates on their lids.

Auke-Florian Hiemstra

A faded plastic foam McChicken box was trickier, bearing only a 1996 copyright date.

“I was in a very deep rabbit hole at a certain point, just talking with these people who were vintage McDonald’s collectors,” Mr. Hiemstra said. “It really felt like a kind of McDonald’s archaeology.”

Auke-Florian Hiemstra

Ultimately, he was unable to pinpoint a clear date for the container, although the item remained an evocative artifact. “We found the packaging of one bird, a McChicken, as part of the nest of another bird,” he said.

And then, in a deep, mud-covered layer of the nest, which emitted “a distinct canal smell,” was the Mars wrapper that started it all. It carried a FIFA logo with a 1993 copyright and promoted the 1994 World Cup, which took place in the United States and was won by Brazil.

Mr. Hiemstra had no memory of that World Cup; he was born in 1992. “So this wrapper was just as old as I am, almost,” he said.

Auke-Florian Hiemstra

The soccer player Ronaldo, in blue, hoisting the World Cup trophy with the Brazilian national team in 1994. The researcher’s native Netherlands lost in the quarterfinals to the eventual champions, though he was too young to remember it. Onze/Icon Sport, via Getty Images

The nest is a testament to the ingenuity and resourcefulness of wild animals in a rapidly changing world. “It really tells the story of the Anthropocene, but then from the bird’s perspective,” Mr. Hiemstra said.

It’s also a physical embodiment of how profoundly humans are reshaping the environment, he added, and how long-lasting the effects can be.

Roughly three decades ago, some weary commuter or ravenous teenager ordered a McChicken sandwich and then tossed out the box that it came in. “Just one meal from one person — the packaging is still here,” Mr. Hiemstra said. “A bird has been breeding on it for 30 years.”

A coot’s nest reveals that what humans throw away doesn’t really go away.

One Bird Nest, 30 Years of Human Trash

After listening to hundreds of hours of ape calls, a team of scientists say they have detected a hallmark of human language: the ability to put together strings of sounds to create new meanings.

The provocative finding, published Thursday in the journal Science, drew praise from some scholars and skepticism from others.

Federica Amici, a primatologist at the University of Leipzig in Germany, said that the study helped place the roots of language even further back in time, to millions of years before the emergence of our species. “Differences between humans and other primates, including in communication, are far less distinct and well-defined than we have long assumed,” Dr. Amici said.

But other researchers said that the study, which had been conducted on bonobos, close relatives of chimpanzees, had little to reveal about how we use words. “The present findings don’t tell us anything about the evolution of language,” said Johan Bolhuis, a neurobiologist at Utrecht University in the Netherlands.

Hundreds of hours of recordings suggest that the apes can generate meaning by stringing sounds together in pairs. But some scholars are skeptical.

In the Calls of Bonobos, Scientists Hear Hints of Language

Ground sloths emerged in South America tens of million years ago, eventually ranging as far north as Canada. While their modern relatives dwell in trees and top out at the size of a dog, ground sloths also occupied land and seashores and, at their largest, rivaled today’s elephants.

We know from preserved pelts that some ground sloths had lots of fur. But did other sloths evolve to be nearly hairless?

Research published earlier this year in The Journal of Mammalian Evolution offers new insights into what the extinct sloths may have looked like, and how furry they really were.

Developing a more accurate picture of ground sloths is not simple. Their habitats varied from tropical to cold climates across the Americas. And they came in a variety of sizes. The megatheres, known as Eremotherium and Megatherium, were among the biggest terrestrial mammals of their time.

New research painted a more accurate picture of the megafauna that spread widely around the Americas before they went extinct.

Giant Sloths’ Hairy Truth Revealed by Scientists

Forty years ago, it would take a four-drawer filing cabinet to store 10,000 documents. You would need 736 floppy disks to hold those same files; now it takes up no physical space at all to store 10,000 documents on the cloud.

As data storage has evolved, so too has the whole information landscape, and with it the challenges of storing, transferring and appropriately using people’s personal data.

An exhibition by the Information Commissioner’s Office (ICO), which opened at Manchester Central Library this week, charts the evolution of data privacy through 40 items, each chosen to illustrate how access to information has evolved, or how data has been at the heart of some of the biggest news events of the past four decades.

“I think the wonderful thing about the exhibition is that the world that we occupy, like any specialty, is filled with jargon and technicalities,” the information commissioner, John Edwards, said.

“[People] won’t know what a data controller is, they don’t know what a data processor is, they don’t know what a data subject is, we have to use some of these specialist terms. What the exhibition shows is what we do is about people, and it’s about real human impacts.”

Items in the exhibition, which is also available to view online, include a Pokémon toy, a floppy disk, a Tesco Clubcard, a modem, a millennium bug pamphlet, a football shirt and a Covid vaccination card.

Other exhibits highlight how the ICO has made changes in society; from ending the “employment deny list” in the construction industry, to the introduction of public food hygiene ratings for restaurants.

Edwards’ favourite item? A pair of spiked lawn aerator shoes, which illustrate an early example of enforcement action when, in the 1980s, the company behind them was found to be making almost as much money from selling its customers’ information as it was from selling the shoes.

“They were fined quite significantly for their exploitative marketing techniques,” Edwards said.

View image in fullscreen Information commissioner John Edwards’ favourite item in the exhibition is a pair of spiked lawn aerator shoes. Photograph: Christopher Thomond/The Guardian

The 40th plinth in the exhibition remains empty, with members of the public asked to put forward their own ideas for objects that have shaped the data landscape.

“That’s to reflect the notion that privacy is personal, subjective,” Edwards said. “We each have our own expectations and experiences.”

The ICO was founded 40 years ago, in a small office near Manchester, as the UK’s data protection regulator, responsible for presiding over a new Data Protection Act. Since then the landscape overseen by the regulator has changed beyond all recognition.

View image in fullscreen An NHS Covid vaccination card. Photograph: Andy Rain/EPA

Today, people have “tens of thousands of times” more personal data out in the world than they did when the role was created, Edwards said.

“You just move through the world today shedding data wherever you go,” Edwards said. He asks how the Guardian travelled to this interview (by bus). “Did you tap on?” he asked. “You created data there.

“You used your cell phone on the way, so you’re pinging off cell towers the whole way. Forty years ago, none of this was a thing.

“We’ve had this, kind of, data world sneak up on us, and most of that data, for most of those years, was an accidental byproduct.

“We’ve now got to a stage where all these companies are going, hang on a minute, we could make money out of that.”

Every year, he said, “hundreds of billions of data transactions” took place, and with “infinite variety”. The ICO was regulating everything from small schools, GP surgeries and libraries to huge “fiercely competitive” social media companies, which Edwards said often “don’t pause long enough to test the privacy implications of what they’re doing”.

“The biggest challenge has been trying to keep up with the pace of change,” Edwards said. “Companies innovate very quickly, we regulate and investigate very slowly.

“I think part of the objective of an investigation is to put some lines in the sand for companies to say: ‘You can’t do this,’” he added. “But if it takes us three years to do that, then all the companies have moved on by the time we learn that lesson. So that’s the biggest challenge, and that’s something we’ve got to get better at.”

What will data and privacy look like in 40 years’ time? “Look, I don’t know where we’re going to be four weeks from now,” Edwards said.

“The geopolitical situation is really kind of volatile at the moment,” he said. “We’ve got an environment where US tech firms are pushing back on a lot of regulation that affects them, even when it’s in a jurisdiction that they want to do business in. We haven’t seen how that’s going to play out.”

“Quantum computing has potential to change everything,” he added. “Agentic AI is the next AI coming down the pipeline.

“It’s going to be really fascinating even the next 12 months,” he said, “let alone the next 40 years.”

Forty items on display in Manchester, collated by information commissioner, chart evolution of personal data usage over 40 years

Floppy disks and vaccine cards: exhibition tells tale of privacy rights in UK

After our children left home, my wife and I decided to treat ourselves and buy a new car for a driving holiday in Europe. We’d been driving a family estate car for years, loading it up with kids and making trips to and from universities, but we wanted something for ourselves.

As a surprise, she booked a test drive for the Tesla Model S for my birthday. It was unlike any car I’d been in before. I thought “Wow, this is amazing.” It felt like the future: a computer on wheels that was constantly updating with new features. I can’t say I feel that way now – and many people seem to share that view. Tesla sales figures declined by 13% in the first few months of this year. Others feel even more uneasy: more than 200 demonstrations happened last weekend outside company facilities around the world to protest against Elon Musk and the wrecking ball he has taken to the federal government.

I had no complaints about the car itself: it was quiet. It was large. And, importantly for us, as an electric vehicle it was environmentally friendly. Some people go for the sports car when they gain a new sense of freedom in later life, but I had done a bit of research and wanted to move away from diesel and petrol.

Despite people voicing concerns about a lack of charging infrastructure, I found Tesla’s network to be wonderful. We drove all over Europe – driving to Switzerland one year, Italy another, and last summer to the south of France. At one point we even packed in five adults and all their suitcases after an airport pickup.

I would consider myself left-leaning, but buying an electric car wasn’t a huge political statement for us at the time, but rather an acknowledgment of the way things are going to be in the future. Little did I know how our choice would be politicised for us six years down the line.

In the early years of owning the car I remember a neighbour saying: “What is it like to own a car built by a madman?” But I brushed it off. I had watched videos of Musk at Tesla events, and he seemed like a poor man’s Steve Jobs. Much like the Apple CEO, I saw Musk as a frontman rather than the designer and didn’t take much notice of the business owner.

Soon it was impossible to ignore him. My wife and I were horrified as he publicly adopted rightwing positions – with his endorsement of Donald Trump for US president and then for the far-right AfD in Germany, as well as his support for Tommy Robinson. It seemed so odd to us that a man who would champion electric vehicles would also hold such reactionary views.

I started to notice a shift in the Tesla owners’ group on Facebook. Having not owned an electric car before, the group was a useful resource. People were very helpful and even recommended places to stop off during our European holiday. When Musk’s politics began to grow more extreme, members began to post that they no longer felt comfortable owning a Tesla. And then others would respond: ““We’re not interested in your politics.” I saw someone giving stick to the Guardian and even calling the BBC “BBC Stalin”, which was funny for me as I always personally felt it went too far the other way.

View image in fullscreen Protesters demonstrate against Tesla CEO, Elon Musk, during a nationwide Tesla Takedown rally outside a dealership in Pasadena, California. Photograph: Mario Tama/Getty Images

I started to reflect on owning a Tesla and it was my daughter who made the case for selling it. But judging by people posting secondhand cars in the Facebook group, prices have been dropping. I feel trapped. I cycled to meet a friend recently, and he took one look at my reasonably nice bike and said “It’s probably worth more than your Tesla now.” Someone else said to me he was interested in buying a Tesla, but he was going to wait until they were virtually worthless.

It’s a shame as in every other way the car is a perfect match for our lifestyle. It’s cheap to run, easy to service and Tesla has a mobile repair service that dispatches a “ranger” to your house to fix any issues.

If I was buying a car from scratch I would boycott the company, but we are where we are. For now I am going to have to keep hold of it. Next time, maybe I’ll get a European electric vehicle. Who knew owning a car would prove so difficult?

I bought it to be part of a greener future, but that was before Musk proved so awful. I’d sell it now, but prices have dropped, says the secret Tesla driver

Trapped with a Tesla: my dream car has become a living nightmare | The secret Tesla driver

A leading academic tasked by the UK government with reviewing the effects of smartphones on teenagers has suggested blanket bans are “unrealistic and potentially detrimental”.

Amy Orben, from the University of Cambridge, will lead the work on children and smartphone use that has been commissioned by the Department for Science, Innovation and Technology (DSIT) along with a team of other academics from a number of UK universities.

Ministers have so far been resistant to implementing any new legal restrictions on social media and smartphones for children that goes further than the current Online Safety Act, which clamps down on harmful content.

Some MPs have been pushing for further restrictions that go beyond harmful content – including on access to social media for those under 16, full bans on smartphones in schools or restrictions on social media algorithms that are able to train addictive content on young teenagers.

In a paper Orben published this week with four co-authors in the British Medical Journal (BMJ), they said bans and restrictions were unlikely to be effective – though they did advocate for children and teens to have phone-free spaces.

“Bans and restrictions have been successfully used for public health issues such as smoking. But smoking is not comparable with smartphone and social media use because the harms from smoking are extensive, clear-cut and by far outweigh the benefits,” the paper said.

“Prescribing abstinence from all technologies to protect against harms is unrealistic and potentially detrimental in a society where technology use is a practical necessity and confers various benefits, including information access and social support.”

The paper said that there had been “increased public pressure to mitigate the potential harmful effects of smartphones and social media on health, wellbeing … academic performance, disruptive behaviours and bullying.”

But it said the evidence suggested there were “no simple, one-size-fits-all answers” and that while most parents and policymakers were “primed to believe arguments that smartphones and social media are inherently harmful, the evidence about their overall effect on children is not clear-cut”.

Bans on devices could “undermine children’s rights to technology design and education that will help them thrive as adults”, the academics warned.

The paper acknowledged that for some more vulnerable children “access to certain digital content can result in grave harm” but said legislators should also acknowledge restricting access “can be harmful to other high risk populations”.

The researchers did raise concern about “trending content” aimed at younger users to encourage recurrent use, saying those designs were “purposefully not supporting the development of healthy tech habits”.

Peter Kyle, the science and technology secretary, appointed Orben in January to lead further research into the effects of smartphones on children’s health and wellbeing. Orben, a leading academic in the field, said she did not herself advocate for specific policy positions. Her work has been cited by both opponents and proponents for bans, including by the Australian government, which has banned social media for under-16s.

She told the Guardian: “Understanding the digital world’s impact on young people is complex and there are no simple solutions. The report my team is currently leading was commissioned by government to produce recommendations on the types of research which could help generate high-quality evidence, not to provide policy recommendations.”

“As an independent researcher my role is to examine and communicate evidence clearly, not advocate specific policies. In the BMJ article, we highlighted the benefit of multiple approaches noting technology-free spaces are important but smartphone or social media bans are not a complete solution, as other areas such as digital literacy and platform safety are also important parts of the online safety picture.”

Ministers have come under pressure in recent months to go further on smartphone use for teenagers, especially in the wake of the Netflix drama Adolescence which depicts a murder committed by a 13-year-old in thrall to toxic “manosphere” content online.

The Labour MP Josh MacAlister was forced to drop plans in a private member’s bill on restricting social media algorithms for teenagers when ministers opposed the plans.

The education secretary, Bridget Phillipson, is reviewing her department’s guidelines on smartphone bans in schools, by looking at a small cohort of schools to see if the current rules are working effectively.

The prime minister, Keir Starmer, convened a roundtable with campaigners and the makers of Adolescence in No 10 on Monday, where he suggested that it should be discussed more widely in schools. But he said there “isn’t one single policy lever to pull” when it came to addressing potential harms.

The DSIT has been contacted for comment.

Dr Amy Orben says there are no ‘one-size-fits-all answers’ given importance of access to online information

Blanket ban on teen smartphone use ‘potentially detrimental’, says academic

Meta faces a $2.4bn (£1.8bn) lawsuit accusing the Facebook owner of inflaming violence in Ethiopia after the Kenyan high court said a legal case against the US tech group could go ahead.

The case brought by two Ethiopian nationals calls on Facebook to alter its algorithm to stop promoting hateful material and incitement to violence, as well as hiring more content moderators in Africa. It is also seeking a $2.4bn “restitution fund” for victims of hate and violence incited on Facebook.

One of the claimants is the son of Prof Meareg Amare Abrha, who was murdered at his home in Ethiopia after his address and threatening posts were published on Facebook in 2021 during a civil war in the country. Another claimant is Fisseha Tekle, a former researcher at Amnesty International who published reports on violence committed during the conflict in Tigray in northern Ethiopia and received death threats on Facebook.

Meta has argued that courts in Kenya, where Facebook’s Ethiopia moderators were based at the time, did not have jurisdiction over the case. The Kenyan high court in Nairobi ruled on Thursday that the case fell within the jurisdiction of the country’s courts.

Abrham Meareg, the son of Meareg, said: “I am grateful for the court’s decision today. It is disgraceful that Meta would argue that they should not be subject to the rule of law in Kenya. African lives matter.”

Tekle said he cannot return home to Ethiopia because of Meta’s failure to make Facebook safe. “Meta cannot undo the damage it has done, but it can radically change how it moderates dangerous content across all its platforms to make sure no one else has to go through what I have,” he said. “I look forward to this matter now being heard by the court in full.”

The case, supported by non-profit organisations including Foxglove and Amnesty International, also demands a formal apology from Meta for the murder of Meareg. The Katiba Institute, a Kenya-based NGO focusing on the Kenyan constitution, is the third claimant in the case.

In 2022 an analysis by the Bureau of Investigative Journalism and the Observer found that Facebook was letting users post content inciting violence through hate and misinformation, despite being aware that it was fuelling tensions in Tigray.

Meta rejected the claims at the time, saying it had “invested in safety and security measures” to tackle hate and inflammatory language along with “aggressive steps to stop the spread of misinformation” in Ethiopia.

In January the company said it was removing factcheckers and “dramatically” reducing the amount of censorship on the platform, although it would continue to tackle illegal and high severity violations.

Meta said it did not comment on ongoing legal matters.

Son of murdered academic calls on Facebook owner to ‘radically change how it moderates dangerous content’

Meta faces £1.8bn lawsuit over claims it inflamed violence in Ethiopia

Child safety campaigners have warned the government against watering down landmark online laws as part of a UK-US trade deal, describing the prospect of a compromise as an “appalling sellout” that would be rejected by voters.

A draft transatlantic trade agreement contains commitments to review enforcement of the Online Safety Act, according to a report on Thursday, amid White House concerns the legislation poses a threat to free speech.

The Molly Rose Foundation, a charity established by the family of Molly Russell, a British teenager who took her own life after viewing harmful online content, said it was “dismayed and appalled” at the prospect of the act being a bargaining chip in a deal.

The MRF said it had written to the business secretary, Jonathan Reynolds, outlining its concerns and urging him “not to continue with an appalling sellout of children’s safety”.

The commitment to review enforcement of the OSA and another tech-focused piece of legislation – the Digital Markets, Competition and Consumers Act – was reported by the online newsletter Playbook, which said the legislation would undergo a review of how it is implemented and not a “do-over”.

This week, the Guardian reported that the US state department had challenged Britain’s communications regulator, Ofcom, over the OSA’s impact on freedom of expression.

The Online Safety Act focuses heavily on child safety and requires that tech platforms should shield under-18s from harmful content such as suicide and self-harm-related material. Companies that breach the act face fines of up to £18m or 10% of worldwide revenue, which in the case of those such as Facebook’s owner, Meta, or Google would equate to billions of pounds. In extreme cases, services can also be taken down in the UK.

Beeban Kidron, a cross-bench peer and internet safety campaigner, said: “The Labour party has lost its way. It’s pretty clear from my inbox this morning that their voters could not conceive that they would trade child safety for crumbs from Trump’s table.” The NSPCC, a UK child safety charity, said the government “must not roll back” on its commitment to improving the online environment for children.

Asked in the House of Commons on Thursday whether the digital safety and competition laws and the digital services tax were part of the trade talks, the business secretary said there were “real differences of opinion” on some issues such as VAT but refused to “go into the content of all of the negotiations”. A source close to Reynolds did not deny the Playbook report.

The technology secretary, Peter Kyle, told LBC last month that the government was standing behind its online safety measures and “none of our protections for children and vulnerable people are up for negotiation”.

The prime minister’s spokesperson said the government’s position on online safety had not changed.

“We’ve said previously our Online Safety Act is all about protecting children online, making sure what is illegal offline is illegal online,” they said. “The technology secretary has reiterated that our basic protections for children and vulnerable people are not up for negotiation.”

Child protection charities say watering down Online Safety Act would be an ‘appalling sellout’ by government

Don’t weaken online safety laws for UK-US trade deal, campaigners urge

Tony Blair’s thinktank has urged the UK to relax copyright laws in order to let artificial intelligence firms build new products, as it warned a tougher approach could strain the transatlantic relationship.

The Tony Blair Institute said enforcing firm copyright measures would strain ties with the US, which is poised to announce tariffs on UK goods on Wednesday.

Warning that geopolitical considerations require “urgent and adequate attention” while AI policy is being drafted, TBI said: “Without similar provisions in the United States, it would be hard for the UK government to enforce strict copyright laws without straining the transatlantic relationship it has so far sought to nurture.”

The thinktank has said that if the UK went down the route of demanding licensing of all UK content used in AI models, it would simply push that development work to other territories where there are less strict copyright laws. To enforce a strict licensing model, the UK would also need to restrict access to models that have been trained on such content, which could include US-owned AI systems.

In a report published on Wednesday, TBI said it backed government proposals to let AI firms train their models on copyright-protected material without permission, unless creatives have signalled they want to opt out of the process. With the Trump administration signalling it will not pursue strict AI regulations and China pursuing AI growth at “breakneck speed”, the UK could weaken its economic and national security interests by lagging in the AI race, said TBI.

“If the UK imposes laws that are too strict, it risks falling behind in the AI-driven economy and weakening its capacity to protect national security interests,” said TBI.

The report said arguing that commercial AI models cannot be trained on content from the open web was close to saying knowledge workers – a broad category of professionals ranging from lawyers to researchers – cannot profit from insights they get when reading the same content.

Rather than fighting to uphold outdated regulations, said TBI, rights holders and policymakers should help build a future where creativity is valued alongside AI innovation.

A more restrictive approach to copyright than the one taken by the EU, Singapore or Japan risked pushing AI developers out of the UK, said TBI.

Fernando Garibay, a record producer who has worked with artists including Lady Gaga and U2, said history has been dotted with “end-of-time claims” related to technological breakthroughs, from the printing press to music streaming. Those fears were unfounded, said Garibay in an introduction to the report, and the creative community should not “cling” to a legal status quo.

“The progressive solution is not about clinging to copyright laws designed for an earlier era but allowing them to co-evolve with technological change, to remain effective in fulfilling their regulatory and protective aims in the age of AI,” said Garibay.

The report, called Rebooting copyright: How the UK can be a global leader in the Arts and AI, also calls for the creation of a Centre for AI and Creative Industries that will encourage collaboration between the tech and creative sectors.

Beeban Kidron, a cross-bench peer and prominent campaigner against the government proposals, said the report was effectively telling artists to “put up or shut up” because AI’s advance is inevitable.

“It cannot be that the sum total of the government’s imagination is that the UK become an AI hub for Silicon Valley,” she said. “That has got to be the most depressing vision of the UK imaginable.”

Kidron added that TBI’s acceptance of donations from the US tech billionaire Larry Ellison, which reached $270m (£210m) last year, were “the very definition of conflict of interest”. According to documents filed in the US last year, Ellison’s foundation donated more than $52m (£40m) to TBI in 2024, with a further $218m (£169m) committed. TBI said the $270m commitment was for helping governments around the world, particularly in Africa, with “reform and change programmes”.

“We maintain intellectual independence over our policy work,” said TBI.

Tony Blair Institute says enforcing stricter licensing rules for copyright-protected material will threaten national security interests

UK needs to relax AI laws or risk transatlantic ties, thinktank warns

Donald Trump is preparing to consider a final proposal to decide the future of TikTok before a deadline for the app to find a non-Chinese buyer or face a US ban.

The US vice-president, JD Vance, the commerce secretary, Howard Lutnick, the national security adviser, Mike Waltz, and the director of national intelligence, Tulsi Gabbard, will meet in the Oval Office on Wednesday to discuss the issue, Reuters reported.

In the closely watched sale of TikTok, the White House is playing the role of an investment bank, with Vance running the auction.

The private equity firm Blackstone is discussing joining ByteDance’s existing non-Chinese shareholders, led by Susquehanna International Group and General Atlantic, in contributing fresh capital to bid for TikTok’s US business, Reuters reported last week.

On Sunday, Trump said a deal with ByteDance to sell the short video app used by 170 million Americans would be struck before Saturday’s deadline.

Trump is preparing to announce global tariffs on what he has called ‘liberation day’ on Wednesday. Last week, he said he would be willing to reduce tariffs on China to get a TikTok deal done.

Trump set the deadline in January for TikTok to find a non-Chinese buyer or face a US ban on national security grounds due to have taken effect that month under a 2024 law.

The US venture capital firm Andreessen Horowitz is in discussions to invest in TikTok as part of an effort led by Trump to gain control of the app, the Financial Times reported on Tuesday.

The Silicon Valley Trump supporter and Andreessen Horowitz’s co-founder, Marc Andreessen, is in talks to add new outside investment that will buy out TikTok’s Chinese investors, as part of a bid led by Oracle and other American investors to carve it out of its parent company, ByteDance, according to the FT.

Blackstone is reportedly evaluating making a small minority investment in TikTok’s US operations.

skip past newsletter promotion Sign up to Business Today Free daily newsletter Get set for the working day – we'll point you to all the business news and analysis you need every morning Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

The talks on the future of TikTok are coalescing around a plan for the biggest non-Chinese investors in ByteDance to raise their stakes and acquire the short-video app’s US operations, Reuters reported last month.

Trump said last month his administration was in touch with four different groups about a prospective TikTok deal, without identifying them.

TikTok and Andreessen Horowitz did not immediately respond to Reuters’ requests for comment.

Owner ByteDance required to find non-Chinese buyer for video app’s American operations by Saturday

Trump to consider final proposal on TikTok future as US ban deadline looms

Tesla reported a 13% drop in vehicle sales in the first three months of the year, making it the electric vehicle maker’s worst quarter since 2022. It’s another sign that Elon Musk’s once high-flying electric car company is struggling to attract buyers.

The drop is likely due to a combination of factors, including its ageing lineup, competition from rivals and a backlash from Musk’s embrace of rightwing politics. It also is a warning that the company’s first-quarter earnings report later this month could disappoint investors.

Tesla reported deliveries of 336,681 vehicles globally in the January-March quarter. Analysts polled by FactSet expected much higher deliveries of 408,000. The figure was down from sales of 387,000 in the same period a year ago. The decline came despite deep discounts, zero financing and other incentives.

Tesla’s stock has plunged by roughly half since hitting a mid-December record as expectations of a lighter regulatory touch and big profits with Donald Trump as president were replaced by fear that the boycott of Musk’s cars and other problems could hit the company hard. Teslas around the country were vandalized in protest of Musk after he dismantled entire federal agencies in his role as the head of the so-called “department of government efficiency”. Musk also made a gesture at a rally that his estranged daughter called “definitely a nazi salute”.

Despite Donald Trump’s attempts to shore up company sales with a Tesla presentation in front of the White House and Musk’s assurances to employees that the company has a “bright and exciting” future, the drop in sales was striking.

Matt Britzman, a senior equity analyst at Hargreaves Lansdown, said: “The scale is worse than many had expected.

“There’s no way to sugarcoat it, Tesla’s first-quarter delivery numbers are a disappointment, though many investors were already preparing for a soft number,” Britzman said.

Analysts are still not sure exactly how much the fall in sales is due to the protests or other factors. Electric car sales have been sluggish in general, and Tesla in particular is suffering as car buyers hold off from buying its bestselling Model Y because of plans for an updated version later this year. Even before Musk took up his role in government alongside Trump, Tesla had struggled with meeting its delivery targets. Earnings calls and delivery reports failed to match analysts’ expectations throughout 2024.

“Headlines will point to branding issues, and it’d be naive to assume that’s not a factor here, but it misses the key point,” Britzman said. “Deliveries have been significantly impacted by downtime at factories as Tesla launched the long-awaited refreshed version of the Model Y, its bestselling car.”

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

The Austin, Texas, electric vehicle maker has also lost market share to rivals in recent months as their offerings improve, including those of BYD. The Chinese EV giant unveiled in March a technology that allows its cars to charge in just a few minutes.

Despite what analysts describe as strong demand for the Model Y, they expect to see continue volatility for the company for a while. Tesla’s most recent release, the futuristic Cybertruck, has failed to find widespread adoption.

The “brand is under pressure”, Britzman said. “It’s rare to see sentiment toward a company so closely tied to a polarising White House, and until Musk pulls his focus back to Tesla, shares will remain volatile.”

Drop is likely combination of ageing lineup, increased competition and backlash to Musk’s politics

Tesla quarterly sales slump 13% amid backlash against Elon Musk

Italy’s interior ministry has written to police forces across the country to increase security at Tesla dealerships after 17 of the electric cars made by Elon Musk’s company were destroyed in a fire in Rome.

Italy’s state police anti-terrorism unit, Digos, is investigating whether the fire at the Tesla dealership in Torre Angela, a suburb in the east of the capital, was started by anarchists.

Firefighters worked for hours to put out the blaze in the early hours of Monday. Drone images showed a row of the burnt-out remains of the vehicles in a parking area of the dealership. Using his social media platform, X, Musk referred to it as “terrorism”.

There are 13 Tesla dealerships in Italy, all managed by the parent company, the majority of them in Rome, but also in other cities including Florence and Milan.

An interior ministry source said the circular was aimed at “raising awareness” of possible anti-Tesla protesters amid a global wave of vandalism in response to Musk’s political activities in the US. If needed, surveillance of dealerships would be increased, it said.

skip past newsletter promotion Sign up to Headlines Europe Free newsletter A digest of the morning's main headlines from the Europe edition emailed direct to you every week day Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

Since the inauguration of Donald Trump as US president in January, Musk has been slashing the federal workforce as chief of the administration’s so-called “department of government efficiency”, prompting the emergence of “Tesla Takedown”, a boycott movement that started in the US before spreading to Europe.

Although most protests have so far been peaceful, Tesla dealerships and cars have increasingly been targets of vandalism. Seven vehicles were set alight at a dealership in Ottersberg, Germany, on Saturday, and two Tesla stores in Sweden, one in the capital, Stockholm, and the other in the coastal city of Malmö, were vandalised with orange paint on Monday.

Musk has nurtured close relations with European far-right party leaders including the Italian prime minister, Giorgia Meloni, who in an interview in early January described him as “a brilliant man”.

Matteo Salvini, who leads the far-right League, a member of Meloni’s ruling coalition, expressed solidarity with Musk after the Rome incident.

“Too much unjustified hatred against the Tesla car company,” Salvini wrote on X. “The season of hate and conflict must come to an end as soon as possible. My solidarity goes out to Elon Musk and to all the workers who have been threatened and attacked.”

State police anti-terrorism unit investigating whether blaze in Torre Angela was started by anarchists

Italian police increase security at Tesla dealerships after 17 cars destroyed in Rome fire

Parents can now block their children from communicating with specific friends or playing certain games on Roblox, an online gaming platform popular with children.

The changes form part of a suite of safety updates intended to give parents more control over their child’s experience on the platform.

From Wednesday, parents and caregivers who identify themselves with an ID or credit card will have access to three new tools. The friend management tool means they can block anyone on their child’s friends list, preventing their child from exchanging direct messages with that account, and report people they believe are violating Roblox policies.

They can also review and change the content maturity level for their child’s account, determining which games their child can access, and obtain detailed screen-time insights.

Under the Online Safety Act, which came into force this year, tech companies must tackle harmful content on their platforms or face fines of up to £18m or 10% of global revenue.

There have been reports of bullying and grooming on Roblox and fears that children are being exposed to explicit or harmful content on the site, which is the most popular platform in the UK among gamers aged eight to 12.

Roblox’s chief safety officer, Matt Kaufman, said safety was at the company’s core and its mission was to be “the safest and most civil online platform in the world”.

The US-based company is one of the world’s largest games platforms, with more monthly users than Nintendo Switch and Sony PlayStation combined. In 2024, the site averaged more than 80 million players a day, and roughly 40% of those were under 13.

Roblox introduced 40 safety updates last year, including preventing users under 13 from sending direct messages. Roblox has also updated its voice safety technology, which uses a machine-learning model to moderate chat between players more accurately than human moderators.

Andy Burrows, the chief executive of the Molly Rose Foundation, said he welcomed the safety improvements but that “Roblox still needs to get to grips with substantial problems with harmful and age-inappropriate content”.

He added: “Extensive research has shown Roblox is awash with age-inappropriate games and communities, including depression rooms that can compound misery and offer no support to vulnerable children.

skip past newsletter promotion Sign up to Pushing Buttons Free weekly newsletter Keza MacDonald's weekly look at the world of gaming Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

“This content raises fundamental questions about Roblox’s broader commitment to safety and shows it cannot just rely on parental controls but must take decisive action to make the platform safe for its young users.”

Last month, Roblox’s co-founder and chief executive, David Baszucki, said the platform was vigilant in protecting its users and that “tens of millions” of people had “amazing experiences” on the site.

He added: “My first message would be: if you’re not comfortable, don’t let your kids be on Roblox. That sounds a little counterintuitive, but I would always trust parents to make their own decisions.”

Parents can block children from talking to certain people and get detailed screen-time insights on site popular with under-13s

Roblox gives parents more power over children’s activity on gaming platform

The UK government is trying to placate peer and Labour backbencher concerns about copyright proposals by pledging to assess the economic impact of its plans.

Creative professionals including Sir Paul McCartney, Sir Tom Stoppard and Kate Bush have strongly criticised ministers’ proposals to let artificial intelligence companies train their models on copyright-protected work without permission, unless the rights holder opts out.

Their stance has been supported by peers, who have passed amendments pushing back on the proposals, and by some backbench MPs.

It is understood that concessions offered to MPs and peers this week include an economic impact assessment, with a report that could address issues such as how AI developers access data to train their models and transparency around use of copyright-protected works.

Ministers are hoping the concessions will allow thethe data (use and access) bill to pass. The bill has been used as a vehicle for fighting back against the proposals but now could be stuck in a process known as “ping pong”, where a bill is batted between the House of Lords and the House of Commons while amendments are added and removed. The government’s overtures were first reported by the Times.

View image in fullscreen Beeban Kidron. Photograph: Antonio Olmos/The Observer

Beeban Kidron, the cross-bench peer who successfully introduced amendments to the bill that battle the proposals, only for them to be removed by the government, said the Department for Science, Innovation and Technology was “almost exclusively” focused on the interests of the US tech lobby. Her amendments on transparency – allowing copyright owners to know when, where and how their work is used in AI models – will at least give copyright holders some protection, she said.

“In the absence of a UK government willing to protect its workers and wealth creators, at least that way UK-based copyright holders would have some of the tools to protect themselves.”

James Frith, a Labour member of the culture, media and sport committee, said: “I remain encouraged by the government’s commitment to a dialogue on this.”

But, he added, there needed to be complete transparency on how copyright protected work was used and use of such material needed to be compensated.

skip past newsletter promotion Sign up to Business Today Free daily newsletter Get set for the working day – we'll point you to all the business news and analysis you need every morning Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

“I have yet to see an explanation as to why AI companies are held back by copyright or by having to pay for content,” said Frith.

A government spokesperson said: “No final decisions have been taken. We are carefully considering the consultation responses and continue to engage with tech companies, the creative industries, and parliament to inform our approach.

“We have always been clear that no changes will be made until we are absolutely confident we have a practical plan that delivers on each of our objectives.”

The government is expected to publish its response to a consultation on the proposals over the summer or by October.

Economic impact assessment is one concession aiming to head off opposition from MPs, peers and creatives such as Paul McCartney and Tom Stoppard

UK government tries to placate opponents of AI copyright bill

When lightning strikes a tree in the tropics, the whole forest explodes.

“At their most extreme, it kind of looks like a bomb went off,” said Evan Gora, a forest ecologist at the Cary Institute of Ecosystem Studies in Millbrook, N.Y. Dozens of trees around the one that was struck are electrocuted. Within months, a sizable circle of forest can wither away.

Somehow, a single survivor stands, seemingly healthier than ever. A new study by Dr. Gora, published last week in the journal New Phytologist, reveals that some of the biggest trees in a rainforest don’t just survive lightning strikes. They thrive.

The rainforest in Panama’s Barro Colorado Nature Monument is the perfect place to study whether some trees are immune to lightning. It’s home to the Smithsonian Tropical Research Institute and one of the most closely studied tropical forests in the world. Dr. Gora set out to study whether individual trees in the forest benefit from being struck by lightning. And if they did, does that help the population of the species survive at a larger scale?

Early on, he spent much of his time climbing trees, looking for signs of lightning damage. But making critical observations could be painfully inefficient. Dr. Gora would begin climbing one tree, convinced it was the struck trunk, only to get 50 feet up and see he actually wanted to be up the neighboring tree. Honey bees would also swarm Dr. Gora’s eyes and ears.

Before a discovery in a Panamanian rainforest, “it seemed impossible that lightning could be a good thing for the trees,” a scientist said.

This Tree Wants to Be Struck by Lightning

More than 600 miles separated the epicenter of Friday’s earthquake in Myanmar from the skyscrapers of Bangkok, which lurched and swayed on the skyline of the vast metropolis. A 33-story high-rise under construction even collapsed. How could the shaking in Bangkok, the Thai capital, be connected to an earthquake so far away?

The answer involves low frequency seismic waves, which are capable of traveling vast distances and making high-rise buildings sway.

When a large earthquake strikes, it simultaneously radiates different frequencies of shaking. Some produce a rapid back-and-forth rattling, others low-frequency swaying.

That was true on Friday, when the earthquake in Myanmar produced violent, high-frequency seismic waves; they destroyed low-rise buildings, Buddhist pagodas and other structures near to the quake’s epicenter just outside Mandalay, Myanmar’s second largest city. Many of the destroyed places were made with brick and masonry, brittle materials that are susceptible to that kind of shaking.

Large quakes produce shaking at a variety of frequencies. Some waves can travel hundreds of miles, and are amplified by local geological conditions.

How the Myanmar Earthquake Shook Skyscrapers in Bangkok

A professional YouTuber in Queensland has been ordered to pay $350,000 plus interest and costs to the former world record score holder for Donkey Kong, after the Brisbane district court found the YouTuber had defamed him “recklessly” with false claims of a link between a lawsuit and another YouTuber’s suicide.

William “Billy” Mitchell, an American gamer who had held world records in Donkey Kong and Pac-Man going back to 1982, as recognised by the Guinness World Records and the video game database Twin Galaxies, brought the case against Karl Jobst, seeking $400,000 in general damages and $50,000 in aggravated damages.

Jobst, who makes videos about “speed running” (finishing games as fast as possible), as well as gaming records and cheating in games, made a number of allegations against Mitchell in a 2021 YouTube video. He accused Mitchell of cheating, and “pursuing unmeritorious litigation” against others who had also accused him of cheating, the court judgment stated.

The court heard Mitchell was accused in 2017 of cheating in his Donkey Kong world records by using emulation software instead of original arcade hardware. Twin Galaxies investigated the allegation, and subsequently removed Mitchell’s scores and banned him from participating in its competitions.

The Guinness World Records disqualified Mitchell as a holder of all his records – in both Donkey Kong and Pac-Man – after the Twin Galaxies decision.

The judgment stated that Jobst’s 2021 video also linked the December 2020 suicide of another YouTuber, Apollo Legend, to “stress arising from [his] settlement” with Mitchell, and wrongly asserted that Apollo Legend had to pay Mitchell “a large sum of money”.

The judgment noted multiple versions of the video were published, with the allegations removed or addressed, before Jobst issued a retraction video in July 2021, two months after the original. The first video was viewed by more than 500,000 people, including 200,000 in Australia, the court heard.

Judge Ken Barlow found Jobst had made five defamatory imputations about Mitchell, including that Mitchell had required Apollo Legend pay him a large sum of money and thus implying he “in essence, hounded Apollo Legend to death”.

Barlow found this was “based on a fallacy” – and that Jobst, in not making further inquiries before publishing the video, had been “recklessly indifferent” to if it was true.

View image in fullscreen Jobst (middle) leaves the Brisbane district court on Tuesday. Photograph: Jono Searle/AAP

Jobst denied the imputations had been made in the video. He argued Mitchell had a pre-existing bad reputation because he had been previously exposed as a cheat, the court heard.

Barlow found Mitchell did have an existing reputation as a cheat and for suing people who alleged he was a cheat, and found that Mitchell had expressed joy when he believed – incorrectly – on an earlier occasion that Apollo Legend may have died. But Barlow found Jobst had severely damaged Mitchell’s reputation and caused distress.

He described Jobst as having a “self-aggrandising and perhaps self-protective tendency not to admit error and not to back down once he has taken a stance”.

Barlow framed Jobst’s actions as a “crusade” against Mitchell, stating that he was trying to “[show] his audience that he is the knight who slew the Mitchell dragon”.

The judgment summary noted that the court was not called on to decide, and did not decide, if Mitchell had cheated in his world record scores.

The court awarded Mitchell $300,000 in damages for non-economic loss, and an additional $50,000 in aggravated damages due to Jobst’s publishing the video twice, mocking Mitchell’s complaint about it, failing to apologise and withdraw the allegations, and his “clear malice” towards Mitchell.

Jobst was ordered to pay more than $40,000 in interest on the damages from the date of publication, as well as Mitchell’s costs.

Mitchell had previously sued Twin Galaxies in the US in a case which settled in January 2024. As part of the settlement, Twin Galaxies published a statement that Mitchell had “produced expert opinion that the game play on the tapes of Mr Mitchell’s record game plays could depict play on an original unmodified Donkey Kong arcade hardware if the hardware involved was malfunctioning, likely due to the degradation of components”.

The company said it would reinstate all of Mitchell’s previous scores on its website’s official historical database, and remove the dispute thread about Mitchell’s records. Mitchell’s Guinness World Records were also reinstated in June 2020.

Queenslander ordered to pay $350,000 plus interest and costs as judge finds his video about American gamer Billy Mitchell was ‘recklessly’ based on false claim

Donkey Kong champion wins defamation case against Australian YouTuber Karl Jobst

OpenAI said it had raised $40bn in a funding round that valued the ChatGPT maker at $300bn – the biggest capital-raising session ever for a startup.

It comes in a partnership with the Japanese investment group SoftBank and “enables us to push the frontiers of AI research even further,” OpenAI announced, adding it would “pave the way toward AGI (artificial general intelligence)” for which “massive computing power is essential”.

SoftBank said it wanted to realise “artificial super intelligence” (ASI) surpassing human intelligence and OpenAI was the partner closest to achieving that goal.

SoftBank is to put $10bn at first into OpenAI and $30bn more by the end of 2025 if certain conditions are met.

Also on Monday, OpenAI announced it was building a more open generative AI model as it faces growing competition in the open-source space from DeepSeek and Meta.

OpenAI previously was a fierce defender of closed, proprietary models that do not allow developers to modify the basic technology to adapt AI to their goals. “We’ve been thinking about this for a long time, but other priorities took precedence. Now it feels important to do,” said the OpenAI chief executive, Sam Altman.

OpenAI and defenders of closed models – which include Google – have often decried open models as riskier and more vulnerable to malicious and non-US government use.

Elon Musk, a former OpenAI investor, has called on OpenAI to “return to the open-source, safety-focused force for good it once was”.

Large companies and governments have proven reluctant to use AI models they have no control over, especially when data security is a concern. Meta and DeepSeek let companies download and modify their models.

Meta’s CEO, Mark Zuckerberg, said this month that Llama had hit 1 bn downloads. DeepSeek’s lower-cost R1 model in January rocked the world of artificial intelligence.

OpenAI has been riding on the success of its latest image-generation features in ChatGPT. Altman claimed the tool helped add “one million users” in one hour and overwhelmed OpenAI’s graphics processing units.

With Agence France-Presse

Japanese investment group says it wants to realise ‘artificial super intelligence’ – smarter than people – in biggest capital raising ever for a start-up

OpenAI raises up to $40bn in record-breaking deal with SoftBank

The chief executive of Channel 4 said that artificial intelligence companies are “scraping the value” out of the UK’s £125bn creative industries, and urged the government to take action.

Alex Mahon told MPs that if the government pursues its proposed plan to give AI companies access to creative works unless the copyright holder opts out, it would put the UK creative industries in a “dangerous position”.

Speaking on the work of Channel 4 at a culture, media and sport select committee meeting on Tuesday, she said: “AI is clearly absolutely critical to the future of our industry, and many industries. The debate of the day is we need very clear terms. UK copyright law is very, very clear. And what is happening at the moment is the scraping of value from our creative industries.”

Critics of the government’s opt-out proposal, issued in a consultation that closed in February, argue that it is unfair and impractical.

Generative AI models, the term for technology that underpins powerful tools such as ChatGPT, are trained on vast amounts of data to generate highly realistic responses.

Mahon said that allowing large language models (LLMs) to continue to freely scrape data poses a big threat to the creative industry, which generates £125bn in gross value added (GVA), a measure of how much value companies add through the goods and services they produce.

“The creative industries account for 6% of the UK’s GVA and is growing 1.5 times faster than other sectors,” she said. “If we continue in a world where LLMs can scrape and use that data without paying for it properly we are in a dangerous position for the industry.”

Authors, artists and executives from the film and TV industries are among those to have argued that allowing companies to train AI programs on their works without a licence posed a threat to their livelihoods.

Mahon said Channel 4 was “very clear” that the copyright regime should be “opt-in”.

She said: “The burden should be on them, not us. We are very clear we think that LLMs need to licence what they use and pay properly for it. We can’t have automated scraping; we need a proper payment and licensing regime.”

Mahon also said that Channel 4 – which is state-owned but fully commercially funded, primarily through advertising revenue – said that the broadcaster broke even last year.

The broadcaster, which will release its 2024 results in the summer, reported a deficit of £52m in 2023.

skip past newsletter promotion Sign up to Business Today Free daily newsletter Get set for the working day – we'll point you to all the business news and analysis you need every morning Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

This followed three years of financial surpluses, totalling £178m. Mahon said at last year’s results press conference that she expected deficits in 2024 and this year.

“I’m pleased to say that 2024 was a much better year,” she said. “We [will be] pretty much on a flat deficit [when we report]. Pretty much breakeven.”

Mahon also said that regulations governing the programming prominence that public service broadcasters [PSBs] such as Channel 4, the BBC and ITV enjoy on TV platforms may need to be extended.

“Without prominence we will disappear,” she said. “We need to think about reach and prominence on new platforms. We do need to think about regulation over time for promotion [of PSBs] rather than just preservation [of them]. We should think about how to extend prominence [regulations] to social platforms and how we would do that in the UK.”

Government plan over copyright-protected work would put industries in ‘dangerous position’, Alex Mahon tells MPs

AI firms are ‘scraping the value’ from UK’s £125bn creative industries, says Channel 4 boss

After a stay aboard the International Space Station that unexpectedly stretched to nine and a half months, the NASA astronaut Suni Williams is back to doing something she enjoys on Earth.

“I actually went out and ran three miles yesterday,” Ms. Williams, who returned to Earth two weeks ago, said on Monday during a news conference at NASA’s Johnson Space Center in Houston. “So I will give myself a little pat on the back.”

Ms. Williams and her fellow astronaut Butch Wilmore repeatedly expressed gratitude. During Ms. Williams’s remarks, she thanked the two astronauts who had shared the SpaceX Crew Dragon capsule that gave them a ride back to Earth. She thanked NASA. She thanked SpaceX. She thanked Boeing. She thanked the medical team that helped them get accustomed to gravity again.

Those words of gratitude may have been appreciated by people at the space agency who — like many federal workers — are uncertain about their mission, their direction or even their continued employment since President Trump’s inauguration.

Suni Williams and Butch Wilmore spoke in their first news conference since returning to Earth two weeks ago from an unexpectedly long I.S.S. stay that lasted more than nine months.

Astronauts Emphasize Gratitude as NASA Contends With Uncertainty

For a creature made up of only a single cell, the stentor is a giant. This trumpet-shaped organism is among the largest unicellular organisms, stretching as long as a sharpened pencil tip. But sometimes it has a hard time vacuuming up the swimming bacteria and microscopic algae it eats to survive.

New research reveals that stentors, which are part of a group called protists, may address this challenge by eating “family style.” In a paper published on Monday in the journal Nature Physics, scientists shared the discovery that colonies of stentors can make water flow more quickly around them, helping them suck up more prey.

The new findings suggest that, although they lack neurons or brains, stentors can cooperate with one another.

“These single-cell organisms can do things that we assume are limited to more complex organisms,” said Shashank Shekhar, a biophysicist at Emory University in Atlanta who is the lead author of the new paper. “They form this higher order structure, like what we do as humans.”

Under microscopes, scientists found that giant single-cell organisms were able to vacuum up more food when they are stuck together.

Eating ‘Family Style’ May Have Set the Stage for Life as We Know It

Some 1,900 leading researchers accused the Trump administration in an open letter on Monday of conducting a “wholesale assault on U.S. science” that could set back research by decades and that threatens the health and safety of Americans.

The letter’s signatories, all of them elected members of the National Academies of Sciences, Engineering and Medicine, warned of the damage being done by layoffs at health and science agencies and cuts and delays to funding that has historically supported research inside the government and across American universities.

“For over 80 years, wise investments by the U.S. government have built up the nation’s research enterprise, making it the envy of the world,” the letter said. “Astoundingly, the Trump administration is destabilizing this enterprise by gutting funding for research, firing thousands of scientists, removing public access to scientific data and pressuring researchers to alter or abandon their work on ideological grounds.”

Nearly 2,000 scientists urged that Congress restore funding to federal agencies decimated by recent cuts.

Trump Administration Has Begun a War on Science, Researchers Say

Ten years ago, this was just a big hole in the ground behind a Lowe’s home improvement store in southern New Jersey, an unlikely place to find what might be one of the world’s most important fossil sites.

But 66 million years ago, tantalizingly close in time to when the dinosaurs went extinct, a multitude of sea creatures died here — a “mass death assemblage” — and sank to the bottom of what was then a shallow sea.

Because of its prehistoric past as a possible mass extinction gravesite, the hole that was once a quarry has become the Edelman Fossil Park & Museum.

Built in Mantua, N.J., about 20 miles from Philadelphia, the museum welcomed its first paying customers this past weekend. For Kenneth Lacovara, a professor of paleontology and geology at nearby Rowan University and the museum’s executive director, it is the culmination of a decade of work.

The museum hopes that after learning about the planet’s prehistoric past, people will do more to preserve Earth’s future.

A New Dinosaur Museum Rises From a Hole in the Ground in New Jersey

Elon Musk gave out $1m checks on Sunday to two Wisconsin voters, declaring them spokespeople for his political group, ahead of a Wisconsin supreme court election that the tech billionaire cast as critical to Donald Trump’s agenda and “the future of civilization”.

“It’s a super big deal,” he told a roughly 2,000-person crowd in Green Bay on Sunday night, taking the stage in a yellow cheesehead hat. “I’m not phoning it in. I’m here in person.”

Musk and groups he supports have spent more than $20m to help conservative favourite Brad Schimel in Tuesday’s race, which will determine the ideological makeup of a court likely to decide key issues in a perennial battleground state. Musk has increasingly become the center of the contest, with liberal favourite Susan Crawford and her allies protesting Musk and what they say is the influence he wants to have on the court.

“I think this will be important for the future of civilization,” he said. “It’s that’s significant.”

He noted that the state high court may well take up redistricting of congressional districts, which could ultimately affect which party controls the US House.

“And if the [Wisconsin] supreme court is able to redraw the districts, they will gerrymander the district and deprive Wisconsin of two seats on the Republican side,” Musk claimed. “Then they will try to stop all the government reforms we are getting done for you, the American people.”

A unanimous state supreme court on Sunday refused to hear a last-minute attempt by the state’s Democratic attorney general to stop Musk from handing over the checks to two voters, a ruling that came just minutes before the planned start of the rally.

Two lower courts had already rejected the legal challenge by Democrat Josh Kaul, who argues that Musk’s offer violates a state law. “Wisconsin law prohibits offering anything of value to induce anyone to vote,” Kaul argued in his filing. “Yet, Elon Musk did just that.”

But the state supreme court, which is now controlled four-to-three by liberal justices, declined to take the case as an original action. The court gave no rationale for its decision. All four liberal justices have endorsed Dane County Judge Susan Crawford, the Democratic-backed candidate.

Kaul had no immediate comment on the court’s order.

Musk’s attorneys argued in filings with the court that Musk was exercising his free speech rights with the giveaways and any attempt to restrict that would violate both the Wisconsin and US constitutions.

The payments are “intended to generate a grassroots movement in opposition to activist judges, not to expressly advocate for or against any candidate,” Musk’s attorneys argued in court filings.

Musk’s political action committee used a nearly identical tactic before the presidential election last year, offering to pay $1m a day to voters in Wisconsin and six other battleground states who signed a petition supporting the First and Second amendments. A judge in Pennsylvania said prosecutors failed to show the effort was an illegal lottery and allowed it to continue through Election Day.

Musk’s attorneys, about four hours before the rally was to begin, asked that two liberal justices who have campaigned for Crawford – Jill Karofsky and Rebecca Dallet – recuse themselves from the case. His attorneys argued their work for Crawford creates “the spectre of inappropriate bias.” If they did recuse, that would leave the court with a three-two conservative majority.

Both justices rejected the request and said they would spell out their reasons why at a later date.

One of the court’s conservative justices has endorsed Schimel, who wore a “Make America Great Again” hat while campaigning Sunday.

Schimel said in a national television interview that he does not control “any of the spending from any outside group, whether it’s Elon Musk or anyone else” and that all Trump asked was whether he would “reject activist judges” and follow the law.

“That’s exactly what I’ve committed to anybody, whether it’s President Trump, Elon Musk or any donors and donors or supporters or voters in Wisconsin. That’s my commitment,” Schimel told Fox News Sunday.

The contest has shattered national spending records for a judicial election, with more than $81m in spending.

It comes as Wisconsin’s highest court is expected to rule on abortion rights, congressional redistricting, union power and voting rules that could affect the 2026 midterms and the 2028 presidential election in the state.

Musk denied he was buying votes but said the court election outcome would be critical to Trump’s agenda and ‘the future of civilization’

Elon Musk hands out $1m checks to voters amid Wisconsin supreme court election race

AI tools will soon be in use in classrooms across England, but the education secretary, Bridget Phillipson, has one big question she wants answered: will they save time?

Attending a Department for Education-sponsored hackathon in central London last week, Phillipson listened as developers explained how their tools could compile pupil reports, improve writing samples and even assess the quality of soldering done by trainee electrical engineers.

After listening to one developer extol their AI writing analysis tool as “superhuman”, able to aggregate all the writing a pupil had ever done, Phillipson asked bluntly: “Do you know how much time it will have saved?”

That will be our next step, the developer admitted, less confidently.

In an interview with the Guardian, Phillipson said her interest in AI was less futuristic and more practical. Could classroom AI tools free teachers from repetitive tasks and bureaucracy, allow them to focus on their students and ultimately help solve the recruitment crisis that bedevils England’s schools?

“I think technology will have an important role to play in freeing up teachers’ time, and in freeing up that time, putting it to better use with more face-to-face, direct teaching that can only ever be done by a human,” she said.

“This is less about how children and young people use technology, and more about how we support staff to use it to deliver a better education for children. I think that’s where the biggest potential exists.

“In the next few years I want to see AI tech embedded across schools, with staff supported to use the best technology to improve children’s outcomes but also to make teaching a more attractive career for people to go into and stay.

“It’s not about replacing teachers. It’s about how the use of technology can complement the very human face-to-face contact that can’t be replaced.”

Some of the tools are being designed to solve very specific problems. Jessica Leigh Jones, the chief executive of iungo Solutions, showed Phillipson and Peter Kyle, the secretary of state for science, innovation and technology, a tool to assess how well students had wired circuit boards for vocational qualifications such as T-levels.

Using a digital microscope, the tool can rapidly scan dozens of boards and provide feedback to students in a fraction of the time needed by a qualified and experienced electrical engineer – a specialist role in short supply, Jones noted.

Jade Lesh, iungo’s digital growth engineer, told Phillipson the tool “minimises the time spent marking each piece of work”, and that their tests have found it to be as accurate as a trained human’s eye.

Time saved was also Phillipson’s interest in an observation tool being developed by TeachScribe, giving early-years teachers earpieces to record their comments about each child’s development – how well they are playing with others or being creative – which are then uploaded to a database and compared with the government’s early-years curriculum goals.

skip past newsletter promotion Sign up to First Edition Free daily newsletter Our morning email breaks down the key stories of the day, telling you what’s happening and why it matters Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

The developer, Ewan Dobres, said it would save teachers a “phenomenal amount of time” that could be spent with pupils because they would no longer have to stop and write down each observation.

Other applications on display included tools that would allow teachers to scan and transcribe pupils’ handwritten work, and use AI to assess it against national benchmarks in English and literacy.

Helen Williams, chief executive of the Inmat multi-academy trust in Northamptonshire, said her schools had been working with a company called Stylus on a writing assessment tool that not only saved time but allowed teachers to set pupils more of the writing tasks that are a burden for them to mark.

Williams said using AI tool for marking had advantages other than saving time. They “levelled the playing field” between experienced and early-career teachers, and removed the traps teachers can fall into when assessing their own pupils.

“If you’ve got a class of 30, you’re always going into marking with an idea of ‘well, this one will be all right’. AI doing it takes away that unconscious bias and is actually just marking what’s there, which might throw up some things that a teacher hasn’t picked up in the past,” she said.

Exclusive: education secretary exploring tools to compile student reports and assess writing and vocational skills

Bridget Phillipson eyes AI’s potential to free up teachers’ time

Over the weekend, protesters gathered at Tesla showrooms in hundreds of cities across the world to demonstrate against Elon Musk laying waste to the US government in alliance with Donald Trump. Their goal: stigmatize Tesla’s cars. One sign in Manhattan read: “Burn a Tesla, save democracy.” Protesters are using the commercial democracy of consumer products to influence US political democracy.

***

My colleagues Dara Kerr and Edward Helmore report:

In New York City, several hundred anti-Tesla protesters gathered outside the EV company’s Manhattan showroom on Saturday. Sophie Shepherd, 23, an organizer with Planet Over Profit, explained that the rally was not about protesting electric cars.

“We’re here to protest Musk, who has essentially held a Tesla car show on the White House lawn,” she said. “We want to disrupt his business as much as possible, so that includes all Teslas, and not just the Cybertruck.”

The demonstrations did not draw huge crowds. Nonetheless, Musk’s fortune has dropped more than $100bn as Tesla’s stock has fallen by nearly half since its peak value in December 2024.

Musk’s political ventures are forcing many Tesla owners to either be a spokesperson for or against the company

Musk is synonymous with Tesla. In its comprehensive annual report, the company discloses in the section on risks that it is “highly dependent on the services of Elon Musk, Technoking of Tesla and our Chief Executive Officer”. Technoking of Tesla is one of Musk’s official, if juvenile, titles.

By extension, consumer sentiment regarding Musk’s public persona is inextricable from Tesla. When he was known primarily as a multi-hyphenate technical genius, the inspiration for Iron Man, that perception boosted Tesla’s brand like an engine from his rocket company.

In light of the polarization of Musk as a public figure, however, reading Tesla’s risk assessments in its 10-K form takes on a different tone: “We will need to maintain public credibility and confidence in our long-term business prospects in order to succeed.”

“We are targeting a global mass demographic with a broad range of potential customers,” reads Tesla’s disclosure.

Musk’s political ventures are forcing many Tesla owners to either be a spokesperson for or against the company. Individual owners are being asked to justify their purchases or publicly declare the separation of their beliefs from their vehicles. Of particular concern are the dates of the purchases, with “I bought this before Elon went crazy” bumper stickers proliferating as a sign of rebuke of the so-called “department of government efficiency”. Yet there is a wide swath of the US public that seems to want their car to just be a car, not a political statement.

For his own part, Musk is targeting a narrower and narrower audience, hosting a Tesla sales event on the White House lawn alongside one of the most poorly rated and divisive presidents in modern political history. I can’t predict Tesla’s fate, but for now, widespread social souring on Tesla is making a dent in its finances.

Read the full story on Saturday’s protests here.

More on Elon Musk

Does Donald Trump know who Totoro is?

View image in fullscreen Photograph: X, The White House

On Thursday, the White House posted a cartoon depicting the arrest of a wailing woman beside an American flag. The image was captioned: “Virginia Basora-Gonzalez, a previously deported alien felon convicted of fentanyl trafficking, was arrested by @ICEgov in Philadelphia after illegally reentering the U.S. She wept when taken into custody (picture attached).”

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

By ghiblifying this woman, the White House inadvertently made a previously deported felon and literal fentanyl trafficker sympathetic

The White House’s image is shocking. The contrast between the harshness of deportation and the round softness of the characters rendered in a warm color scheme, evocative of children’s animated classics such as Spirited Away and My Neighbor Totoro, is ghoulish. That seems to be the point. The image is both viral nonsense – there may be little meaning to extract from it other than its capacity to appall – and also a communication from the most powerful office in the United States. We may be obligated to attempt to understand it.

Earlier last week, OpenAI upgraded ChatGPT with better image generation. Soon, there was a deluge on Twitter/X of real photographs reproduced as still images in the distinctive style of Studio Ghibli, the Japanese animation studio behind, most recently, the Oscar, Bafta and Golden Globe-winning film The Boy and the Heron. I made one of the images. OpenAI CEO Sam Altman changed his profile photo to a character he described as “a twink Ghibli style”, too. The White House appears to have done the same. (OpenAI did not respond to a request for comment on the cartoon depiction of the arrest and deportation.)

Trump has never seen Ghibli’s Princess Mononoke, I would venture. Part of what is supremely bizarre in the image is visual; another part is political. Hayao Miyazaki, the 84-year-old founder of Ghibli, has criticized nationalism and championed environmentalism, expressing opposition to positions easily attributable to Trump’s administration.

In a 2018 documentary, Miyazaki called one young animator’s use of artificial intelligence to produce a grotesque moving figure “an insult to life itself”. Whether OpenAI violated Ghibli’s copyright in equipping its AI model to imitate the studio’s style remains an open legal and philosophical question. I recommend reading Walter Benjamin’s The Work of Art in the Age of Mechanical Reproduction, which concerns recreations of the Mona Lisa.

Mike Solana, the chief marketing officer of Peter Thiel’s Founders Fund and the founder of the conservative newsletter Pirate Wires, tweeted that the image was ineffective in its mockery because it was too kawaii.

“24 hours in and we face the first great example of a person working with a form they didn’t earn, and therefore don’t understand. by ghiblifying this woman the white house inadvertently made a previously deported felon and literal fentanyl trafficker sympathetic… this specific aesthetic is innately endearing. you can’t make me want to see a ghibli character in pain,” Solana tweeted.

I have written before about how US conservatives love AI while liberals shun it, which I think remains the case. The jarring admixture of seemingly silly internet aesthetics with the Trump administration’s punitive approach to immigration will not disappear soon. The White House made a similar post last month. “ASMR: Illegal Alien Deportation Flight 🔊”. I expect more.

The wider TechScape

As demonstrations against Elon Musk and his political ties spread, Tesla owners find themselves caught in a growing debate – turning a once-iconic brand into a symbol of division

How Tesla became a battleground for political protest

Since the release of ChatGPT in late 2022, generative artificial intelligence has trickled down from adults in their offices to university students in campus libraries to teenagers in high school hallways. Now it’s reaching the youngest among us, and parents and teachers are grappling with the most responsible way to introduce their under-13s to a new technology that may fundamentally reshape the future. Though the terms of service for ChatGPT, Google’s Gemini and other AI models specify that the tools are only meant for those over 13, parents and teachers are taking the matter of AI education into their own hands.

Inspired by a story we published on parents who are teaching their children to use AI to set them up for success in school and at work, we asked Guardian readers how and why – or why not – others are doing the same. Though our original story only concerned parents, we have also included teachers in the responses published below, as preparing children for future studies and jobs is one of educators’ responsibilities as well.

Some parents and teachers told us they are going full steam ahead, integrating ChatGPT into everyday interactions with their children to explain new concepts, answer incessant questions and illustrate fanciful tales. Others, wary of the dangers AI can pose to young people, are demonstrating it only a little at a time in closely supervised settings.

Still other parents and teachers are refraining from showing their children and students how to use AI at all, concerned about harms and the impossibility of ethical use of a technology still rife with unresolved copyright questions. We felt their responses made valuable contributions to the discussion, too.

The answers below have been edited for length and clarity.

The parents explaining the world with AI’s help

I swear, watching my kids discover AI has been one of those unexpected parenting joys nobody tells you about. Remember when we all used to say, ‘Let’s Google it’? Well, my nine-year-old has completely ditched that phrase. Now it’s always ‘Daddy, can we ChatGPT it?’ whenever he’s stuck on homework. The best part? He’s learning to ask for hints instead of answers – figuring things out himself but with a little AI nudge.

Then there’s my six-year-old – this kid’s questions could outlast anyone’s patience! You know those evenings when your brain is completely fried but they’re still going strong? That’s when our AI buddy steps in. He will chat away for half an hour, bouncing from dinosaurs to space to ‘why is water wet?’ while I secretly recharge my mental batteries.

My daughter is only three but already thinks she’s royalty. Instead of reading the same princess books for the millionth time, we now create stories where SHE’S the princess having these wild adventures.

–Matt, consultant, 44, Palm Beach Gardens, Florida

I use AI as a kind of intellectual backstop when I’m having a conversation with my children and I can’t answer one of their questions. We have AI voice assistants at home and in the car so this has recently involved asking Alexa for the population of China and examples of how you can use the word ‘credential’ for homework. It fast forwards the conversation and keeps the learning flowing. My children are aware that AI can hallucinate, however, so you need to sense check responses. Alexa recently came up with some incorrect answers about the book Ann of Green Gables, which my eight year old challenged!

–Graham, writer, Lamberhurst Quarter

I use AI tools with my three-year-old to help explore the world. We use apps to identify birdsongs and plants on our daily walks, and ChatGPT to answer questions he has about the world (“what are bones for? Where did my bones come from?”) especially when we get new books from the library on something he hasn’t learned much about yet. I want him to feel comfortable turning to AI to augment his curiosity. At the same time, we minimize all passive play, including screen time.

For now, I try to make time for us to tell stories to each other and do imaginative play for the topics he asks AI about, so he has a chance to do some generative engagement.

AI is a daily part of our routine, but it’s not a particularly high priority.

–Nate, data scientist, East Bay, California.

I am currently introducing OpenAI’s ChatGPT to my oldest daughter, describing it as a creative helper that can be both friend or foe and teaching her basic prompts to make an AI work and benefit you. I show her my daily usage of AI in my working area, creating mails for me, optimising event planning, structuring data. I am teaching her that you have to be always sceptical when using AI and should not start to rely completely on its answers, which in fact can be wrong.

–Ben, 47, entrepreneur, Germany

My son understands Italian (my language) but speaks back in Dutch. My Dutch is very poor. I am occasionally making use of ChatGPT voice interaction tool to translate more complicated things from Italian/English into Dutch – with the upside that the output can be tailored to a six-year-old.

–Name withheld, Netherlands

To introduce them to AI, we logged into ChatGPT and generated images based on their prompts. While their understanding is still very limited, they are beginning to grasp the idea that you can create fake images and videos online.

We then worked together to create a story, which we gave to ChatGPT to generate a screenplay of a news report. Taking this a step further, we used NotebookLM to convert it into a podcast, which I played through our car stereo. The podcast was formatted as a legitimate-sounding news report, but the story itself was absurd – our Shetland pony turning into a unicorn and teaming up with a toy gibbon to fight evil sheep in Co Kerry.

After listening, I asked them whether we should believe the news report. We then searched online for evidence of the story and, unsurprisingly, found nothing. This helped illustrate the importance of verifying information, even when it sounds official. I have no idea how much of this lesson will stick, but it’s an ongoing conversation.

My five-year-old, in particular, is starting to grasp that AI can generate fake content, and my hope is that this awareness will help him recognise or at least question misleading material in the future.

–David, software developer, Ireland

For pupils struggling with writing, I encourage them to ask AI for feedback at a sentence level, requesting information about why sentences don’t parse, together with grammatical explanations.

I encourage them not to use AI to produce whole essays because that’s not how they will learn. Some still ignore this advice, of course!

–Jenny, English and drama teacher, Valencia, Spain

I have a girl, six and a boy, 11. Both access chatGPT via myself. They know if dad does not know something they may insist he asks chatGPT. And I often do, while curating the response. The boy may also access it himself, most often to ask some questions about the world around.

I want kids to understand that LLMs are a tool to be used to achieve our goals. We also talked about kids who do homework with AI. I explained that we humans, also have computers in our heads which have to be trained in order for us to be successful and own tools of our success.

–Anton, fintech director, Geneva

I have been frequently rendered speechless over these past two years by all the changes introduced in my profession by AI. I have an 11-yo and I am staunchly trying to ensure that the first (and possibly only) intelligence she uses must be her own; she is aware of the existence of ChatGPT, AI translation engines and the like, but does not use them for schoolwork or any other writing.

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

–Name withheld, translator, Slovenia

The teachers making AI a part of their curricula

As a guideline, I generally tell students to ask AI the questions that they would ask their teachers. Would they ask their teacher to write them an essay? No. Would they ask their teacher how to improve their essay? Yes.

At the same time, I remind students that AI is inaccurate. It will lie and provide false information. They will need to exercise their critical thinking and fact-checking skills. While we do have guidelines in place to cite AI use, I tell my students they should never cite AI, similar to my guidelines around Wikipedia.

–Adam, secondary English teacher, Vancouver, BC, Canada

Whenever I have used AI in the classroom it has been to spark children’s imaginations, rather than replace them. Adobe Firefly is great for encouraging descriptive writing in English, I simply type in children’s noun-phrases or lists of adjectives and AI creates an image based on the description. We can then adjust our vocabulary and take great delight in how AI interprets it quite literally. This is also a really important lesson as children can see first hand that AI has its flaws.

I have also used character.AI to bring historical characters to life to generate quotations and responses for children, based on their questions.

The children always ask for the website address so they can have a go at home, but I have not shared these directly. I advise the children that sometimes AI can appear so human, that children can become confused and start believing that the voice that answers is real. So AI tools and websites should always be used with an adult who can remind them and keep them safe online.

–Angie, primary teacher, Tunbridge Wells

I have chosen to introduce AI in generative ways – making clear that this is a tool, and it should be used. You need to know how to yield a hammer before you can use it; similarly, I see AI in the same way. Students more or less see it the same. Their teachers often don’t. I predominantly teach Māori and Pasifika students, and although they are often – and rightfully! – annoyed that their culture is sometimes misunderstood by these AI models, the technology is so inherently compelling that students take to it faster than their teachers.

I have done simple things like, for example, during a speech unit, having students use it to generate a speech on a random topic, from a list of random unfamiliar topics, to get them up and talking.

–Adam, 28, high school teacher, Christchurch, New Zealand

The parents using AI to help with homework

My 11-yo son uses it to help with homework and studying. He needs to get ideas and then reformulate the answer in his own words, so it’s not a copy and paste job.

–Joanna, housewife and carer, Bath

I started introducing AI to my first born who is in secondary school while doing his school homework.

My approach has been to show him how AI, particularly ChatGPT, can help to understand better concepts and encourage responsible use. The reason is because as ardent user of technology l believe the earlier he gets to know it will make him better since the AI revolution is here to stay

–Richard, university lecturer, Uganda

The parents who are not introducing AI to their children

My five-year-old son barely even knows the internet exists. Computers are for video calls, games, movies, and reading the day’s temperature off the Met Office website. He definitely doesn’t need to know anything about LLMs at this stage and certainly shouldn’t use them. He needs to learn how to write his own words and draw his own pictures, not use machines for plagiarising and stunting his imagination. If LLMs come up, I would strongly encourage my son not to use them.

They are built on the backs of labour and art from writers and artists who have not licensed their work for this purpose. They are plagiarism machines. They can’t create anything new.

I would tell him to be suspicious of any machine wanting to write words for him, which is tantamount to thinking his thoughts for him. Young children deserve opportunities to develop their critical thinking skills and imaginations.

–Name withheld, higher education administrator, Oxford

I am not introducing it to mine – 10 yrs old and eight. They are extremely sophisticated readers but not great writers. One has ADHD and is always attracted to distraction and short cuts, ie, instinctively avoids hard work.

–Name withheld, writer, London

Guardian readers share the ways and reasons they are preparing their children and students for a future that may necessitate familiarity with generative artificial intelligence

How and why parents and teachers are introducing young children to AI

Elon Musk’s xAI artificial intelligence firm has acquired Musk’s X – the social media platform formerly known as Twitter – for $33bn, marking the latest twist in the billionaire’s rapid consolidation of power.

The all-stock deal announced on Friday combines two of Musk’s multiple portfolio companies, which also include automaker Tesla and SpaceX, and potentially eases Musk’s ability to train his AI model known as Grok.

Musk announced the transaction in a post on X, saying: “The combination values xAI at $80bn and X at $33bn ($45B less $12B debt).”

“xAI and X’s futures are intertwined,” he wrote. “Today, we officially take the step to combine the data, models, compute, distribution and talent.”

Neither X nor xAI spokespersons immediately responded to requests for comment. Much of the deal’s specifics remained unclear, such as how investors may be compensated, how X’s leaders would be integrated in the new firm or the prospect of regulatory scrutiny.

“This development feels surprising and somewhat unexpected,” PP Foresight analyst Paolo Pescatore said. “To a certain extent, it closes a chapter in the turbulent saga of X.”

“The choice of $45bn is not a coincidence,” said DA Davidson & Co analyst Gil Luria. “It is $1bn higher than the take-private transaction for Twitter in 2022” and he can share the value of the xAI business with X co-investors.

Musk, the world’s wealthiest man, has also consolidated his power in Washington DC by overseeing the Trump administration’s cost-cutting efforts as head of the so-called Department of Government Efficiency, or Doge. That has also put him in the position to potentially influence agencies that oversee his business dealings.

An investor in xAI, and now in the combined entity, told Reuters they were not surprised by the deal, viewing it as Musk consolidating his leadership and management at his own companies. The investor declined to be named.

Musk did not ask investors for approval but told them that the two companies had been collaborating closely and the integration would drive deeper integration with Grok.

Musk’s xAI startup was launched less than two years ago and recently raised $10bn in a funding round that valued the company at $75bn, according to a media report.

In February, Musk, 53, made a $97.4bn bid with a consortium for the ChatGPT maker OpenAI, which was rejected, with OpenAI saying that the startup was not for sale. Musk co-founded OpenAI with CEO Sam Altman in 2015.

Musk competes directly with the popular OpenAI platform and has sued in federal court in California to prevent his rival from converting from a non-profit to a for-profit business. A judge earlier this month denied Musk’s request for a preliminary injunction that would prevent the changeover.

The wide release of AI software has set off a flurry of investment and competition in Silicon Valley. In an effort to be more efficient, companies are seeking ways to integrate the software into nearly every part of their operations.

As competition in AI intensifies, xAI has been ramping up its data centre capacity to train more advanced models, and its supercomputer cluster in Memphis, Tennessee, called Colossus, is touted as the largest in the world.

xAI introduced Grok-3, the latest iteration of its chatbot, in February, as it tries to compete with Chinese AI firm DeepSeek and Microsoft-backed OpenAI. The X platform can serve to further distribute xAI products, while also providing a real-time feed of users’ musings, screenshots and other data.

Musk clinched a deal in 2022 to buy X, then Twitter, for $44bn, ending its run as a public company since its 2013 initial public offering, declaring that “the bird is freed” once the acquisition closed.

He gutted the company’s workforce after the acquisition, prompting advertisers to flee the platform and a rapid decline in revenue. Recently, brands have been returning to X as Musk’s influence grows in the Trump administration.

The seven banks that extended $13bn in loans to Musk to buy X kept the debt on their books for two years until they could sell it all at once last month, according to a source familiar with the transactions.

This was made possible after a surge in investor interest for exposure to AI companies along with X’s improved operating performance over the previous two quarters, among other factors, according to two people familiar with the matter.

After the merger, investors who bought the debt from the banks will profit, said Espen Robak, founder of Pluris Valuation Advisors, which specialises in illiquid assets. “For sure the debt is worth more now, if not fully paid off.”

Separately, a US judge on Friday rejected a bid by Musk to dismiss a lawsuit claiming he had defrauded former Twitter shareholders by waiting too long to disclose his initial investment in the company.

Specifics of deal remain unclear, including how X’s leaders will be integrated into new company

Elon Musk’s xAI firm buys social media platform X for $33bn

Tesla showrooms across the world are expected to face anti-Elon Musk protests on Saturday, as Musk’s senior role in the Trump administration has contributed to a European consumer backlash by some Tesla owners and prospective buyers.

It follows a 44% drop in Tesla sales in Europe on average last month, according to the research platform Jato Dynamics. Tesla’s European market share fell to 9.6% last month, the lowest it has registered in February for five years.

However, in the UK the Society of Motor Manufacturers and Traders reported an almost 21% rise in the number of new Tesla cars registered in February, with the Model 3 and Model Y proving the second and third most popular after the Mini Cooper.

Hundreds of Tesla owners got in touch with the Guardian to share how their feelings about owning Teslas have changed over time. Here are some their views.

‘Our daughter keeps asking us to sell it’

As a birthday present in 2019, my wife booked me a test drive for the Tesla Model S. I thought, “My God, what an amazing bit of kit”. It was very smooth, very fast, and I was keen to drive something more environmentally friendly. So we bought one.

We’ve now done about 80,000 miles on it – driving to Switzerland one year, Italy another, and last summer to the south of France. It’s been incredibly cheap, reliable, spacious, and with virtually no servicing.

But we are horrified by Trump and Musk’s extreme politics. Our 26-year-old daughter keeps asking us to sell it. When I saw Trump and Musk showcasing the models outside the White House, I felt a bit sick. I’m on Tesla owners’ forums, it’s getting very polarised.

We’re now in a series of constant dilemmas. I don’t want to have to put stickers on my Tesla like others are, with things like, “Love the car, not the CEO”. I feel very torn. The seconhand market has plummeted. Our warranty runs out in 2027, and then we’ll see. I never thought owning a car would be this difficult.

Jonathan, 64, Oxford

View image in fullscreen An anti-Musk poster in Broadstairs, Kent. Photograph: Krisztián Elek/SOPA Images/Rex/Shutterstock

‘Do we regret it? Morally, yes. Practically, no’

Last spring, my girlfriend and I were looking for a new car. A Tesla wasn’t even on our radar, but it turned out to be a cheaper option than the electric alternatives, and during a test drive it felt like the best car I’d ever driven … unfortunately. We thought, let’s go for it.

When Musk started campaigning for Trump last summer, I began to get questions about the car. Now it’s even worse. Friends or colleagues joke around, saying: “Ah! So you’re for Mr Musk?” But for me, having something that’s in any way a symbol of the far right is extremely painful. We feel a moral pain at being owners.

Do we regret it? Morally, yes. Practically, no. I hope this will be a passing moment, that the brand will outlive Musk. We hope someone can take over as CEO from Musk.

Patrick*, 40, Belgium

‘I’m a Tesla fanboy who jumped ship’

When I first drove a Tesla, while I was visiting Las Vegas for work in 2019, I was absolutely blown away. Going back to my Ford Mondeo felt like travelling back in time. I got the Model 3, which was a joy to drive, like a go-kart around the country roads, and later we got the Model Y.

I didn’t see the rightwing stuff coming. When Musk aligned with Trump, we thought, well … the car’s great. I was still prepared to stick with Tesla right up until Musk’s so-called “Roman salute”. The day I saw that, I thought, “Oh, man”.

My car was up for renewal a few weeks ago. I was on the phone to Tesla – they were really persistent, trying to convince me to stay – when a Polestar 4 [a competitor] drove by. I booked a test drive the same day and now I’m waiting for my Polestar 4 to arrive. I’m prepared to take a slight step down in software.

I was like, “Actually, I’ve got to stick to my morals, without that I haven’t got anything”. Even so, the Polestar is made in China, and we thought, is that any better? It’s almost impossible to have a fully ethical supply chain today. But Elon really put me off staying with Tesla – I’m a Tesla fanboy who jumped ship.

Henry, 50, Bedfordshire

‘There’s no way I’d sell my Tesla because of the CEO’s political views’

They are incredible cars. I bought a Model 3 Performance in 2019, and get software updates on a regular basis so the car consistently improves and offers new features. For me, once you’ve owned a Tesla, there’s no going back. It’s the iPhone equivalent of cars – everything works really well and is easy to use.

We’ve seen on the news the fools who think it’s OK vandalising Teslas. I get quips from friends, saying, “Has anyone keyed yours yet?” But this hasn’t worried me about my car. I don’t think these activists represent more than a fraction of a per cent of the population.

There’s no way I’d sell my Tesla because of the CEO’s political views. I disagree with some things Musk says and does, but I could say the same for everyone, and I don’t see what the fuss is about with Doge. The attacks he’s endured have been disproportionate and dishonest. What I worry about is how easily people can be programmed to hate a company to such an extent. It’s been pretty astonishing. Tom, 51, Spain

View image in fullscreen The burnt-out remains of a Tesla in the Steglitz district of Berlin. Police were investigating after four Tesla cars parked in different streets went up in flames earlier this month, Photograph: Omer Messinger/Getty Images

‘It really hurts to think that some of our savings have gone to someone like Musk’

I’m not rich, and my wife and I planned for a long time before we put our savings together to buy an electric vehicle. A car is not a status symbol for me, it’s a practical object and we wanted to reduce our carbon emissions as much as possible.

There’s no Tesla showroom in Slovakia yet, so a couple of years ago we had to get a company in Bratislava to import one from Austria. Tesla drivers here still wave at each other because we’re so rare. While buying it, a friend warned me about Musk. Back then I just thought, probably if you dig deep enough into any corporation you’ll find someone with unpleasant views.

But if I were buying a car today, there’s no way I’d get a Tesla. It really hurts to think that some of our savings have gone to a company led by someone like Musk. When he started supporting the AfD in Germany, or his associations with Trump, or what they’re now doing to USAid and federal employees – I’ve just felt worse and worse about giving the company thousands of euros.

It’s now worth a lot less than we paid. And it was a lot of money to us, we can’t just ditch it. It’s a symptom of a rather sad time that we live in, really.

Steven, 50, Slovakia

*Some names changed.

Some voice regret over Musk’s backing of Trump, but others say CEO’s views do not detract from the car’s appeal

‘Love the car, not the CEO’: how Europe’s Tesla owners feel about their cars - and Elon Musk

Advocates of “incel” ideology are rebranding as “Sub5s” and adopting the language of self-improvement to push their content on TikTok, according to a study.

Hateful material is banned from the social media site but accounts disseminating the beliefs are said to be hiding behind new terms and “socially palatable” language.

One of the fundamental tenets of the involuntary celibate, or incel, ideology is the belief that society is organised based on a looks-based hierarchy.

Pseudo-scientific theories about self-improvement and attractiveness that are often misogynist in nature are now being promoted and “normalised”, it is claimed.

Anda Iulia Solea, a researcher at the School of Criminology and Criminal Justice at Portsmouth, said this included a theory referenced in the Netflix drama Adolescence that suggests that 80% of women were attracted to 20% of men.

Another concept pushed by accounts on TikTok is that of the PSL scale, under which it is claimed that men can be ranked from one to eight according to attractiveness.

The PSL acronym comes from the first letters of three incel sites: PUAhate, Sluthate and Lookism.net.

Men are rated from being “PSL gods” (exceptionally attractive men) to “Chads” (generally attractive men) and finally “Sub5s” (unattractive men).

Men are encouraged to believe they can move up through the scale by a process called “looksmaxxing”, which can be as unproblematic as going to the gym or as damaging as changing the structure of the jaw using a hammer.

Such theories are said to be bound up with the belief that there is an ideal form of masculinity and that women are to blame.

When searching for the term “incel” on TikTok, an error message appears alongside a content warning about the phrase being associated with hateful behaviour. However, should a user search for “Sub5” content, they are offered a stream of material on the platform.

“We argue that facilitated by its rebranding through the frameworks of looksmaxxing and the PSL scale, misogynist incel ideology is being mainstreamed”, write Solea and Dr Lisa Sugiura.

“These mechanisms reframe overtly toxic, misogynistic and extremist rhetoric into socially palatable discourses of self-improvement and biological and racial essentialism, enabling its diffusion into broader digital spaces.”

TikTok has more than 1.9 billion users worldwide. A third of users in the UK are said to fall between the ages of 13 and 17, with 40.32% in the 18 to 24 age group.

The conclusions of the new paper, submitted to peer review in the Crime, Media, Culture journal, followed analysis of videos on five accounts on TikTok identified as promoting incel theories.

Solea said: “What they try to do is to attract wider audiences and also to be permitted on the platform and accepted by others through rebranding. So you’ll see Sub5s, you’ll see the PSL scale, and then also they are being less directly offensive and violent towards women.

“I’m just going to mention the Adolescence show because everybody’s talking about that: the 80:20 rule, where 80% of women go for the 20% of men. There are lots of these theories. You’re more likely to believe that women are bad when they show you that, rather than just saying: ‘Oh, women are such sluts,’ especially when those words and insults are not allowed on TikTok.”

A TikTok spokesperson said: “Misogyny is strictly prohibited on TikTok and we take swift action to remove content that violates our policies.

“We continually refine our moderation using technology and expert insights, and 93% of hateful content that was removed from our platform last quarter was removed before it was reported to us.”

Misogynist ideas being normalised as accounts rebrand, with focus on ‘Sub5s’ and ‘looksmaxxing’, say researchers

‘Incel’ accounts using self-improvement language to avoid TikTok bans – study

Blaring car horns on the three-lane A40 in west London are nothing new. However, on Saturday, they weren’t aimed at other drivers for a change; instead it was Tesla’s CEO, Elon Musk, who was the target of their anger as part of the “Takedown Tesla” movement, which has spread from the United States.

“It’s too overwhelming to do nothing,” said Louise Cobbett-Witten, who has family in the US. “There is real solace in coming together like this, everyone has to do something. We haven’t got a big strategy besides just standing on the side of the street, holding signs and screaming.”

The protest was part of a global day of protests planned under the umbrella of the Tesla Takedown movement. Organizers say the rallies will take place in front of more than 200 Tesla locations worldwide, including nearly 50 in California. Musk has not commented on the demonstrations.

Cobbett-Witten has family in Washington DC, and is planning to move back to the US. The 39-year-old NHS worker, who lives in south London, said: “The checks and balances have just failed. As much as people are trying to not say these words, they are fascists, they are white supremacists, they’re xenophobes, they’re misogynists, and they’re coming for everyone. And what starts in America comes over here.”

In the last fortnight, Tesla has responded to the protests outside its showroom and charging point in Park Royal by stationing a lone security guard at its gate, who said protesters had been friendly and peaceful. Dozens turned up on Saturday, their largest turnout since they began weeks ago.

While Tesla sales have fallen in Europe, they rose in the UK by more than a fifth in February, according to new car registration figures from the Society of Motor Manufacturers and Traders.

View image in fullscreen A protester at the demonstration on Saturday in London. Photograph: Thomas Krych/AP

Gay rights campaigner Nigel Warner MBE was attempting to hand out stickers to Tesla drivers entering or leaving the site in Park Royal. “This is the only thing you can do to make a difference,” the 77-year-old retired accountant from London said. “We are pretty helpless over here, the same as Europe, the only thing we can do is try to affect Tesla’s share prices and sales. It is something that has been done already with the Tesla sales dropping in many places. If he can’t sell his cars he is finished.”

Documentary film-maker Jim Green, 56, who lived in New York and Los Angeles before moving back to the UK 18 months ago, had worked with Musk on a film a decade ago.

Green said: “He was a different person, and he was very charismatic. He was talking when I was hanging out with him about the gigafactory where the batteries were being built, and he had a very compelling argument to make about the importance of batteries, an argument he made extraordinarily articulately. So I was very much leaning in to believing this guy wouldn’t turn into the fascist he has become.”

He said Musk had attacked typical Tesla buyers, whom he described as wealthy liberals who care about the environment: “Musk has gone out of his way to insult that exact group of human beings. I lived in LA during the time when everyone who was wealthy and liberal traded their Toyota Prius in for their Tesla during 2014-2015.”

Retiree Anne Kajava, 59, who is originally from Minnesota but lives in Cambridgeshire, said she was concerned about the United States’ change in policy on Europe and Ukraine.

She said: “I am truly concerned about a world war three. I am concerned about a civil war within the United States. You could say those are extreme views but Trump is talking about war. You have JD Vance in Greenland; it’s not impossible.”

Holding a banner attached to a Donald Trump toilet brush, she said: “I used to not hesitate to say I’m an American. Now actually I’m working with an acting coach, to fake a British accent so I can turn it on and off when I want to. I don’t want to be identified as American.”

Alarmed by the havoc the Tesla CEO has wreaked in the US, Britons and Americans are showing their displeasure

Protesters picket London Tesla showroom on global anti-Musk day

The motor manufacturer Hyundai faces legal action over allegations it failed to warn its customers that one of its most popular models of electric cars could be stolen “effortlessly in seconds”. Elliott Ingram, an expert in digital security, was stunned when a CCTV camera installed at his home recorded a hooded thief stealing his Hyundai Ioniq 5 car in less than 20 seconds.

The thief is believed to have used a device, available online, to mimic the car’s electronic key. It is the latest in a spate of thefts involving the vehicle, and many owners now resort to a steering lock. Ingram’s car was later recovered by police, but he is terminating the lease and seeking compensation from the motor company. He says the South Korean car giant should have alerted customers to the security vulnerabilities.

“This security system has been completely blown open, so anyone can attack it,” he said. “It’s no longer fit for purpose.”

Hyundai promotes the convenience of its digital and smart keys, which allow motorists to lock or unlock the doors and start the engine while carrying just a key fob or digital key. The new technology includes several security measures but can be defeated by criminal gangs.

Ingram discovered a key emulator device being marketed online for €15,000 (£12,500), which claimed to be able to gain entry to the Hyundai Ioniq 5 and a number of other models. The physical appearance of the device is based on the Nintendo Game Boy games console, and it can be operated in either English or Russian. “The device records the signal from the car and within 10 seconds to two minutes completely [duplicates] the native one,” says the website promoting the device. “The key can be stored in the device’s memory. Open, close, start and go at any time.”

The Observer revealed last year that the car industry ignored warnings more than a decade ago that keyless technology risked a surge in vehicle thefts. Experts warned keyless systems could be “successfully undermined” and vehicles stolen without forced entry.

Ingram, 38, who lives in north London, says Hyundai had warned him of other modifications the vehicle required, but failed to alert him and other motorists that its security systems were compromised. He intends to file a claim against the company over the theft unless he receives compensation, citing a breach of the Consumer Rights Act. In a letter to Hyundai, he wrote: “The thief was able to enter the vehicle and steal it effortlessly in seconds. Had I been informed of this specific security risk, I could have taken additional deterrent measures, such as installing a secondary immobiliser or using a steering lock.” He said the motoring manufacturer should “urgently inform” its customers of the vulnerability and ways to avoid theft until the security risk is remedied. Hyundai said in its response to Ingram that there was an industry-wide issue of organised criminal groups deploying electronic devices to illegally override smart key locking systems. It said it was working with police to understand more about the devices and track vehicles as possible.

View image in fullscreen CCTV footage shows Elliott Ingram’s Hyundai Ioniq 5 was stolen in less than 20 seconds. Photograph: Elliott Ingram

Hyundai said its vehicles met all the required security standards for the UK market, and it was developing updates to reduce the risk of keyless thefts. It said vehicles sold from February 2024 have the latest updates to “mitigate” risk and it was working on offering retrospective action to customers with cars placed in the UK market before February 2024.

Car thefts in England and Wales have risen from 70,053 in the year to March 2014 to 129,127 in the year to March 2024, a rise of 84%. According to the 2022 to 2023 Crime Survey for England and Wales, remote devices were used by offenders in 40% of vehicle thefts.

Under the crime and policing bill currently going through parliament, a ban on the electronic devices used to steal vehicles with keyless ignition will be introduced. Under the new laws, anyone who is found in possession of such a device, or to have manufactured, imported or distributed them, could receive five years’ imprisonment and an unlimited fine.

Ministers hope the new laws will disrupt the distribution of car theft devices to organised crime groups. The RAC says car thefts are driving higher insurance costs and the new laws “cannot come soon enough”.

Ken Munro, a founding partner of Pen Test Partners, a cyber security testing house, said car manufacturers were improving car security, but there was a lucrative criminal industry in designing products that exploit “cracks” in the technology. “The average car thief doesn’t need to be a hacker,” he said. “They just need to buy a product that allows them to hack and steal a car.”

Hyundai said it was working hard to combat the thieves, but was not planning a recall of any vehicles. “The software and hardware updates that we have and are implementing will significantly reduce the risk of this particular form of vehicle theft occurring in the UK,” said a spokesperson. “However, this is a race against very determined, well-funded individuals who stop at nothing to steal vehicles for multiple purposes.”

Elliott Ingram was stunned at how a thief made off with his Ioniq 5 deploying a device to mimic the smart key – and says he should have been warned

Hyundai facing legal action over car that can be stolen ‘effortlessly in seconds’

Protesters gathered outside Tesla showrooms around the world on Saturday as part of a global day of action against billionaire chief executive Elon Musk. The protest is part of the Tesla Takedown movement — a grassroots campaign that calls for people to boycott Tesla, sell their shares and join local picket lines in a peaceful demonstration against Musk’s influence

Protesters gathered outside Tesla showrooms around the world on Saturday as part of a global day of action against billionaire chief executive Elon Musk

Global protests against Tesla CEO Elon Musk – video

Hundreds of protests at Tesla showrooms are planned across the US and internationally on Saturday. Organizers have dubbed it Tesla Takedown’s Global Day of Action, the latest and largest in a series of demonstrations that began shortly after Donald Trump was inaugurated. Organizers say the rallies will take place in front of more than 200 Tesla locations worldwide, including nearly 50 in California alone.

The protesters’ goal is to send a message to the Trump administration that they’re against what the Tesla CEO, Elon Musk, is doing with the US federal government – laying off thousands of workers, cutting department budgets, giving fascist salutes and getting rid of entire agencies.

“Nobody voted for this, and nobody voted for Elon,” said Vickie Mueller Olvera, who has been organizing Tesla Takedown protests in the Bay Area. “He’s an unelected super-billionaire and he’s a thug.”

Tesla Takedown describes itself as a decentralized grassroots movement that will “protest Tesla for as long as Elon Musk continues to shred public services”. The group says on its organizing page that Musk is “destroying our democracy using the fortune he built at Tesla” and so, in turn, they are “taking action at Tesla”. Local organizers are planning their own demonstrations rather than coordinating with one national group.

Going after Musk’s bottom line

Olvera said that demonstrators were asking people to do three things: don’t buy a Tesla, sell off Tesla stock and join the Tesla Takedown protest movement.

Musk did not respond to a request for comment. On X, the social media platform he owns, he has said that the work he is doing for the federal government heading the so-called “department of government efficiency” (Doge) is “literally improving government efficiency, like the name says”.

As Musk has made his mark on the federal government and Tesla Takedown protests have grown, his car company’s financial performance has suffered. Since Trump’s inauguration, Tesla’s stock has plummeted more than 35%, sales have nosedived and the resale value for used Teslas appears to have hit an all-time low. Musk’s own net worth has fallen by 25% since then, about $100bn. Some Tesla owners have said they sold their cars after Musk gave a speech on inauguration day that included two Nazi-style salutes.

Earlier this month, Trump and Musk showcased Tesla vehicles on the driveway of the White House. The president touted the electric cars, staging a photo op inside a red Tesla sedan where he exclaimed, “Wow … everything’s computer,” and declared he would buy one. The two also condemned the anti-Tesla protests, and Trump said he would designate any violence against showrooms as domestic terrorism.

Jonathan V Last, the editor for the Bulwark, a politically conservative and anti-Trump website, has credited much of Tesla’s decline to the takedown protests.

“I would go so far as to say the Tesla takedown protests have been one of the most consequential protest movements in the shortest amount of time ever,” Last said during a Bulwark podcast that aired this week. “They helped to obliterate one-third of Elon Musk’s net worth, in like 100 days. And when you attack his money stack, you attack his power.”

The protest movement grows

Tesla Takedown protests started popping up in towns and cities across the US in mid-February. On social media, people posted photos of small gatherings in front of Tesla showrooms with demonstrators holding signs that depicted Musk as a Nazi and said things like “Diabolical Oligarchs Gaslighting Everyone” and “No Swastikars”, a new nickname for Teslas.

Olvera said that’s around the time she heard about a protest happening near her in the Bay Area, so she checked it out. She stood around with the other demonstrators in front of a showroom, holding her anti-Musk sign and joking around, and then something unexpected happened.

“People would drive by, you know, just passersby,” Olvera said. “And some of them just stopped and joined in.”

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

The protests kept growing each week with the same crowd coming back and newcomers participating. Olvera, who now organizes demonstrations, said that Tesla owners started joining the protests as they gained momentum and put bumper stickers on their cars saying “I bought this before Elon went crazy.” At one point, she said, even the owner of a Cybertruck – the massive boxy metallic trucks that have become synonymous with Musk – joined the protest.

“I see Elon Musk as hijacking our government, and he’s just dismantling everything that we hold dear,” Olvera said. “Everything that people have fought long and hard for, like social security, Medicare and Medicaid and our beautiful national parks … it just feels like the rug is getting pulled out from under us.”

Anti-Tesla sentiment – and protests – go global

The same pattern has taken shape across the US. Cities like New York and Chicago now see several hundred people joining in the weekly Tesla Takedown protests. On social media, the Bluesky account for Tesla Takedown has gained nearly 25,000 followers in just a matter of weeks. And one of the group’s top promoters, Alex Winter, an activist and the actor who played Bill in Bill & Ted’s Excellent Adventure, has amassed nearly 200,000 followers on Bluesky with his daily posts of anti-Musk protests.

Tesla Takedown organizers have been careful to distance themselves from the violent vandalism that has also been carried out against Tesla showrooms. Dozens of Tesla facilities have been attacked in the middle of the night with molotov cocktails, gunshots or graffiti saying things like “Fuck Elon” and “Tesla is Fascist.”

On Monday, the FBI launched a taskforce to go after the alleged vandals, calling them “domestic terrorists”, which Trump, Musk and the US attorney general, Pam Bondi, have also called them. The FBI director, Kash Patel, posted on X: “This is domestic terrorism. Those responsible will be pursued, caught, and brought to justice.” On Wednesday, police arrested a suspect in a high-profile attack against a Tesla showroom in Las Vegas that involved molotov cocktails. The suspect was charged with 15 counts.

Tesla Takedown organizers also condemn the vandalism. “We are a non-violent grassroots protest movement,” the group says. “We oppose violence and destruction of property. Peaceful protest on public property is not domestic terrorism.”

For Saturday, Tesla Takedown organizers say they expect to see the biggest turnout yet, with 213 planned protests from Waterville, Maine, to Honolulu, Hawaii – that includes 32 international locations in countries such as Australia, Finland and Germany.

“LFG!” the group posted to Bluesky. “Let’s send these techno-fascists, broligarchs and old fashioned Nazis a message loud and clear.”

Tesla Takedown’s Global Day of Action will be the largest in a series of demonstrations that began after Trump term 2.0

Global anti-Elon Musk protests planned at nearly 200 Tesla showroom locations

Monday

With all the other conflicts going on in the world right now, Elon Musk v Jeremy Clarkson is one we could probably safely afford to sit out. I am weak-willed, however, and click through to the story in the Times to test the principle of “my enemy’s enemy is my friend”. Musk is a real villain and Clarkson is just a motormouth, but I suspect the latter – for reasons of basic functionality and the sort of flippant humour with which Musk seems ill-equipped to cope – is capable of getting the better of the world’s richest man, should these latest remarks of Clarkson’s come to his attention.

Certainly, Clarkson has upset Musk before, exciting a libel action out of him in 2011 for comments Clarkson made about Tesla on Top Gear. A judge ruled against Musk then, a fact Clarkson is only too happy to remind him about now. In his Sunday Times column, Clarkson referred to the “pan-global decision to uncrowdfund Tesla”, and took a victory lap around news of Teslas being set on fire and vandalised the world over. “I’d love to remind all you Tesla drivers that I warned you 17 years ago that no good would come of your buying choice,” he wrote in his customary low-key style. “But you didn’t listen.”

Musk, as we know, is apoplectic about the hatred being directed towards Tesla right now, as is Donald Trump. Defending his buddy last week, the US president threatened, nonsensically, to send any “sick terrorist thugs” caught vandalising Teslas to “the prisons of El Salvador”. Clarkson, meanwhile, strove to point out that Musk, once popular on the left for promoting green vehicles, is now targeted by exactly the kind of people Clarkson hates, forcing him to a conclusion I fully support: that my enemy’s enemy is still kind of a jerk. There’s quite enough dislike to go round.

View image in fullscreen ‘Can we somehow sell the spring statement as “cheeky”?’ Photograph: Malcolm Park/Alamy Live News

Tuesday

It’s a hard video to watch: two RAF engineers caught on CCTV kicking poor old Paddington while he sits blamelessly on a bench in Berkshire. The violence escalates to such a degree that by the end of the assault, there’s nothing of the bear statue left but a hollow, metal shell. I guess at least it speaks to the fitness of the British armed forces.

The incident happened at 2am on a drunken night out this month, and on Tuesday the pair were ordered by a judge to pay £2,725 and do community service. It was the judge’s emotional remarks about the target of their vandalism, however, that made one realise not all criminal damage is equal. The two men had, according to Judge Goozee, attacked “a beloved cultural icon”, that represents “kindness, tolerance and promotes integration and acceptance in our society”. On the night in question, he said, they had behaved in a way that was “the antithesis of everything Paddington stands for”.

Which rather makes one hope that the next time someone in those parts rips the wing mirror off a Tesla or otherwise puts the boot into “everything it stands for”, they find themselves appearing before this same, symbolism-minded judge.

Wednesday

The discourse around the Netflix show Adolescence continues to rage and after receiving blanket praise for two weeks, some dissent is starting to show. All the usual caveats first: extraordinary television, brilliant performances, breathtaking vision. Adolescence is smartly and interestingly done, although it remains the case that if every girl who was called an ugly bitch online killed the boy or man who said it, there’d be no men left to make the TV show.

The presence of Brad Pitt as an executive producer, meanwhile, rings slight alarm bells. I’m not sure any of us needs the origins of misogyny explained, even tangentially, by a man whose kids have dropped his surname and whose ex-wife has alleged that he abused her. Pitt does, however, belong to what he has called a “really cool men’s group”, so I’m guessing that his “remarkably hands-on” involvement with the show might have brought to the project valuable insights into the self-victimising process.

Thursday

On the Americans in London social media feeds, people are sharing tips for getting in and out of the US without being cuffed and dragged to a detention centre in Louisiana. “If you plan to go, get a burner phone,” writes one frequent traveller which, blimey, I mean, I guess. Lots of people have popped up to report flying recently through Newark or Miami without incident, but the anxiety is enough to have changed the experience. Flight bookings between Canada and the US are down 70% and US tourist operators are braced for a worldwide downturn.

We’re flying to the US in May and, even as citizens, I’m nervous. Six years ago, while still on a green card, I got the “step this way” nod at JFK and was escorted to the exceedingly grim Congratulations, You Have Problems With Your Paperwork side room, where a woman relieved me of my phone and shouted at me and everyone else in the room to sit down and be quiet. For travellers to the US it seems a sensible precaution to have a lawyer’s number at the ready.

View image in fullscreen ‘Tell me again how you’d rather have a president than this comfy old scene.’ Photograph: Chris Jackson/Getty Images

Friday

I haven’t had to sleep with the hall light on since the Blair Witch Project came out in 1999 – oh man, that was a bad one because my flatmate was away that weekend and after seeing the film, I felt the dank presence of something watching me from the corner of my room. This week a babysitter in Kansas tried to vanquish a child’s fear of monsters under the bed by showing them there was nothing there. Unfortunately, in this case a 27-year old man called Martin Villalobos Jr was hiding under the bed. After a scuffle that knocked over the child, he was arrested and charged with aggravated battery and child endangerment. A mere externalisation of what, at the moment, we know to be true: the monsters are real.

Motormouth v Musk is a hard spectacle to resist and, in the end, it turns out the monsters are real

Digested week: I agree with Jeremy Clarkson – my enemy’s enemy is still kind of a jerk | Emma Brockes

Dolphins whistle. Whales sing. Fish croak, chirp, grunt, hum and growl. But in the chatter of the sea, one voice has been missing — until now.

Sharks have long been seen as the silent killers of the water. But scientists at the University of Auckland in New Zealand recently recorded a rig shark, or Mustelus lenticulatus, making a sharp clicking sound, most likely by snapping its teeth together, according to findings published in the journal Royal Society Open Science on Wednesday. They believe it’s the first time a shark has been recorded actively making noise.

The lead researcher, Carolin Nieder, first heard the sound while she was researching the hearing abilities of sharks. While she was handling one shark, it made a clicking, snapping sound similar to that of an electric spark, she said.

The noise came from a rig shark, a fairly small shark common in the waters around New Zealand that grows to up to five feet and mainly eats crustaceans. It is eaten by bigger shark species — and by New Zealanders, who use it to make fish and chips.

Researchers in New Zealand have recorded what they believe is the first instance of a shark actively making noise.

A Shark’s Sounds Are Recorded for What Is Believed to Be the First Time

From ancient star streams to the innards of white dwarfs, the Gaia space telescope has seen it all.

On Thursday, mission specialists at the European Space Agency sent Gaia, which is low on fuel, into orbit around the sun, and switched it off after more than a decade of service to the world’s astronomers. They also overwrote some of Gaia’s onboard memory with the names of team members, some of whom wrote personal goodbyes.

Gaia has charted the cosmos since 2014, creating a vast encyclopedia of the positions and movements of celestial objects in our Milky Way and beyond. It is difficult to capture the breadth of development and discovery that the spinning observatory has enabled. But here are a few numbers: nearly two billion stars, millions of potential galaxies and some 150,000 asteroids. These observations have led to more than 13,000 studies, so far, by astronomers.

After more than a decade of mapping the stars, the European spacecraft was shut down on Thursday. But its legacy lives on.

Gaia to Shut Down After More Than a Decade of Mapping the Milky Way

First Close-Up of Star Outside Our Galaxy Shows a Giant About to Blow

Astronomers zoomed in on a stellar behemoth in the Larger Magellanic Cloud, a galaxy that orbits about 160,000 light-years from the Milky Way.

By

Breaking news, videos and photos on space and astronomy, including NASA, SpaceX, black holes, the moon, Mars, Jupiter, Saturn, the International Space Station, the sun, exoplanets and the stars.

Space and Astronomy

Donald Trump has said he would be willing to reduce tariffs on China to get a deal done with TikTok’s Chinese parent company ByteDance to sell the social media app used by 170 million Americans.

He acknowledged the role China would play in any agreement. “With respect to TikTok, and China is going to have to play a role in that, possibly in the form of an approval, maybe, and I think they’ll do that,” Trump told reporters on Wednesday. “Maybe I’ll give them a little reduction in tariffs or something to get it done,” he added.

Trump’s comment suggests the sale of TikTok’s is a priority for his administration and important enough to use tariffs as a bargaining chip with Beijing.

TikTok did not immediately comment.

ByteDance has a 5 April deadline to find a non-Chinese buyer for TikTok or face a US ban on national security grounds that was supposed to have taken effect in January under a 2024 law.

The move is the result of concern in Washington that TikTok’s ownership by ByteDance makes it beholden to the Chinese government and that Beijing could use the short video app to conduct influence operations against the US and collect data on Americans.

In February and earlier this month, Trump added levies totalling 20% to existing tariffs on all imports from China.

Getting China to agree to any deal to give up control of a business worth tens of billions of dollars has always been the biggest sticking point to getting any agreement finalised. Trump has used tariffs as a bargaining chip in the TikTok negotiations in the past.

On 20 January, his first day in office, he warned that he could impose tariffs on China if Beijing failed to approve a US deal with TikTok.

Vice-president JD Vance has said he expects the general terms of an agreement that resolves the ownership of the social media platform to be reached by 5 April.

Reuters reported last week that White House-led talks among investors are coalescing around a plan for the biggest non-Chinese backers of ByteDance to increase their stakes and acquire the video app’s US operations, according to two sources familiar with the discussions.

The future of the app used by nearly half of all Americans has been up in the air since a law – passed with overwhelming bipartisan support – required ByteDance to divest TikTok by 19 January.

The app briefly went dark in January after the US supreme court upheld the ban, but flickered back to life days later once Trump took office. Trump quickly issued an executive order postponing enforcement of the law to 5 April and said last month that he could further extend that deadline to give himself time to shepherd a deal.

The White House has been involved to an unprecedented level in the closely watched deal talks, in effect playing the role of investment bank.

Free speech advocates have argued that the ban unlawfully threatens to restrict Americans from accessing foreign media in violation of the first amendment of the US constitution.

US president acknowledges Beijing’s role in any sale as app’s owner faces 5 April deadline to find non-Chinese buyer or be hit with US ban

Trump floats easing tariffs on China in return for TikTok deal

Fortunately, the Webb telescope, launched in 2021, has come to the rescue.

Heidi Hammel, an astronomer at the Association of Universities for Research in Astronomy and another of the study’s authors, has been studying Neptune since the 1980s. She thought that if Webb “was powerful enough to see the earliest galaxies in the universe, it’d better be powerful enough to see things like aurorae on Neptune,” she said. “And by golly, it was.”

Using the telescope’s Near-Infrared Spectrograph, astronomers caught Neptune’s infrared auroras in June 2023. And unlike Earth’s, they dance not above the poles, but its mid-latitudes. That’s because Neptune has a wonky magnetic field that is tilted by 47 degrees from the planet’s spin axis.

The new Webb observations also reveal why Neptune’s auroras have been invisible until now. Nearly 40 years ago, Voyager 2 recorded a temperature of around 900 degrees Fahrenheit for Neptune’s upper atmosphere. But the Webb telescope shows that the temperature has dropped, to close to 200 degrees. That lower temperature means the auroras are dimmer.

In fact, Neptune’s aurora is glowing “with less than 1 percent of the brightness we expected, explaining why we haven’t seen it,” said James O’Donoghue, a planetary astronomer at the University of Reading in England and one of the study’s authors. “However, that means we now have a new mystery on our hands: How has Neptune cooled down so much?”

With the detection of Neptune’s strange light show, answers may be forthcoming.

“Auroras are like a TV screen,” said Leigh Fletcher, a planetary scientist at the University of Leicester in England and one of the study’s authors. They are “allowing us to watch the delicate dance of processes in the magnetosphere — all without actually being there.”

The James Webb Space Telescope identified the lights in the distant planet’s atmosphere, which could not be seen by earlier telescopes or spacecraft.

NASA’s Webb Telescope Spots Auroras on Neptune for the First Time

Masaki Kashiwara, a Japanese mathematician, received this year’s Abel Prize, which aspires to be the equivalent of the Nobel Prize in math. Dr. Kashiwara’s highly abstract work combined algebra, geometry and differential equations in surprising ways.

The Norwegian Academy of Science and Letters, which manages the Abel Prize, announced the honor on Wednesday morning.

“First of all, he has solved some open conjectures — hard problems that have been around,” said Helge Holden, chairman of the prize committee. “And second, he has opened new avenues, connecting areas that were not known to be connected before. This is something that always surprises mathematicians.”

Mathematicians use connections between different areas of math to tackle recalcitrant problems, allowing them to recast those problems into concepts they better understand.

Masaki Kashiwara received the honor, often regarded as the Nobel Prize in mathematics, for work that combined different mathematical fields to solve challenging problems.

Masaki Kashiwara, Japanese Mathematician, Wins 2025 Abel Prize

The government’s ambition to boost efficiency by embedding AI in all aspects of its work risks being undermined by out-of-date technology, poor quality data and a lack of skilled staff, an influential Commons committee has warned.

The report by the cross-party public accounts committee (PAC) found that more than 20 government IT systems identified as “legacy”, meaning out of date and unsupported, have yet to be given funding to improve them.

Government research cited by the PAC in the report found that almost a third of central government IT systems met this definition in 2024.

Keir Starmer’s government has repeatedly stressed its desire to increase economic growth through the mass take-up of AI systems, including in the public sector.

An official plan for the technology published in January called for the government to “rapidly pilot” AI-powered services, saying this would both increase productivity and improve people’s experience of dealing with officialdom.

In a speech earlier this month, Starmer said AI should replace the work of government officials where it can be done to the same standard, with 2,000 new tech apprentices to be recruited to the civil service.

However, the PAC report also warned about “persistent digital skills shortages in the public sector”, in part because of civil service pay levels “that are uncompetitive with the private sector”.

The Department for Science, Innovation and Technology (DSI), which is spearheading AI roll-outs in government, says it has previously recognised many of the same worries in two reports published in January, the State of Digital Government and Blueprint for a Modern Digital Government.

However, the PAC report makes a series of new recommendations, including setting a six-month deadline for the department to specifically set out how it will fund replacements for the highest-risk legacy technology, and to also assess the costs of failing to take action.

It also calls for action to improve public confidence in the transparency and standards for how AI is used in government, saying that as of January this year, just 33 official records had been published setting out algorithm-assisted decisions and how they are made. The report recommends this be sped up.

The report also identifies a lack of coherent systems to learn from the mass of different AI pilots taking place across government, calling for action to address this.

Sir Geoffrey Clifton-Brown, the Conservative MP who chairs the PAC, said: “The government has said it wants to mainline AI into the veins of the nation, but our report raises questions over whether the public sector is ready for such a procedure.

“A transformation of thinking in government at senior levels is required, and the best way for this to happen is for digital professionals to be brought round the top table in management and governing boards of every department and their agencies. I have serious concerns that DSIT does not have the authority over the rest of government to bring about the scale and pace of change that’s needed.”

A government spokesperson said: “These findings reflect much of what we already know, which is why we set out a bold plan to overhaul the use of tech and AI across the public sector – from doubling the number of tech experts across Whitehall, to making reforms to replace legacy IT systems more quickly and building new tools to transform how people interact with the state.”

Public accounts committee also flags ‘persistent digital skills shortages’ and uncompetitive civil service pay rates

Government AI roll-outs threatened by outdated IT systems

Welcome to Opt Out, a semi-regular column in which we help you navigate your online privacy and show you how to say no to surveillance. The last column covered what to do with your 23andMe account after the company filed bankruptcy. If you’d like to skip to a section about a particular tip, click the “Jump to” menu at the top of this article.

If you’re a visa or green card holder with plans to travel to the US, reports of people being turned away at airports over messages found on their devices might be prompting you to second-guess your travel plans. You might be asking whether Customs and Border Protection (CBP) can search your phone, whether you can opt out and what you should do to minimize your risks.

The short answer is that yes, CBP can search your devices. Constitutional protections are generally weaker at US borders, including airports. You can try to opt out, but depending on your specific circumstances, you might not be willing to risk the potential ramifications of not complying, which can include the confiscation of your devices.

Privacy experts say everyone should conduct a personal risk assessment – which should include your immigration status, travel history and what data you might have on your phone. There’s not a one-size-fits all solution because data that may seem sensitive to some may not be to others, depending on your circumstances. That assessment might affect your calculus of whether to push back if CBP attempts to search your phone, for instance, or how much you want to lock down your devices before heading to the airport.

While CBP said it only searched about 47,000 devices of the 420 million people who crossed the US border in 2024, experts the Guardian spoke to say border enforcement has been unpredictable under the Trump administration, so figuring out whether you’re at risk of a device search is not as straightforward as it once was. French officials said a French scientist was recently turned away at an airport in Texas because immigration officers found texts that were critical of Trump on his phone.

“The super-conservative perspective is to assume they are completely unhinged and that even the most benign reasons for travel are going to subject non-citizens to these device searches,” said Sophia Cope, a senior staff attorney at the Electronic Frontier Foundation (EFF), a non-profit digital rights group.

If you’re a US citizen, you must be admitted into the country. That said, some jurisdictions allow CBP to work with the FBI or local police to advance domestic investigations, so there are still some risks of your devices being searched for domestic reasons.

There are steps you can take to make it harder for CBP officers to access your device and the data on it. So what should you do to protect the data on your phone from being searched? The main thing is to prepare ahead of heading to the airport. Here is what you should be thinking about:

Decide if you will comply with a phone search

Before you travel, start to prepare for the possibility of being pulled into secondary screening. First, you should decide if you’re going to comply if an immigration officer asks if they can search your device. They may ask for your phone password or for you to unlock the device. Ideally, you would unlock the device yourself and not share your password. You can decide not to give your consent, but that does come with its own risks.

From a guide to the border from the EFF: “This presents a no-win dilemma. If a traveler complies, then the agents can scrutinize and copy their sensitive digital information. If a traveler declines, then the agents can seize their devices and subject the traveler to additional questioning and detention.”

It’s possible that, if you refuse the search, the officer might decide that you are not worth the trouble because you do not present a high risk. They might let you go. On the flip side, though, declining could make the process longer or could result in border agents confiscating your device. If they do confiscate your device, make sure you ask for a property receipt so you can document that they have your device when you try to get it back. Even if you don’t give up your password, immigration officials can use various tools on your device to unlock it themselves. They can also try to guess your password, so make sure you have a strong and long password.

There are a lot of reasons you might not want to risk being held back longer than you already have been or risk having your device confiscated – including that you don’t know how long border agents will keep your phone.

If you plan to comply with a phone search to avoid any further complications, your phone might be searched either manually or with forensic tools. It is worthwhile to prepare for both types of searches.

Turn your phone and Face ID off before entering the US

The EFF recommends turning your devices completely off before entering the US. This could return your phone or laptop to a heightened security state and, as a result, could make it harder for anyone to break the encryption on your device.

Privacy advocates also recommend making sure your device requires a password to decrypt or unlock. If you use Face ID or a fingerprint to unlock your phone, for instance, it would be easier for an officer to use it to gain access to your device.

Do not wipe your phone

You might think the most protective options are to completely wipe your phone before traveling, use a burner or travel without a phone. But the EFF’s Cope said that could actually raise suspicions.

“People are damned if they do and damned if they don’t,” Cope said. “If you cross the border with no data on your device, that itself can be seen as suspicious.”

Instead, if you want to seem cooperative but do have data or texts stored on your phone that you wouldn’t want to be accessed, Cope suggests deleting that information selectively rather than wiping your whole device.

skip past newsletter promotion Sign up to TechScape Free weekly newsletter A weekly dive in to how technology is shaping our lives Enter your email address Sign up Privacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy . We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply. after newsletter promotion

Encrypt your data and use a strong password

The most important step to take before you travel is to encrypt the data on your device, which is different than using encrypted messaging services like Signal. Device encryption can make it harder for CBP officers to access files on your phone or laptop or recover deleted files, even if they confiscate the device and subject it to sophisticated forensic tools.

Fortunately, all recent models of both iPhones and most Android phones come with full-device encryption automatically turned on. On an Android, double check that yours is on in the “advanced settings” tab of your “Security” menu. You will want to choose a strong password that is not easy to guess so CBP can’t walk in the front door to your device. Here’s a good primer on how to make a strong password.

“This encryption is only as good as the encryption passphrase someone uses on their device, though,” said EFF senior staff technologist Bill Budington. “So the best advice is to choose a strong, nine- to 12-random-character (or four- to five-word) passphrase for the device, and make sure that biometric unlocks like Face ID or Touch ID are turned off when going through sensitive areas like checkpoints or somewhere your device could be confiscated.”

Laptops, on the other hand, do not all come with full-device encryption. You can use the encryption tools some of them offer to encrypt your data. MacOS has a tool called FileVault, which you can access by searching for it in the top right corner of your screen, and some Windows computers come with a tool called Bitlocker, which can be used to encrypt your device. The EFF has a full list of tools you can use on various operating systems here.

For those of you traveling with a device owned by your employer or someone else, you will want to make sure to have a conversation with them before you travel to ensure your device is sufficiently protected.

How to securely delete your data

In addition to encrypting your devices, you should delete any specific texts, apps, photos, etc that you feel are sensitive or you wouldn’t want a government agent to see.

Securely deleting this data requires a few steps and comes with limitations. If you are not wiping your phone entirely, as that may raise suspicions, you will probably opt to delete specific files. That more practical option may be effective for a manual or cursory search but may not be sufficient in the event of a more advanced search by US immigration personnel. Files may not be fully deleted, or there may be references to these files that remain on your device.

On top of ensuring your device is encrypted, you will want to make sure that you’ve deleted your files from any trash folders as well. On iMessage, for instance, if you click on “filters” in the top left corner you’ll find a “recently deleted” folder. Make sure you’ve cleared texts from there as well. On iPhones, once a file is deleted from both the main iMessage interface and the “recently deleted” file, it is permanently deleted, according to the company.

Cope suggests pre-emptively deleting some apps your don’t want to be searched. This protective method is imperfect because an advanced search could reveal that an app was installed, but it would be a way to avoid having your WhatsApp messages searched, for instance, in the case of a manual search.

Move things on to a cloud storage server

During law enforcement searches inside the boundaries of the US, a cloud storage server is not more protected than your devices. At the border, however, there are currently policies in place that prohibit CBP from searching online cloud services. In practice, that means that immigration officers will have to put your phone in airplane mode before searching it.

“They do specifically say officers are only authorized to look at data that are ‘resident on the device’,” Cope said. “So that is data that is actually on the hard drive of your phone, laptop or camera. They’re supposed to disconnect it from the internet, if it’s an internet-connected device.”

If you have data that you don’t want to or can’t delete permanently for any reason, you can delete it off your device and store it on your cloud storage like iCloud, Google Drive or Microsoft One Drive.

This is a high-level guide that may not touch on the specifics of your situation. For a full comprehensive guide on how to protect your devices at US borders, please visit the Electronic Frontier Foundation.

With reports of people being turned away at airports over messages found on devices, here’s what to do to minimize risks