Meta AI agent’s instruction causes large sensitive data leak to employees
An AI agent instructed an engineer to take actions that exposed a large amount of Meta’s sensitive data to some of its employees, in the latest example of AI causing upheaval in a large tech company,The leak, which Meta confirmed, happened when an employee asked for guidance on an engineering problem on an internal forum,An AI agent responded with a solution, which the employee implemented – causing a large amount of sensitive user and company data to be exposed to its engineers for two hours,“No user data was mishandled,” a Meta spokesperson said, and they emphasised that a human could also give erroneous advice,The incident, first reported by The Information, triggered a major internal security alert inside Meta, which the company has said is an indication of how seriously it takes data protection.
This breach is one of several recent high-profile incidents caused by the increasing use of AI agents within US tech companies.Last month, a report from the Financial Times said Amazon experienced at least two outages related to the deployment of its internal AI tools.More than half a dozen Amazon employees later spoke to the Guardian about the company’s haphazard push to integrate AI into all elements of their work, leading, they said, to glaring errors, sloppy code and reduced productivity.The technology that underlies all these incidents, agentic AI, has evolved rapidly over the past months.In December, developments in Anthropic’s AI coding tool, Claude Code, triggered widespread hubbub over its ability to autonomously book theatre tickets, manage personal finance, and even grow plants.
Soon after was the advent of OpenClaw, a viral AI personal assistant that ran on top of agents such as ClaudeCode but could operate entirely autonomously – trading away millions of dollars in cryptocurrency, for example, or mass-deleting users emails – leading to heady talk about the advent of AGI, or artificial general intelligence, a catch-all term for AI that is capable of replacing humans for a wide number of tasks.In the weeks that followed, stock markets have wobbled over fears that AI agents will gut software businesses, reshape the economy and replace human workers.Tarek Nseir, a co-founder of a consulting company focused on how businesses use AI, said these incidents showed that Meta and Amazon were in “experimental phases” of deploying agentic AI.“They’re not really kind of standing back from these things and actually really taking an appropriate risk assessment.If you put a junior intern on this stuff, you would never give that junior intern access to all of your critical severity one HR data,” he said.
“The vulnerability would have been very, very obvious to Meta in retrospect, if not in the moment.And what I can say and will say is this is Meta experimenting at scale.It’s Meta being bold.”Jamieson O’Reilly, a security specialist who focuses on building offensive AI, said AI agents introduced a certain kind of error that humans did not – and this may explain the incident at Meta.A human knows the “context” of a task – the implicit knowledge that one should not, for example, set the sofa on fire in order to heat the room, or delete a little-used but crucial file, or take an action that would expose user data downstream.
For AI agents, this is more complicated.They have “context windows” – a sort of working memory – in which they carry instructions, but these lapse, leading to error.“A human engineer who has worked somewhere for two years walks around with an accumulated sense of what matters, what breaks at 2am, what the cost of downtime is, which systems touch customers.That context lives in them, in their long-term memory, even if it’s not front of mind,” O’Reilly said.“The agent, on the other hand, has none of that unless you explicitly put it in the prompt, and even then it starts to fade unless it is in the training data.
”Nseir said: “Inevitably there will be more mistakes.”
UK government borrowing costs soar to highest since 2008 amid Iran war
Investors wary of the impact of the Iran conflict dumped UK government bonds on Friday, pushing the yield, or interest rate, on 10-year borrowing to its highest level since 2008.The market move followed the Bank of England’s decision on Thursday to leave interest rates on hold and hint at a future increase. By Friday morning, markets were pricing in as many as three interest rate rises in 2026.Higher gilt yields create a headache for the chancellor, Rachel Reeves, by pushing up the cost of servicing the government’s debt pile.The 10-year yield was 4
UK borrowing costs hit highest since 2008, as money markets predict three interest rate rises this year – business live
A key measure of UK government borrowing costs has hit its highest level since 2008, as traders bet that the energy price shock will push up interest rates.The yield, or interest rates, on 10-year UK gilts has risen to 4.927% this morning, a rise of 9 basis points (0.09 percentage points). That’s the highest level since July 2008, in the run-up to the financial crisis
Fire experts ‘kept awake’ over growing hazard of lithium-ion batteries
Lithium-ion batteries represent a new technological hazard that one fire science expert has said keeps him awake at night, as fire service chiefs warn the ubiquity of the batteries in everyday products is outpacing public understanding and safety regulations.The blaze that devastated a historic building in Glasgow and resulted in the continuing closure of Central Station, Scotland’s largest rail interchange is believed to have started in a shop selling vapes, which are powered by lithium-ion batteries. The latest data reveals a sharp increase in battery-related fires across Scotland, while firefighters in London attend an e-bike or e-scooter fire every other day.Paul Christensen, a professor of pure and applied electrochemistry at the University of Newcastle, underlined that, while the probability of a fire from a lithium-ion battery is very low, the hazard is “very, very high, as we’ve seen with this fire in Glasgow”.Guillermo Rein, a professor of fire science at Imperial College London, said: “It’s a new technology that comes with an unintended new hazard, that keeps me awake at night
Essex police pause facial recognition camera use after study finds racial bias
Essex police have paused the use of live facial recognition (LFR) technology after a study found cameras were significantly more likely to target black people than people of other ethnicities.The move to suspend use of the AI-enabled systems was revealed by the Information Commissioner’s Office (ICO), which regulates the use of the technology deployed so far by at least 13 police forces in London, south and north Wales, Leicestershire, Northamptonshire, Hampshire, Bedfordshire, Suffolk, Greater Manchester, West Yorkshire, Surrey and Sussex.The ICO said Essex police had paused LFR deployments “after identifying potential accuracy and bias risks” and warned other forces to have mitigations in place. LFR systems are either mounted to fixed locations or deployed in vans. In January, the home secretary, Shabana Mahmood, announced the number of LFR vans would increase five-fold, with 50 available to every police force in England and Wales
For Mexico and Canada, injuries are striking just as World Cup hosting duty looms
When Marcel Ruiz slumped into the grass of San Diego FC’s Snapdragon stadium late in the first half of Toluca’s Concacaf Champions Cup game last Wednesday, he seemed to already know. He covered his mouth with his left hand and clutched his right knee – first the back of it, then the front – with his other hand. He turned his head every which way, perhaps hoping that he might scan something or someone who would tell him that this was not in fact happening. That his World Cup on home soil was not already over three months before it was to even start. That Mexico’s injury crisis had not just deepened further
No Limit: can rap mogul Master P really become an elite basketball coach
The 55-year-old has is an assistant coach at the University of New Orleans. Now he believes he can take the step up to the top of his sportYou are Arizona State athletics director Graham Rossini, more of a forward-thinking sports executive than a classic campus administrator. The Sun Devils basketball team have just staggered through another middling season, missing the NCAA tournament for a third straight year. You’ve just fired coach Bobby Hurley, but the vacancy isn’t what anyone in the sport would call coveted – not compared to a blue-blood program like Duke or Kentucky, or even the cross-state rival Arizona Wildcats men’s basketball, standard-bearer of the old Pac-10.You could hire another hardwood hero like Hurley, a Duke Blue Devils men’s basketball supervillain whose winning pedigree as a player surfaced only in flashes over 11 uneven years on the sideline
Work from home and slow down on the road: world’s energy watchdog advises emergency measures as oil prices rise
High charges, poor service: NCP hits the skids as drivers change habits
Shrinkflation takes a bite out of Easter eggs as shoppers pay more for less
Markets keep the faith – but oil staying above $100 could test that optimism | Nils Pratley
US moves to soften capital rules: ‘Big banks can declare mission accomplished’
Bank of England tipped to raise UK interest rates twice this year to fight inflation shock from Middle East crisis, as oil and gas prices rise – as it happened