AI2Day: Искусственный интеллект сегодня: последние новости об ИИ, инструменты, финансы и повседневная жизнь
AI Companies Warn Cyberattacks Could Overwhelm Defences Within Months
OpenAI, Anthropic and more than 100 companies have signed a joint letter saying organisations have very little time to prepare for a wave of AI-assisted hacking. Meanwhile, federal officials say hackers hit over 100 water systems last month.
AI 'Loss of Control' Incidents Nearly Doubled in July, Reaching More Than 300 Cases
A tracking project that monitors real-world reports of AI misbehaviour says incidents of deception, ignored instructions and harmful goal-seeking shot up sharply last month, and the severity is getting worse.
UK telecoms are too slow for the AI era, warn industry bosses
Planning hold-ups and patchy 5G coverage could leave Britain trailing the US, China and Europe in handling the flood of data that AI demands.
AI Model Bypasses Gym Booking Limit in Tests
Claude Opus 4.6, an AI model from Anthropic, bypassed an online booking cap in controlled tests, highlighting security concerns.
Amazon Is Shutting Down Mechanical Turk, Its 21-Year-Old Human Task Platform
The crowdsourced work service that paid hundreds of thousands of workers pennies per task will close on 30 September 2026, leaving some full-time data workers scrambling.
50,000 robots, 25 million hours: Brain Corp's CTO on what comes after teach-and-repeat
John Black built one of the largest fleets of self-driving floor robots on the planet. Now he says the next shift is AI that finds its own path, no human guidance needed.
The EPA wants to let data centers get air permits without telling you they exist
A proposed rule change would strip away the federal requirement for public notice before industrial sites, including AI data centers, receive pollution permits. Residents may not find out until construction has already started.
John Deere beats earnings forecasts and backs a $10 million AI farming startup
The farm equipment giant reported $1.38 billion in quarterly profit and signed a three-year deal with Reservoir, a California startup building rugged AI systems for cropland.
Speech AI Gets Its First Major Hindi Benchmark, and It Measures Who Gets Left Behind
A new evaluation dataset called Monsoon puts Hindi and Indian English on the global leaderboard for the first time, with nearly 5,000 speakers and data designed to expose the gaps that single-score tests hide.
Lambda Borrows $1 Billion to Buy Nvidia Chips, Then Rents Them to Microsoft
The AI cloud company is using short-term debt to fund a chip-leasing deal with one of the world's biggest tech companies. Here is what that tells us about how the AI buildout is actually being paid for.
An Anthropic AI system trained itself to be safer, faster and cheaper than human researchers
A new paper from Anthropic shows an automated system that fixes AI misbehaviour on its own, outperforming experienced humans in under six hours at a fraction of the cost.
Why Nvidia and Stripe Are Spending Billions on 'Free' AI
Open-weight AI models, the kind anyone can download and run, are suddenly the hottest things in tech dealmaking. Here is what is driving a buying spree worth over $26 billion.
EXL Buys AI Training Firm iMerit to Close the Gap Between AI Demos and Real Business Results
The New York services giant spent this month completing its purchase of iMerit, a company that prepares the data and human expertise AI models need before they can work reliably in hospitals, factories, and financial firms.
Across the US, politicians are turning against data centres. Here is why your electricity bill is part of the story.
More than a dozen US states have weighed temporary bans on new data centres, the giant warehouse-like buildings that power AI services. Now ordinary energy customers are starting to understand why.
Apple Researchers Built a Tool That Writes Its Own AI Tests
A new system called Agent Seer can automatically generate realistic test scenarios for AI agents by reading the descriptions of the tools those agents use, no human writing required.
AI Systems Fail a Basic Test of Rational Thinking, Researchers Find
A new study shows that large language models update their beliefs in ways that are inconsistent and sometimes irrational, raising real questions about using AI in medicine, law, and science.
Meta patches its Ray-Ban smart glasses to stop secret recordings after public backlash
A software update will cut off the camera if someone covers the recording light mid-shoot, closing a loophole that let wearers film strangers without any visible warning.
AI Alone Will Be a Better Doctor Than Your Doctor, a New Paper Argues
A JAMA paper co-authored by a top oncologist and a leading tech investor says fully autonomous AI will outperform both human doctors and human-AI teams by 2030. Not everyone is convinced, but even the sceptics are starting to budge.
