Google, OpenAI, Anthropic advance industry-run AI safety body
The group could launch by year-end or early 2027 and would set rules for model testing, audits, and incident reporting.
Tech stories and the sources behind them.
402 stories
Newest first
Sunday
The group could launch by year-end or early 2027 and would set rules for model testing, audits, and incident reporting.
Axios reported Trump personally invited him; the president told reporters he opposes slowing AI development.
Problems with the robot's intricate hands, automated equipment, and suppliers are complicating the push to manufacture…
The Information reports the year-old AI chip startup, founded by three former Tesla Dojo leaders, is in advanced…
DeepMind chief Koray Kavukcuoglu says the model is in post-training as Google works to close the gap with Anthropic…
The Palantir-style structure would preserve the seven co-founders' control despite their relatively small ownership…
Enthusiasm around its Jev model is driving discussions of valuations above $10 billion, per The Information.
The viral post says China's AI process now runs with zero dependency on the United States, echoing predictions the…
OpenAI says it cannot reconnect the files to the affected users, complicating direct notification.
The federal jury found unintentional infringement; Apple plans to appeal, so the award is not final.
Saturday
Every's Opus 5.5 vibe check argues lower per-token costs make extra code reviews and speculative pull requests worth…
Gergely Orosz said the feature copies Kiro and GitHub Workflows, which he argues proved dead on arrival outside…
Showing 373 to 384 of 402 stories