Anthropic gave up its veto over what its safety testers publish.
Anthropic said on Thursday that six researchers from the nonprofit tester METR now work inside the company. They get laptops, desks and internal tools roughly equal to Anthropic's own risk team. Anthropic can only blank out narrow items like customer data, and METR may say publicly whether that was fair.
Why it mattersUntil now, outsiders mostly saw what AI labs chose to show them. If METR's reports turn critical, you get a rare independent check on the safety claims behind tools you use.
Read in the app — free, in 9 languages
Related stories
TSMC kept more than half of every dollar it earned as profit.2026-10-15Anthropic told investors it turned a profit on nearly $17 billion.
2026-10-15Nvidia will give British labs $250 million in computing for AI safety.
2026-10-14Microsoft fixed more flaws in one month than in all of 2020.
2026-10-14ASML says AI demand will outrun its chip machines for two years.
2026-10-14