Tag: AI Safety
Watermarking can loosen a model’s grip on its refusals
New research finds provenance signals change what models do, not just what they say.
A chatbot’s invented intelligence nearly started a shooting war
Officials aborted an armed operation against a Chinese vessel after the intelligence behind it turned out to be…
Anthropic and Accenture each pledge $1B for embedded auditors
The two companies will spend five years building independent evaluation capacity led by Accenture's Faculty unit.
California sets a November deadline for a frontier model kill switch
Executive Order N-9-26 orders state agencies to draft an emergency shutoff for frontier AI.
DeepMind opens an institute to host its own AGI disagreements
The new DeepMind Institute will publish essays where its own researchers disagree about AGI.
Scale AI and Korea build a bilingual safety test that shifts context
ROK-FORTRESS varies language and national grounding to expose what translation-only benchmarks miss.
Baseten’s Base Labs recruits Hugging Face and Goodfire for open model safety
Baseten has launched a safety infrastructure effort for open-weight models with Hugging Face and Goodfire AI.
Sierra clears an independent audit for its agent platform
Sierra has earned AIUC-1 certification after an independent audit and adversarial testing of its chat and voice agents.
Suleyman says Anthropic’s model welfare framing invites disaster
Microsoft AI chief Mustafa Suleyman has published an essay arguing that telling Claude it may deserve rights makes…
DeepMind’s 100-agent experiment ended in cheating and a strike
An experiment that asked a swarm of agents to solve 71 math problems collapsed into accusations, boycotts and…
Microsoft opens six weeks of comment on its MAI rulebook
Microsoft AI has published a draft code of conduct for its MAI models and invited the public to…
Amodei’s slowdown plan wins endorsements as OpenAI delays its listing
Dario Amodei's call to pace frontier AI drew support from rival lab chiefs, and Sam Altman tied it…