Anthropic has circulated a draft AI safety bill for Congress to consider. It would allow a model's release to be blocked when a foreseeable, imminent risk could kill more than 50 people or cause over $1 billion in property damage, if the risk stems from a biological weapon, a model evading developer controls or a cyberattack.
WHO:Anthropic
WHAT:Anthropic circulated a draft bill proposing conditions for blocking an AI model's release.
WHEN:Reported Oct 20, 2026 at 7:01 PM ET
WHERE:USA
WHY:Anthropic's stated aim was to set out its vision for government regulation of the AI industry.
Reporting from 1 identified publisher for this event.
