Anthropic Pushes Draft AI Safety Bill to Congress, Including a Kill Switch for Extreme-Risk Models
Understand the story
On October 21, 2026, TechCrunch, citing Politico, reported that Anthropic is promoting a draft AI safety bill to Congress. The draft would allow the release of an AI model to be blocked when it poses a foreseeably imminent risk of causing more than 50 deaths or more than $1 billion in property damage, where the risk stems from biological weapons, the model evading developer control, or cyberattacks. Only this one report has been published so far; there has been no response from Congress or further information on the bill's progress.
AI overview of reporting · Updated 2 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- TechCrunch · AIAnthropic pitches AI safety bill that may block a model if 50+ people could be killed
According to Politico, Anthropic is pitching a draft AI safety bill to Congress under which a model's release could be blocked if it poses a foreseeable, imminent risk of killing more than 50 people or causing more than $1 billion in property damage, with the risk stemming from bioweapons, the model evading developer control, or cyberattacks.
Story heat trend
Not enough continuous observations to show a trend yet.