The bipartisan bill, introduced July 23 by Rep. Ted Lieu (D-Calif.) and Rep. Nathaniel Moran (R-Texas), would require developers of the most powerful AI systems to maintain the technical capability to throttle, suspend, or fully shut down their models. It would also authorize the Department of Homeland Security, in consultation with the Commerce Department and the Director of National Intelligence, to order a slowdown or shutdown of any AI system capable of causing catastrophic harm, following a graduated response framework matching the severity of an incident.
Companies would be required to report qualifying incidents to DHS and preserve forensic records, including model weights and telemetry data. Noncompliance with a shutdown order would carry penalties of up to $20 million per day, while failing to maintain the required kill-switch capability itself would carry penalties of up to $2 million per day. "We are moving from AI that answers questions to AI that takes actions," Lieu said in a statement, adding that powerful systems can "go rogue, behave in extremely dangerous ways, or even resist human intervention." Moran said the bill reflects "achievable policy" around a core question of stewardship.
The bill's release cited recent developments including a Commerce Department export-control action in June that required Anthropic to suspend some access to its Fable 5 and Mythos 5 models, and unspecified concerns about unexpected AI behavior during testing at another lab. Polling cited in the release found 86% of voters, across party lines, support requiring guaranteed shutdown capability for powerful AI systems. The bill has no companion in the Senate yet and Congress remains without a comprehensive federal AI framework.
