HR 8283 United States House · 119th Congress

Deterring American AI Model Theft Act of 2026

This bill establishes a framework to protect American-owned closed-source AI models from unauthorized extraction by foreign entities, particularly those from China and Russia. It requires the Secretary of State to conduct assessments identifying which foreign entities are conducting model extraction attacks or facilitating them through fraudulent account networks, then publish a public list of these actors for up to five years. The legislation authorizes the President to impose economic sanctions on identified entities and their affiliates, while also creating mechanisms for industry coordination and sharing information about threats. Importantly, the bill distinguishes between legitimate AI research conducted under contractual terms and unauthorized extraction attempts that bypass access controls or violate usage agreements.
Bill status in committee 1 of 4 stages cleared
Introduction
Apr 2026
Committee Review
Floor Vote
President
Introduced Apr 15, 2026 Last action Apr 22, 2026
Floor votes

How they voted

No floor votes recorded yet.
Full legislative history

Actions timeline

Total actions
4
Key actions
1
Committee
2
Amendments
1
Apr 22, 2026
Introduced
Ordered to be Reported (Amended) by the Yeas and Nays: 43 - 0.
lower
Apr 22, 2026
Lower · Passed
Committee Consideration and Mark-up Session Held
lower
Apr 15, 2026
Committee
Referred to the House Committee on Foreign Affairs.
lower
Apr 15, 2026
Introduced
Introduced in House
lower
1 primary · 21 co-sponsors

Sponsors