Sponsor
Gabe Evans
Representative · R-CO
Representative · R-CO
The bill promotes clearer rules, nationwide access, and public transparency for open AI models—supporting innovation and targeted risk management—but does so while raising costs, regulatory complexity, potential privacy/IP and national-security exposure, and limits on regulators' ability to swiftly remove dangerous models.
Tech developers, startups, and small businesses gain clearer federal guidance, definitions, and interstate protections that make publishing, distributing, and commercializing open AI models easier across U.S. states and jurisdictions.
All Americans and critical infrastructure operators get greater transparency because the Secretary must publish annual public assessments of foreign-adversary models, enabling better risk management, procurement decisions, and situational awareness.
State governments and national-security planners get clearer definitions of 'foreign adversary' and 'covered nation', enabling more targeted identification and mitigation of adversary-sourced AI risks.
Financial institutions, utilities, critical infrastructure operators, and the public face increased national-security risk because the bill limits the Secretary's ability to remove or restrict harmful open models and public assessments could expose sensitive vulnerabilities.
Small businesses, startups, state agencies, and taxpayers will incur new compliance, monitoring, and reporting costs as agencies benchmark models and enforce guidance, and federal coordination could shift resources toward particular approaches or firms.
Researchers, tech workers, and foreign-affiliated firms risk chilling cross-border collaboration, privacy and IP harms, and extra compliance burdens because models can be labeled 'foreign adversary' or promoted through cross-border agreements.
Based on analysis of 5 sections of legislative text.
Directs Commerce to promote qualified open AI models, assess and publicly report risks from foreign-adversary AI models, and issue guidance and monitoring tools while not authorizing bans on open models.
Official title: To direct the Secretary of Commerce to support the adoption and use of American open artificial intelligence models, and for other purposes.
Introduced August 27, 2026 by Gabe Evans · Last progress August 27, 2026
Requires the Secretary of Commerce to promote and support adoption of qualified open-source AI models, while assessing and publicly reporting risks tied to AI models controlled by foreign adversaries. It directs the Department to identify barriers, coordinate programs, partner with private entities and qualified foreign partners, and develop best practices and monitoring approaches for open models. Requires an 18-month initial public risk assessment and then annual reports (sunsetting reporting authority after 10 years) about foreign adversary models and comparisons to qualified open models, while clarifying the Act does not give the Secretary authority to ban or restrict open AI models in commerce. The Act also defines key terms such as qualified open model, foreign adversary model, qualified foreign partner, and U.S. person status.