HR-7058-119
Ordered to be Reported by the Yeas and Nays: 45 - 0.
Sponsored by Michael Baumgartner (R-WA)
What it does
This bill would require the Secretary of State, in consultation with other relevant federal agencies, to submit an assessment to Congress within 180 days of enactment — and annually for three years after — on the national security risks posed by foreign adversaries using generative AI for malicious activities. Each assessment would cover AI-enabled disinformation, weapons development assistance, cyberattacks, and military or surveillance enhancements. The unclassified portion of each report would be posted publicly on the State Department's website, though a classified annex may be included to protect intelligence sources and methods.
Who benefits
Congressional foreign affairs committees (House Foreign Affairs, Senate Foreign Relations), which would receive structured, recurring intelligence on an emerging threat area. The general public, which would gain access to unclassified findings. U.S. allies and partners, who are explicitly included in the scope of the threat analysis. Policymakers and researchers working on AI governance and national security. Diplomatic and intelligence professionals who would gain a formal mandate and interagency coordination mechanism for this issue area.
Who is hurt
No group faces direct material harm from this bill. Federal agencies — particularly the State Department and relevant intelligence community components — would bear administrative costs and staff time to produce the assessments. Classified annex requirements could create bureaucratic friction across agencies. Foreign adversary governments (defined by reference to 10 U.S.C. §4872(f)) may face increased diplomatic scrutiny as a result of the findings.
Supporters argue
Supporters argue that generative AI is rapidly expanding the toolkit available to foreign adversaries for disinformation, cyberattacks, and weapons development, and that the U.S. government currently lacks a systematic, public-facing assessment of these risks. They contend that mandating regular, interagency-coordinated reports — with both classified and public components — would improve congressional oversight, inform diplomatic strategy, and help establish international norms around responsible AI use, all at minimal cost. The bill's unanimous 45-0 committee vote suggests broad bipartisan recognition of the threat's seriousness.
Opponents argue
Opponents argue that the bill creates a reporting mandate without any accompanying policy authority, funding, or enforcement mechanism, making it a largely symbolic exercise that adds bureaucratic workload without producing actionable change. They contend that existing intelligence community products — such as the Annual Threat Assessment published by the Director of National Intelligence — already cover foreign adversary AI threats, and that duplicating this function through the State Department risks redundancy, interagency confusion, and the diversion of diplomatic resources from higher-priority engagements.
Constitutional context
The bill falls squarely within Congress's established authority to require executive branch reporting under the Foreign Commerce Clause (Art. I, §8, cl. 3) and its general oversight powers. The bill does not direct diplomatic recognition, command military forces, or implement a treaty, so no significant tension arises with the President's exclusive recognition power under Zivotofsky v. Kerry (2015) or the Commander-in-Chief Clause. The "sense of Congress" provision in Section 2 is advisory and carries no binding legal force.
Checks and balances
Congress gains structured, recurring oversight information on a national security topic; the executive branch (Secretary of State) gains a formal interagency coordination role, but the bill imposes no new policy authority or enforcement power on either branch.
Historical precedent
Congress has previously mandated recurring executive branch threat assessments in specific domains, such as the annual country reports on human rights practices required under the Foreign Assistance Act of 1961, though no directly analogous AI-specific foreign adversary reporting mandate has previously been enacted.