“A bill to provide for limitations on the use of artificial intelligence by Department of Defense.”
No CRS summary available for this bill.
This section expresses the sense of Congress that the Department of Defense (DoD) must aggressively adopt artificial intelligence (AI) in the Armed Forces to maintain military preeminence while ensuring security and reliability, consistent with America's AI Action Plan of July 2025. The section prohibits DoD from using AI (1) to execute the launch or detonation of a nuclear weapon; (2) to monitor, track, profile, or target individuals or groups in the United States without an individualized, articulable legal basis—and never solely for activities protected by the First Amendment or other constitutional or statutory rights; or (3) to employ lethal force by autonomous weapon systems without appropriate human judgment and supervision (with all other uses of AI in such systems required to comply with DoD Directive 3000.09). The Secretary of Defense may waive the prohibition in (b)(3) for up to one year (renewable) for a specific system upon written certification to the congressional defense committees of extraordinary national security circumstances and that the system's error rate does not exceed that of trained human operators under equivalent conditions. For each waiver, the Secretary must notify Congress within five days of issuance for development or fielding, or of any substantial system modification, including the rationale, system description, operational parameters and safeguards (e.g., performance assessments, training/doctrine, activation/deactivation procedures, testing results), anticipated duration, and (in unclassified form with a possible classified annex).