“To direct the Director of the National Security Agency to develop strategies to secure artificial intelligence related technologies.”
No CRS summary available for this bill.
This section directs the Director of the National Security Agency, acting through the Artificial Intelligence Security Center, to develop an "AI Security Playbook" to defend covered AI technologies—defined as advanced AI systems (whether private, governmental, or public-private) with critical capabilities (e.g., matching or exceeding human expert performance in chemical, biological, radiological, and nuclear matters; cyber offense; model autonomy; persuasion; research and development; or self-improvement) that pose a grave national security threat if stolen by threat actors—from technology theft. The Playbook must include (1) identification of vulnerabilities in advanced AI data centers and developers, focusing on unique cybersecurity risks; (2) components or information (e.g., AI models, training insights, engineering data) that would aid threat actors' development of covered AI if accessed; (3) strategies to detect, prevent, and respond to cyber threats; (4) security levels, if any, requiring substantial U.S. government involvement for highly advanced AI; and (5) analysis of such government involvement, including a hypothetical initiative to build covered AI in a highly secure environment addressing cybersecurity, model weights protection, insider threats, access controls, counterintelligence, and contingency plans. The Playbook must contain detailed methodologies and intelligence assessments (potentially in a classified annex) and an unclassified portion with general guidelines for private sector dissemination. In developing it, the Director must engage prominent AI developers and researchers through document reviews, interviews, roundtables, facility visits, and collaboration with a federally funded research and development center; such activities are exempt from the Federal Advisory Committee Act. The Director must submit an initial report on progress to the House and Senate Intelligence Committees within 90 days of enactment and a final report within 270 days, including unclassified and publicly available versions (with a possible classified annex). A rule of construction clarifies that the provision authorizes no regulatory or enforcement actions.