Establishes an Artificial Intelligence Security Center within the NSA. Provides a research test bed, develops counter-AI guidance, promotes secure AI adoption, and coordinates with NIST. Ensures access to proprietary models for qualified researchers and sets usage terms.
Analysis summaries, actor details, and coverage mappings were LLM-classified and may contain errors.
This is a binding federal statute enacted by the United States Congress with mandatory obligations and enforcement authority vested in the Director of the National Security Agency.
The document has good coverage of approximately 6-8 subdomains, with strong focus on AI system security (2.2), malicious actors using AI for cyberattacks (4.2), competitive dynamics (6.4), governance structures (6.5), and AI system safety failures (7.2, 7.3). Coverage is concentrated in security, infrastructure, and governance domains.
The document primarily governs the National Security sector through establishment of the AI Security Center within NSA. It also has significant coverage of Scientific Research and Development Services through the research test bed provisions, and Information sector through coordination with NIST and engagement with commercial AI model vendors.
The document primarily covers the Build and Use Model, Verify and Validate, and Operate and Monitor stages. It establishes infrastructure for AI security research, model testing, and ongoing security guidance for deployed AI systems in national security contexts.
The document explicitly mentions AI systems and AI models, with focus on proprietary models from commercial vendors. It addresses security research infrastructure and counter-AI techniques but does not specify particular AI types (frontier, general purpose, task-specific) or compute thresholds. There is implicit coverage of both open and proprietary models through the voluntary provision mechanism.
United States Congress
The document is a section of the Intelligence Authorization Act for Fiscal Year 2025, which is federal legislation enacted by the United States Congress.
Director of the National Security Agency
The Director of the NSA is given explicit authority to establish the Center, set terms of usage, establish mechanisms for model access, and ensure compliance with the statutory requirements.
Director of the National Security Agency; Artificial Intelligence Safety Institute of the National Institute of Standards and Technology
The Director of the NSA has oversight responsibility for the AI Security Center's operations. The document also establishes coordination with NIST's AI Safety Institute, suggesting a monitoring and coordination role.
National Security Agency; Cybersecurity Collaboration Center; private sector researchers; academic researchers; model vendors; managers of national security systems; elements of the defense industrial base; Federal agencies
The document establishes obligations for the NSA Director to create the AI Security Center and provide services to researchers, model vendors, national security system managers, and other federal agencies. It also governs AI developers who provide proprietary models and researchers who access the test bed.
9 subdomains (5 Good, 4 Minimal)