Establishes a centralized AI program within the Department of Energy to advance AI research and deployment for science, security, and technology. Requires development of AI models, datasets, and infrastructure. Mandates AI risk evaluation, workforce development, and partnerships. Allocates $2.4 billion annually for five years.
Analysis summaries, actor details, and coverage mappings were LLM-classified and may contain errors.
This is a binding legislative act from the United States Congress with mandatory language, specific appropriations, enforcement mechanisms, and legal obligations on the Department of Energy and other federal agencies.
The document has good coverage of approximately 6-8 subdomains, with strong focus on AI system security (2.2), malicious actors and weapons development (4.2), competitive dynamics (6.4), governance structures (6.5), AI safety failures and goal misalignment (7.1), dangerous capabilities (7.2), and lack of robustness (7.3). Coverage is concentrated in security, national security risks, and AI safety domains relevant to Department of Energy missions.
This Act primarily governs AI use within the Department of Energy's operations, which spans Scientific Research and Development Services, Information (computing infrastructure), and Public Administration. It also has regulatory implications for the energy sector through permitting improvements and grid interconnection standards, affecting Trade, Transportation and Utilities.
The document comprehensively covers all AI lifecycle stages from planning through operational monitoring. It emphasizes the complete development pipeline including data collection, model building, validation, deployment, and ongoing monitoring, with particular focus on safe deployment and continuous evaluation of AI systems for science, energy, and national security applications.
The document explicitly defines and covers AI models, AI systems, foundation models, and frontier AI with specific technical definitions. It establishes compute thresholds for frontier AI (models with more than 1 trillion parameters). The document does not explicitly mention general purpose AI, task-specific AI, generative AI, predictive AI, or open-weight models, though these may be implicitly covered under broader AI categories.
United States Congress
This is an Act of Congress, as indicated by the legislative format and structure. Congress is the proposing authority for this federal legislation.
Department of Energy; Secretary of Energy; Federal Energy Regulatory Commission; Congress
The Secretary of Energy has primary enforcement authority for implementing the programs. The Federal Energy Regulatory Commission has rulemaking authority. Congress maintains oversight through required annual reporting.
Congress; Department of Energy; Office of Critical and Emerging Technology
Congress monitors implementation through mandatory annual reports. The Department of Energy monitors through the newly established Office of Critical and Emerging Technology and through coordination mechanisms across Centers.
Department of Energy; National Laboratories; National Nuclear Security Administration (NNSA); Federal Energy Regulatory Commission; private companies; nonprofit organizations; institutions of higher education
The Act primarily targets the Department of Energy and its National Laboratories to develop AI capabilities. It also applies to private entities through partnerships and to other federal agencies through coordination requirements.
11 subdomains (7 Good, 4 Minimal)