“A bill to provide for certain requirements relating to cloud, data infrastructure, and foundation model procurement.”
No CRS summary available for this bill.
This section establishes definitions for terms used in Department of Defense (DoD) artificial intelligence (AI) procurement, including "covered provider" (i.e., cloud, data infrastructure, or foundation model providers with DoD contracts totaling at least $50 million in any of the prior five fiscal years) and "foundation model" (i.e., an AI model with at least 1 billion parameters using self-supervision and applicable across contexts, or modifiable for high-risk tasks). The section requires the Secretary of Defense, in contracts with cloud, data infrastructure, and foundation model providers, to (1) require competitive award processes promoting security, resiliency, and competition; (2) ensure exclusive government rights to access and use government data; and (3) prioritize government roles in intellectual property and data rights, modular open systems, mitigation of barriers for small businesses and nontraditional contractors, and multi-cloud technology unless infeasible or a national security risk. The section further directs the DoD Chief Digital and Artificial Intelligence Office to update the Defense Federal Acquisition Regulation Supplement to prohibit unauthorized use or disclosure of government-furnished data (including for training commercial products), require data protection per DoD principles, impose penalties such as fines and contract termination for violations, and allow national security exemptions with notification. Finally, the section requires the Chairman of the Joint Chiefs of Staff, in coordination with the Under Secretary of Defense for Acquisition and Sustainment, to submit reports to the congressional defense committees by January 15, 2027, and annually thereafter for four years assessing AI market competition, innovation, barriers to entry, and market power, listing exemptions, and providing legislative and administrative recommendations.