What happened
The General Services Administration published a revised internal AI governance directive, CIO 2185.1C, 'Accelerating Responsible Use of Artificial Intelligence at GSA,' in June 2026 (administrative edit published June 2026, superseding CIO 2185.1B). The directive aligns GSA AI governance with OMB Memorandums M-25-21, M-25-22, and M-26-04. It establishes: (1) standards for assessment, procurement, usage, monitoring, and governance of AI systems at GSA; (2) empowers the Chief AI Officer (CAIO) and establishes an AI Governance Board; (3) requires regular AI risk assessments, especially for high-impact AI systems; (4) mandates AI use-case tracking and performance reporting; (5) requires AI-ready data governance and workforce training under the AI Training Act of 2023; (6) instructs GSA to prioritise American-made AI solutions and support a competitive US AI marketplace.
Why it matters
As the federal government's central procurement and real-estate agency, GSA's internal AI governance standards set a practical benchmark for how federal AI is acquired and deployed. The directive's alignment with OMB M-26-04 and its emphasis on Chief AI Officers and AI Governance Boards signals the governance architecture the federal government is building. Vendors selling AI to GSA must understand the risk-management and lifecycle-accountability expectations their agency customer now operates under.
Action needed
AI vendors with GSA contracts or pursuing GSA business should review the directive's risk-assessment, lifecycle-accountability, and US-sourcing expectations to align their offer and delivery practices.