GAO Data Strengthens Calls for Agencies to Measure Federal AI Outcomes

Federal agencies are expanding their use of artificial intelligence while facing questions about whether those investments are producing better public services. A Federal News Network commentary, “Washington’s AI experiment is flying blind,” recommends that agencies publish a service-level scorecard for AI spending. The proposed scorecard would measure outcomes alongside workforce capacity, service performance, and quality. It is an analysis and recommendation—not an announced government policy, formal requirement, or procurement opportunity—so contractors and agency teams should not treat it as an existing mandate.

The commentary’s recommendation comes amid rapid growth in reported federal AI activity. The Government Accountability Office (GAO) data cited in the source show that federal AI use cases increased from 571 in 2023 to 1,110 in 2024. At the same time, 22 major agencies had about 250,000 fewer employees in January 2026 than in December 2024. Together, these figures sharpen a practical acquisition question: whether AI systems measurably improve service delivery and quality, and whether agencies have the people and skills needed to select, implement, oversee, and sustain them.

For procurement professionals, a service-level measurement approach could inform how agencies define requirements, evaluate proposals, and assess contractor performance. Agencies may consider asking offerors to describe expected service outcomes, baseline measures, error rates, workforce and skills needs, and methods for capturing lessons learned. Such measures can help connect technical claims to mission results, while clarifying who is responsible for monitoring system performance after deployment. These are potential uses of the commentary’s recommendation, not requirements established by the source.

GAO has recommended policy updates at the Department of Defense (DoD), Department of Homeland Security (DHS), General Services Administration (GSA), and Department of Veterans Affairs (VA). The source does not specify the content or implementation status of those updates, but their identification makes AI acquisition practices and agency readiness relevant areas for those departments and their suppliers to monitor. Contractors supporting federal AI programs should be prepared to substantiate performance claims with evidence and to explain how their solutions fit agency workforce capacity, operational needs, and service-quality objectives.

• Federal AI use cases rose from 571 in 2023 to 1,110 in 2024, according to GAO data cited in the source.
• Twenty-two major agencies had about 250,000 fewer employees in January 2026 than in December 2024, underscoring workforce-readiness considerations for AI adoption.
• Federal News Network commentary recommends publishing a service-level scorecard for AI spending, measuring outcomes, workforce capacity, service performance, and quality.
• The scorecard is a recommendation, not an announced government policy, agency mandate, solicitation, or procurement action.
• Contractors should be ready to document measurable service outcomes, error rates, workforce and skills needs, and lessons learned from AI acquisitions.
• Procurement teams can use outcome and quality measures to assess AI proposals and monitor performance, while clearly distinguishing proposed practices from current requirements.
• GAO-recommended policy updates at DoD, DHS, GSA, and VA make those agencies relevant for monitoring AI acquisition oversight and readiness.
• The source does not provide the details or status of the recommended agency policy updates; vendors should verify any requirements in official agency materials and solicitations.