GAO Releases Report on Government Agencies’ Implementation of AI
Many federal agencies have begun using artificial intelligence but have not fully implemented safeguards.
Published by The Lawfare Institute
in Cooperation With
On Dec. 12, the U.S Government Accountability Office (GAO) publicly released a report on the implementation of artificial intelligence (AI) in the federal government (the report excluded the Defense Department as the GAO has released two reports exclusively on AI at the Pentagon). The report found that while several federal agencies have begun using AI, many legal requirements and guidance have not been met.
According to the report, 20 of 23 agencies examined have begun implementing or plan to implement AI in the near future. NASA and the Department of Commerce accounted for a majority of the total use cases of AI—of which there were about 1,200—in fiscal year 2022. Only five of the 20 agencies using AI provided the Office of Management and Budget (OMB) with all the information required by the Chief Information Officers Council’s guidance for creating agency AI inventories, which are databases of an agencies’ uses of AI. About half of the agencies have implemented the AI requirements created specifically for their agencies.
Based on its findings, GAO made 35 recommendations in the report, most notably that all agencies fully comply with AI reporting and implementation requirements. GAO also recommends that OMB, the Office of Science and Technology Policy, and the Office of Personnel Management issue more government-wide guidance and establish an occupational series with AI-related positions.
You can read the report here or below:
