Federal government’s use of AI soars, but bottlenecks threaten momentum amid public skepticism: Brookings

The use of artificial intelligence across the U.S. federal government has expanded dramatically in recent years, but significant obstacles, from talent shortages to public skepticism, are slowing the responsible integration of artificial intelligence technology into government services, the agency’s new report finds. brookings institute.

Wednesday’s report is based on an inventory of AI use cases from 2023 to 2025, federal employment data, an Office of Management and Budget memo, and interviews with current and former federal engineers across eight government agencies.

The numbers speak of rapid acceleration. In 2025, 41 government agencies documented more than 3,600 individual AI use cases. This is 69% more than the total reported in 2024 and five times the number reported in 2023. Applications span a wide range of government functions. More than half of the use cases reported by the Social Security Administration support service delivery and benefits processing, and more than half of the Department of Justice’s inventory support law enforcement activities.

However, the growth is far from even. Over the past three years, the five largest agencies accounted for more than half of all reported AI use cases, and in 2025, large agencies accounted for 76% of total inventory. Smaller agencies are barely keeping up. In 2025, the 11 small agencies reported a total of just 60 use cases, representing just 2% of their total inventory.

This report identifies several structural barriers to widespread adoption. One of the most pressing issues is the lack of specialized human resources. Of the more than 56,000 technology job postings posted by the federal government since 2016, just over 1,600 (less than 3%) explicitly mention AI capabilities.

The hiring surge during the Biden administration was aimed at addressing this gap, but the layoffs in early 2025 may have undermined that effort, as at least 25% of AI-specific job listings were posted after 2024. That means many of the newly hired workers could have easily been laid off in the recent past.

Beyond staffing, the report points to a deeply ingrained culture of risk aversion within federal agencies. Nearly 60% of all AI use cases are in the pilot or pre-deployment stage, suggesting that the federal government’s AI landscape remains in a rapid growth phase. Dedicated time to education and experimentation is required, and many agencies struggle to find that time. The report also shows that the Trump administration has made a clear link between AI adoption and the workforce. Government Efficiency Bureau (DOGE) There is a possibility that this hesitancy is increasing.

The accountability gap is another concern. Despite explicit requirements from OMB, over 85% of high-impact AI use cases deployed in 2025 lack some necessary information regarding risk mitigation.

Public trust poses yet another challenge. According to recent data from the Pew Research Center, nearly half of Americans now say they are more concerned than excited about the growing rise of AI, up from 37% four years ago, and only 17% of Americans believe AI will have a positive impact on the United States over the next 20 years.

The report warns that the risks are high. Public trust in the federal government remains near historic lows, with recent data showing just 16% say they trust Washington to always do the right thing. Against this backdrop, the authors argue that while improperly executed AI deployments can cause serious harm, well-designed applications that focus on specific service improvements can conversely help restore trust in government agencies.

To get there, Brookings recommends expanding AI literacy training across government agencies, reforming procurement rules designed for more static software systems, increasing transparency practices around risky AI uses, and prioritizing use cases that have clear, positive benefits for the public.