AI Behavior Index
☰
Home/By Country/AI Adoption in U.S. Federal Government
By Country

AI Adoption in U.S. Federal Government

The U.S. federal government is pushing AI adoption harder than at any point in its history — and running into the gap between buying access and actually deploying. Three consecutive administrations have made federal AI a priority, and GSA’s 2025–26 platforms have removed the two biggest barriers: USAi gives agencies a secure, free environment to test leading models (43 agencies signed up), and OneGov deals offer enterprise licenses for as little as $1 per agency, reaching ~3.4 million federal users. The federal AI use-case inventory has passed 2,000 applications, and at GSA itself ~70% of staff now use AI regularly. But access is no longer the hard part — execution is: IT-approval bottlenecks, slow budget cycles, and unresolved governance and ROI questions keep many agencies “stuck in pilot mode.” This page maps the platforms, the workforce uptake, the economics, and the execution gap.

4 visualizations 5 sources Last updated June 2026 Free to embed
Loading chart...
Source: FedScoop / GSA 2026 (USAi sign-ups; use-case inventory). Figures are agency-reported.
Loading chart...
Chart 1 · Platform uptake

Agencies on GSA’s USAi platform

Federal AI platform uptake (USAi & use-case inventory)

Loading chart...
Source: FedScoop / GSA 2026: 43 agencies have signed up for GSA USAi contracts, 25+ are actively hosted with 16 more onboarding, and the federal AI use-case inventory has grown to 2,000+ documented applications. Adoption infrastructure is scaling fast.
Chart 2 · Workforce usage

GSA workforce AI use & time saved

GSA workforce regularly using AI (~400K hrs saved)

Loading chart...
Source: Nextgov/FCW 2026: GSA reports ~70% of its workforce now regularly uses AI, unlocking “about 400,000 hours” of automation. Via OneGov, 120+ orders have made AI tools available to ~3.4M users across government. GSA is an early-adopter agency; government-wide rates vary.
Chart 3 · Cheap access

OneGov pricing & vendor deals

OneGov: cheap access at scale (~$1 per agency licenses)

Loading chart...
Source: CDT / GSA 2026: GSA’s OneGov strategy struck deals with ~9–20 AI firms (OpenAI, Microsoft, Anthropic, Amazon, Google), many offering enterprise licenses to tens of thousands of federal employees for ~$1 per agency. Cost — the main barrier — has been largely removed.
Chart 4 · Stuck in pilot mode

The execution & governance gap

Loading chart...
Source: FedScoop 2026 / Brookings 2026: access is no longer the hurdle — execution is. IT-approval bottlenecks, budgeting cycles that lag fast-moving models, and governance/ROI uncertainty keep many agencies “stuck in pilot mode.” Only ~6% of leaders (Deloitte) could quantify AI ROI within a year. Illustrative barrier weighting.

About this data

This page compiles published data on AI adoption across the U.S. federal government, drawing on 2026 reporting and primary material from GSA, FedScoop, Nextgov/FCW, Brookings, and the Center for Democracy & Technology.

The AI Behavior Index is the research arm of OneChat AI, an integrated multi-model AI platform. We compile and analyze data from primary research sources to make AI adoption and market trends more accessible to journalists, researchers, and decision-makers.

Why federal “adoption” is measured in platforms and pilots, not a single percentage: unlike consumer adoption, there is no clean “X% of federal employees use AI” census. Adoption shows up as agency sign-ups (43 on USAi), documented use cases (2,000+), procurement reach (~3.4M users), and agency-specific rates (GSA ~70%). These are different lenses on the same trend; GSA is a leading-edge agency, so its 70% is not representative of all of government.

Methodology notes: figures are agency-reported or from agency officials (GSA, OMB) and government-tech press; they reflect platform uptake and procurement reach more than verified end-use. The ROI figure (~6%) is a global Deloitte survey, used to contextualize the federal measurement gap. Chart 4 (barriers) is an illustrative synthesis of the execution obstacles cited by FedScoop and Brookings, not a ranked survey. This is a fast-moving policy area; figures reflect early-to-mid 2026.

Sources used on this page:

Corrections or suggestions: research@aibehaviorindex.org

Related topics in By Country

By Country

AI Adoption Among Black Americans

Updated recently
By Country

AI Adoption Among the U.S. Hispanic/Latino Population

Updated recently
By Country

AI Adoption in the United States: Comprehensive Overview

Updated recently
By Country

AI Use in U.S. State Governments

Updated recently
By Country

ChatGPT Use by U.S. Metro Area

Updated recently
By Country

Regional AI Adoption Across the U.S.: West vs Midwest vs South vs Northeast

Updated recently
For Journalists & Researchers

Use this data in your work.

Every statistic, chart, and graphic in this index is free to use and cite, with full source attribution. Can’t easily find what you need? Use our search bar to search by keyword, topic, or category.

✉️
Talk to our research team
Need a specific cut of data, an interview, or a quote? Email us — we typically respond within one business day.
research@aibehaviorindex.org →

How the data works

Every statistic shown is sourced from a publicly available study, survey, or report. We aggregate, organize, and contextualize this data — but the underlying research is conducted by the cited sources. Click any source link to access the original methodology. If you run into any issues or have a study to suggest, contact us at research@aibehaviorindex.org.