Ai Launches Across The Government

Federal employees will soon be able to try out popular artificial intelligence tools — part of a controversial Trump administration effort to automate more of the government.

The U.S. General Services Administration said Thursday morning it will launch USAi, a secure platform designed to let employees experiment with popular AI models made by OpenAI, Anthropic, Google and Meta.

The website is framed as a way to advance the Trump administration’s broader AI Action Plan and to deliver “a competitive advantage to the American people,” GSA Deputy Administrator Stephen Ehikian said in a statement.

The initiative comes at a time when AI threatens to displace federal workers whose jobs are already under siege by the Trump administration’s Department of Government Efficiency. The emerging technology is also developing at a clip that the government — known for its slow bureaucratic processes and uneven tech adoption rates — has historically struggled to keep pace with.

FAS Commissioner Josh Gruenbaum said the goal is to let employees experiment responsibly and “make day-to-day workflows more efficient.”

To address security concerns about government data, the platform is built on cloud infrastructure managed by the GSA, ensuring that agency data does not feed back in to train the companies’ models—a key distinction from commercially available AI tools.

The initiative comes on the heels of GSA announcing last week that it had added Anthropic’s Claude, Google’s Gemini and OpenAI’s ChatGPT to its Multiple Award Schedule, which establishes a pre-negotiated price baseline for the AI models and acts as a seal of approval for federal buyers looking to adopt the technology.

Both OpenAI and Anthropic quickly announced they would sell their AI to the government for $1 over the next year. Though a good deal for Washington, it also gives the two multi-billion dollar companies a first-mover advantage that could entrench their models within the government, possibly at the cost of smaller competitors and new entrants.

The administration says its new AI program is voluntary, not mandatory, and that more AI models are likely to be added later. Agencies opt in by signing a simple agreement, after which employees can immediately access chat, search and coding features.