OpenAI just locked in a major deal to push ChatGPT Enterprise into federal agencies for only $1 per agency for the next year. This move undercuts rivals like Anthropic and Google, who also got greenlit by the U.S. General Services Administration (GSA) to provide AI tools to government workers.
The deal came one day after GSA added OpenAI, Google, and Anthropic to its approved AI vendor list for civilian agencies. These companies will now sell through the federal Multiple Award Schedule (MAS) platform, which simplifies contracts and avoids one-off negotiations.
OpenAI is sweetening the deal with unlimited advanced model access for 60 days, plus a new user community and custom training for federal employees. 
Security is front and center. The government worries about data leaks into training sets. GSA hasn’t revealed if there’s on-prem or private cloud setups yet. TechCrunch asked for details.
GSA commissioner Josh Gruenbaum pushed other American AI firms to match OpenAI’s pricing.
“The government is taking a cautious, security‑first approach to AI,” a GSA spokesperson told TechCrunch.
“This ensures sensitive information remains protected while enabling agencies to benefit from AI‑driven efficiencies.”
The deal hits amid the Trump administration’s AI Action Plan, which targets more AI in government and bans what it calls “woke AI” or non-ideologically neutral AI models from contracts.
TechCrunch reached out to OpenAI for comments on how the company will respond to the new executive order and forthcoming government AI policies. Anthropic and Google have yet to detail their pricing or approaches.
