GSA Inks Another $1 OneGov Vendor Deal, This Time With Anthropic

Anthropic has become the latest company to benefit from the US government's frenetic AI adoption pace, inking a deal to get its software into the hands of federal agencies at a deep discount.

The General Services Administration on Tuesday announced an agreement with Anthropic that will see the Claude maker open its Enterprise and Government variants to agencies across all three branches of government for just $1 for a year of access. Because this is yet another OneGov deal, agencies interested in trying out Claude products will be able to buy it without needing to establish their own individual purchasing agreement with Anthropic. 

"OneGov is revolutionizing how the federal government acquires AI technology," GSA federal acquisition service commissioner Josh Gruenbaum said in a statement. "This agreement with Anthropic is another major step in the AI-driven transformation of government." 

Anthropic will also provide support to government agencies to help them with onboarding as part of the deal, the GSA said. Claude's Government offering is certified for FedRAMP High, meaning that it meets "the most stringent requirement" for handling sensitive unclassified workloads, according to Anthropic. 

While we didn't receive confirmation from the GSA or Anthropic, this deal also appears to be the first time the government has structured one of several recent AI deals to open discounted AI to groups outside of the federal civilian executive agencies. According to the GSA, members of Congress and the federal judiciary will also be able to purchase Claude for $1, though that depends on the approval of both other branches, suggesting that they haven't had AI made available to them in a similar manner before. 

It's worth questioning whether giving advanced AI to the federal judiciary amid an epidemic of fake AI legal citations is a good idea, and that goes double when it's Claude. Anthropic's own lawyers were called out in court for using fabricated legal citations in a music copyright case, and they blamed Claude hallucinations for the mistake - not exactly a great look for a company whose AI is now available to help judges. 

The Anthropic deal is the latest in a series of GSA AI pushes of late, and comes just a week after the agency added Claude, Gemini and ChatGPT to the Multiple Award Schedule to streamline agency purchase of their products in line with the OneGov strategy. The same day, the GSA also announced a deal with OpenAI to make ChatGPT available to federal agencies for $1 per year, with support, in a manner nearly identical to Tuesday's Anthropic announcement. 

• LLMs can hoover up data from books, judge rules
• Reddit sues Anthropic for scraping content into the maw of its eternally ravenous AI
• Trump administration's whole-government AI plans leaked on GitHub
• Anthropic: All the major AI models will blackmail us if pushed hard enough

The GSA has also announced deals with AWS, which made $1 billion in AWS credits available to federal agencies through the end of 2028, and Oracle, which will offer license-based technology to federal agencies at a 75 percent discount.

As we've noted in previous stories about these deals, none of them is permanent. The AWS credits expire in a couple of years, Oracle discounts aren't locked in permanently, and those $1 AI deals with OpenAI and Anthropic appear to expire after a year. 

What that means for federal agencies that've grown accustomed to using certain premium services at a deep discount is that they'll essentially be locked in come the end of the trial period and forced to pay a premium for software that (hopefully, if you're a vendor) has become central to operations. We checked with GSA to see if it has plans to avoid lock-in issues, but didn't hear back. ®