California Signs Statewide Anthropic Deal Amid Federal Tensions

California expands AI access with Anthropic partnership

California has finalized an agreement with Anthropic to make Claude AI tools available to all state agencies and local governments at a 50 percent discount, marking the first such statewide AI deployment in the nation. The deal includes free workforce training and technical support from Anthropic staff.

According to details first reported by POLITICO, the agreement was negotiated as Anthropic faces ongoing conflict with the Trump administration, which has restricted deployment of the company's most advanced models and designated it a supply chain risk to national security.

Chris Given, California's chief information officer and director of the state Department of Technology, said the discount aims to consolidate AI usage under favorable terms. "When we see that folks are going to be using a tool more, we want to make sure that we, as the state, have negotiated the best possible price for them," Given told POLITICO.

Why it matters

This deal represents a significant test of state-federal tensions over AI regulation and procurement. California's willingness to expand its relationship with Anthropic despite Pentagon concerns signals that states may chart independent courses on AI adoption, potentially fragmenting national technology policy. For enterprise leaders, the arrangement also demonstrates how volume commitments can unlock substantial discounts from AI vendors.

Federal designation not a barrier

The Pentagon formally labeled Anthropic a supply chain risk on March 5, the same day California held its first meeting with the company about this deal. Despite that timing, Given said the federal designation "was not an issue that we faced while signing and negotiating this contract."

In March, Gov. Gavin Newsom signed an executive order directing the California Department of Technology to review federal supply chain designations and issue guidance allowing the state to continue business with companies it deems improperly labeled. Given said his department is finalizing recommendations on that review process, expected next month.

Newsom's administration maintains the Anthropic deal was not intended as a rebuke to Washington, though the March executive order explicitly aimed to separate California's AI procurement process from federal oversight.

Existing deployments across state services

Claude already powers several California government functions. The AI model underpins Poppy, the state's digital assistant that serves as an entry point for employees to access AI tools. It also supports customer service operations at the California DMV and internal workflows assisting Medicaid recipients at the Department of Health Care Services.

The state Department of Technology has established a cybersecurity partnership with Anthropic, using Claude to identify vulnerabilities in state code and critical systems. Additionally, Claude powers a public engagement platform Newsom recently used to gather resident feedback on AI's impact on employment.

Balancing innovation and worker protection

Newsom has issued multiple executive orders promoting generative AI for government efficiency while attempting to address workforce concerns. Last month, he directed state agencies to explore policies for responding to AI-driven layoffs, and last week launched a tool to track AI-related job losses.

The governor has faced criticism from labor leaders who warn they will not support a potential presidential bid unless he enacts stronger worker protections against AI displacement.

"AI should not replace the human work of government," Newsom said in a statement. "It should help our workers move faster, solve problems more effectively, and deliver better results for Californians."

Given indicated California plans to pursue similar discount agreements with other AI companies and technology providers. The state invited AI model providers to negotiate terms in December.

Details of the agreement were first reported by POLITICO.