California Governor Gavin Newsom has announced a strategic partnership with Anthropic, under which state government agencies and local authorities will gain access to the AI assistant Claude at a 50% discount. This is a first-of-its-kind move that, in my view, sets a new trend for integrating large language models into public administration.

The agreement provides not only preferential access to the technology but also free employee training, technical support, and the involvement of Anthropic developers in configuring workflows. Newsom emphasized that AI should not replace human labor in government but rather help employees work faster and solve problems more efficiently.

Claude will become the first AI tool available to all state agencies through the SITeS shared IT services portal of the California Department of Technology. Several government agencies are already using Anthropic's product for customer service, internal processes, and cybersecurity tasks, including code scanning and remediation.

This agreement continues Newsom's push to integrate AI into the public sector, which began in 2023 with a directive to study the use of generative artificial intelligence. In 2025, an executive order was signed to improve government efficiency, and in March 2026, AI procurement rules were updated to account for civil rights, privacy, and employee risks.

Interestingly, this partnership comes amid Anthropic's recent release of two new models in the Claude family — Fable 5 and Mythos 5. Fable 5 is positioned as a safe, Mythos-class solution for general use, while Mythos 5 is a "private" base model with relaxed restrictions. However, shortly after launch, developers restricted access to these tools for foreign nationals due to a directive from the U.S. government, and at the end of the month, Anthropic accused operators linked to Alibaba of distilling Claude.

Expert opinion: California is betting on AI as a tool to improve public sector efficiency, but events surrounding restricted access to Claude for foreign nationals show that geopolitical risks and security concerns remain key challenges. The partnership with Anthropic is not just a technological step but also a signal to the market: government entities are ready to invest in AI, but with strict control conditions.