California Gov. Gavin Newsom signed an executive order on March 30 placing new limits on how artificial intelligence can be used in state government, while also encouraging agencies to expand use of the technology.

      The order directs state agencies to create safeguards aimed at preventing harm from AI systems, including violations of civil rights, unlawful surveillance and the creation of harmful content. At the same time, it calls on agencies to adopt vetted AI tools to improve efficiency and public services.

      “California’s always been the birthplace of innovation. But we also understand the flip side: in the wrong hands, innovation can be misused in ways that put people at risk,” Gov. Newsom said in a statement. 

      “California leads in AI, and we’re going to use every tool we have to ensure companies protect people’s rights, not exploit them or put them in harm’s way,” the governor added.  

      The move reflects California’s effort to balance innovation with oversight as the state remains a global hub for AI development. The order requires agencies to recommend contract standards that address risks such as discrimination, misuse of personal data and the potential for AI to generate illegal content.

      Agencies must also update the state’s digital strategy to identify ways generative AI can improve transparency, accountability and access to government services. The directive includes plans to expand AI tools for public use and to guide employees on labeling AI-generated images and videos with watermarks.

      The order follows a dispute between the U.S. Department of Defense and San Francisco-based AI company Anthropic, which was labeled a supply chain risk after disagreements over contract terms. A judge has since issued a temporary injunction blocking that designation. Under the order, California will review such federal decisions before determining whether to contract with affected companies.

      Newsom’s office criticized federal AI policy under President Donald Trump, saying it does not go far enough to address risks.

      “Unlike the Trump administration, California remains committed to ensuring that AI solutions adopted and deployed by (California)… cannot be misused by bad actors,” the governor’s office said in a statement.

      This is Newsom’s second executive order on AI, following a 2023 directive focused on generative technology. The policy comes as state agencies expand AI use across services and operations.