California Signs Landmark Executive Order on AI, Defying Trump's Light-Touch Approach

California Governor Gavin Newsom signed a sweeping executive order on March 31, establishing the most aggressive state-level artificial intelligence regulations in the United States and setting the stage for a direct confrontation with the Trump administration over who gets to govern the technology industry. The order imposes new safety, privacy, and civil rights requirements on AI companies seeking to do business with the state — a move that carries outsized weight given that 33 of the world's 50 largest AI companies operate within California's borders.

The executive order directs the California Department of General Services and the Department of Technology to develop new certification requirements for AI vendors within 120 days. Companies bidding on state contracts will be required to certify that their systems include safeguards against illegal content, harmful algorithmic bias, and civil rights violations — including protections for free speech, voting rights, and against unlawful surveillance. The order also mandates watermarking guidelines for AI-generated images and video, and calls for the creation of a public-facing AI tool to help Californians access government services.

Perhaps the most provocative provision gives California the authority to override federal decisions on AI supply chain risks. The order instructs the state's Chief Information Security Officer to review any federal designation of AI companies as supply chain threats, effectively reserving the right to continue purchasing from vendors that the Trump administration might seek to block. It is a remarkable assertion of state autonomy that legal experts say could trigger significant constitutional questions about federal preemption.

"California has always been the cradle of innovation. But we also understand the other side: in the wrong hands, innovation can be used in ways that put people at risk," Newsom said in a statement. "While others in Washington are crafting policies and contracts under the shadow of misuse, we're focused on doing this the right way." The governor added that California would "use every tool we have to ensure companies protect people's rights, not exploit them."

The order arrives just weeks after the White House released a national AI framework in March 2026 that characterized artificial intelligence as "an inherently interstate phenomenon with key implications for foreign policy and national security," arguing for federal preemption of state laws deemed "burdensome." That framework, developed under pressure from major technology company lobbying, represents the Trump administration's preference for a single, industry-friendly regulatory regime — a vision now directly challenged by the nation's most economically powerful state.

The stakes of this regulatory standoff are difficult to overstate. California's economy, the fourth largest in the world, gives it enormous purchasing power and regulatory leverage. State procurement standards have historically influenced corporate behavior well beyond California's borders, much as the state's vehicle emissions rules have shaped the auto industry nationwide. AI companies that want access to California's vast government contracts will have little choice but to comply, potentially making the state's standards a de facto national baseline.

Governors and legislators in other states are already watching closely. With Congress still deeply divided on comprehensive AI legislation, California's executive order could serve as a template for other large states seeking to fill the federal regulatory vacuum. The order represents the most significant assertion of state authority over artificial intelligence to date, and it ensures that the debate over who regulates AI in America — and how — will intensify sharply in the months ahead.