California orders strict AI safety for state contracts
Governor Newsom signed an executive order requiring safety and privacy guardrails for any AI firm contracting with the state — including mandatory bias audits for hiring tools, automated‑decision oversight, independent cybersecurity audits, and disclosure to applicants. The move positions California as the strictest U.S. regulator on AI and raises compliance requirements for vendors targeting public‑sector work. (theguardian.com) (nytimes.com)
Executive Order N‑5‑26 was issued and signed by Governor Gavin Newsom on March 30, 2026. (gov.ca.gov) The order tasks the California Government Operations Agency with developing a plan for new state contracting processes to vet AI vendors. (gov.ca.gov) It also directs the California Department of Technology to draft recommendations and best practices for watermarking AI‑generated images and manipulated video. (californianewswire.com) N‑5‑26 cites Governor Newsom’s prior Executive Order N‑12‑23 (Sept. 6, 2023) and notes California is home to 33 of the top 50 AI companies worldwide. (gov.ca.gov) The order requires the Department of General Services and the Department of Technology to submit recommendations for AI‑related vendor certifications within 120 days. (news.yahoo.com) The state will perform its own supply‑chain risk assessments even when a firm is labeled a federal supply‑chain risk, a response referenced against recent Pentagon action involving Anthropic. (news.yahoo.com) The executive action arrives while the White House has pushed a national AI framework (Dec. 11, 2025 executive order) and released a March 20, 2026 AI policy blueprint for Congress, creating an explicit state–federal policy contrast. ( )
