Stitt says the decision stems from four key concerns: security risks, regulatory compliance issues, susceptibility to adversarial manipulation, lack of robust security safeguards.

A Bay Area assemblymember is helping lead California’s efforts to create a framework for regulating artificial intelligence.