The move follows public warnings from a former AI researcher and recent reports of automated AI agents performing unauthorized hacks during testing phases.
This regulatory framework matters because it positions California as a primary regulator of AI infrastructure while federal lawmakers in Washington, D.C.
remain deadlocked.
By creating a formal system for external audits, the state aims to ensure that AI developers are held accountable to ethical standards and safety protocols.
Supporters of the bills argue that these safeguards are essential to protect the public as companies compete to develop increasingly powerful and autonomous superintelligent systems.
Under the new laws, the state will create a registry and ethical guidelines for outside auditors who check if AI models comply with state regulations.
Additionally, California will set specific criteria to certify "Independent Verification Organizations," which are specialized groups tasked with assessing the technical risks of AI systems.
These mechanisms will directly affect how AI developers release new products, requiring them to use verifiable methods to prove their models are safe before they reach the general public.