CAL.Regulation
California requires developers and users of clinical AI tools to identify and mitigate biased impacts
Governor Gavin Newsom signed SB 503 on September 30, 2026. From January 1, 2027, developers of AI clinical decision support systems used in health programs must identify systems with known or foreseeable risk of biased impacts, disclose intended uses and risks to deployers, and make reasonable efforts to mitigate that risk. Deployers must regularly monitor the systems and take reasonable, proportionate steps to mitigate biased impacts.
Enrolled text (California Legislative Information) → · Office of Governor Gavin Newsom
Key points
- SB 503 adds Chapter 25.3, commencing with Business and Professions Code section 22758, effective January 1, 2027.California Legislative Information
- It covers AI systems whose predictions, classifications or recommendations aid diagnosis or treatment decisions, known as clinical decision support systems.California Legislative Information
- Developers must make reasonable efforts to identify systems with known or reasonably foreseeable risk of biased impacts in health programs or activities.California Legislative Information
- Developers must give deployers a statement of intended uses and known or foreseeable risks, plus specified documentation, and make reasonable efforts to mitigate bias.California Legislative Information
- Deployers must regularly monitor clinical decision support systems and take reasonable and proportionate steps to mitigate known or foreseeable biased impacts.California Legislative Information
- Newsom signed SB 503 on September 30, 2026, as part of a package of AI bills.Office of Governor Gavin Newsom
Sources2 articles across 2 outlets
California Legislative Information
Office of Governor Gavin Newsom
Filed under
RegulationCal.California SB 503 AI in health careCivil rightsComplianceHealthcareTechnologyFor practitionersFor governanceFor vendors