California Governor Gavin Newsom signed a package of four laws requiring independent safety assessments and auditor standards for artificial intelligence developers in California. The measures establish protections for minors using companion chatbots and mandate the screening of synthetic nucleic acids to prevent biological threats. These requirements are the first state-level implementation of their kind in the U.S. and utilize independent verification organizations to assess system risks.
OpenAI endorsed the state legislation and urged the U.S. Congress to establish mandatory, capability-based federal safety rules for labs building the most advanced AI. The company proposes requirements for independent testing, stronger cybersecurity, and incident reporting, though it suggests excluding startups and open-weight models. OpenAI also warned against autonomous recursive self-improvement and stated it would halt model development if safety risks became unacceptable.
Key sources
- SUPPORT@kalshi“Newsom signs "first-in-the-nation" AI audit laws in California and calls for immediate federal action”x.com
- SOURCE@politico“Obama calls for Dems to focus on AI”x.com
- SOURCE@thehill“Jeffries says Congress should ‘act urgently’ on AI safeguards”x.com
- SUPPORT@thehill“Obama pushes Jeffries to move AI oversight to center of agenda”x.com
- SUPPORT@clashreport“2028 presidential candidates should make AI a central part of their agendas”x.com
- SOURCEhuggingnewshuggingnews.com