SB 53 - Requires Artificial Intelligence Companies to Publish and Adhere to Safety Guidelines in Order to Mitigate Catastrophic Risks Associated with AI - California Key Vote

Stage Details

Title: Requires Artificial Intelligence Companies to Publish and Adhere to Safety Guidelines in Order to Mitigate Catastrophic Risks Associated with AI

See How Your Politicians Voted

Title: Requires Artificial Intelligence Companies to Publish and Adhere to Safety Guidelines in Order to Mitigate Catastrophic Risks Associated with AI

Vote Smart's Synopsis:

Vote to concur with House amendments and pass a bill that requires artificial intelligence companies to publish a framework describing how they will incorporate national, international, and industry standards into their AI framework in California.

Highlights:

  • Establishes the Transparency in Frontier Artificial Intelligence Act (TFAIA) (Sec. 1).

  • Requires a large frontier developer to write, implement, and clearly and conspicuously publish on its internet website a frontier AI framework that describes how it adheres to national and international standards and implements best practices (Sec. 1).

  • Requires a large frontier developer to transmit to the the Office of Emergency Services a summary of any assessment of catastrophic risk (Sec. 1).

  • Requires the Office of Emergency Services to establish a mechanism for a frontier developer or a member of the public to report a critical safety incident that includes (Sec. 1):

    • The date of the critical safety incident;

    • The reasons the incident qualifies as a critical safety incident;

    • A short and plain statement describing the critical safety incident; and

    • Whether the incident was associated with the internal use of a frontier model.

  • Requires the Office of Emergency Services to establish a mechanism for a large frontier developer to confidentially submit summaries of any assessments of the potential for catastrophic risk (Sec. 1).

  • Exempts the following reports of a critical safety incident submitted to the Office of Emergency Services from the California Public Records Act:

    • A report of assessments of catastrophic risk from internet use, and

    • a covered employee report made pursuant to the whistleblower protections.

  • Establishes a civil penalty for noncompliance with the TFAIA to be enforced by the Attorney General (Sec. 1).

  • Establishes within the Government Operations Agency a consortium required to develop a framework for the creation of a public cloud computing cluster to be known as “CalCompute” (Sec. 1).

  • Requires the Government Operations Agency to, on or before January 1, 2027, submit a report from the consortium to the Legislature (Sec. 1).

  • Prohibits a frontier developer from making, adopting, enforcing, or entering into a rule, regulation, policy, or contract that prevents a covered employee from disclosing, or retaliating against a covered employee for disclosing information to (Sec. 1):

    • The Attorney General;

    • A federal authority;

    • A person with authority over the covered employee; or

    • Another covered employee who has the authority to investigate, discover, or correct the reported issue.

  • Requires a large frontier developer to provide a certain internal process through which a covered employee may anonymously disclose information (Sec. 2).

  • Requires a frontier developer that discovers a critical safety incident that imposes an imminent risk of death or serious physical injury to disclose this information within 24 hours to an authority (Sec. 2).

See How Your Politicians Voted

Title: Requires Artificial Intelligence Companies to Publish and Adhere to Safety Guidelines in Order to Mitigate Catastrophic Risks Associated with AI

Vote Smart's Synopsis:

Vote to amend and pass a bill that requires artificial intelligence companies to publish a framework describing how they will incorporate national, international, and industry standards into their AI framework in California.

Highlights:

  • Establishes the Transparency in Frontier Artificial Intelligence Act (TFAIA) (Sec. 1).

  • Requires a large frontier developer to write, implement, and clearly and conspicuously publish on its internet website a frontier AI framework that describes how it adheres to national and international standards and implements best practices (Sec. 1).

  • Requires a large frontier developer to transmit to the the Office of Emergency Services a summary of any assessment of catastrophic risk (Sec. 1).

  • Requires the Office of Emergency Services to establish a mechanism for a frontier developer or a member of the public to report a critical safety incident that includes (Sec. 1):

    • The date of the critical safety incident;

    • The reasons the incident qualifies as a critical safety incident;

    • A short and plain statement describing the critical safety incident; and

    • Whether the incident was associated with the internal use of a frontier model.

  • Requires the Office of Emergency Services to establish a mechanism for a large frontier developer to confidentially submit summaries of any assessments of the potential for catastrophic risk (Sec. 1).

  • Exempts the following reports of a critical safety incident submitted to the Office of Emergency Services from the California Public Records Act:

    • A report of assessments of catastrophic risk from internet use, and

    • a covered employee report made pursuant to the whistleblower protections.

  • Establishes a civil penalty for noncompliance with the TFAIA to be enforced by the Attorney General (Sec. 1).

  • Establishes within the Government Operations Agency a consortium required to develop a framework for the creation of a public cloud computing cluster to be known as “CalCompute” (Sec. 1).

  • Requires the Government Operations Agency to, on or before January 1, 2027, submit a report from the consortium to the Legislature (Sec. 1).

  • Prohibits a frontier developer from making, adopting, enforcing, or entering into a rule, regulation, policy, or contract that prevents a covered employee from disclosing, or retaliating against a covered employee for disclosing information to (Sec. 1):

    • The Attorney General;

    • A federal authority;

    • A person with authority over the covered employee; or

    • Another covered employee who has the authority to investigate, discover, or correct the reported issue.

  • Requires a large frontier developer to provide a certain internal process through which a covered employee may anonymously disclose information (Sec. 2).

  • Requires a frontier developer that discovers a critical safety incident that imposes an imminent risk of death or serious physical injury to disclose this information within 24 hours to an authority (Sec. 2).

Title: Requires Artificial Intelligence Companies to Publish and Adhere to Safety Guidelines in Order to Mitigate Catastrophic Risks Associated with AI

Title: Requires Artificial Intelligence Companies to Publish and Adhere to Safety Guidelines in Order to Mitigate Catastrophic Risks Associated with AI

arrow_upward