California Judicial Council — Generative AI Use Policy
Annotated summary of the Judicial Council of California's statewide generative AI policy for courts — the first such mandate in the United States.
In 2024 the Judicial Council of California adopted the country's first statewide framework directing trial and appellate courts to either adopt their own generative AI policy or follow the model the Council promulgated. This page summarizes the policy's structure and its lessons for Ohio practitioners.
What the policy requires
- Every California court must adopt a written generative AI policy — either local or the Council's model — by the deadline specified in the rule.
- Confidential and sealed information may not be entered into any public-facing generative AI tool that retains or trains on inputs.
- Any work product produced with the assistance of generative AI must be reviewed by a human for accuracy, bias, and consistency with court rules before issuance or filing.
- Courts must address bias monitoring, accessibility, and public records implications in their local policies.
What it does not require
The Council declined to mandate a uniform attorney disclosure rule. Disclosure requirements were left to local court rules and individual judges. The model policy also stops short of approving or banning specific products — vendor selection remains a local governance decision.
Why it matters for Ohio
California is the template other large state systems are studying. Ohio's task force reviewed the Council's framework before issuing its own guidance. Two California decisions appear to have influenced Ohio's approach: (1) leaving disclosure to local discretion rather than a statewide bright line, and (2) treating procurement and monitoring as governance functions distinct from IT.
Practitioner takeaways
- If you practice in California or before a California tribunal, read the local court's posted AI policy before filing AI-assisted work.
- Where no local policy exists, the Council's model is the default expectation.
- The verification standard — human review for accuracy and bias before issuance — is consistent with ABA Formal Opinion 512 and the Ohio task force report.
The official policy and any updates are available at the source link above.
Attribution: This page reproduces public-record material from the Judicial Council of California. Reproduced and annotated by LegalTek.ai for educational purposes. The original document remains the work of the issuing authority.
Not legal advice: The content on this page is provided for informational and educational purposes only and does not constitute legal advice. No attorney-client relationship is created by reading this material. Consult a licensed attorney in your jurisdiction for advice on any specific matter.
LegalTek.ai is a technology company, not a law firm.