CALIFORNIA LEGISLATION

California AI Transparency Act (AB 853, 2026)

Updated May 15, 2026

Status: Enacted - Signed into law as part of California's 2025 AI legislative package. Effective January 1, 2026.

Bill Number: California AB 853 (California AI Transparency Act)

Jurisdiction: California

Category: AI Content Provenance, Generative-AI Labeling, Platform Disclosure

Industries: Generative AI Developers, Large Online Platforms, Content Distribution, Media

Effective Date: January 1, 2026

Summary: California AB 853 imposes labeling, provenance, disclosure, and risk-governance requirements on generative AI developers and large online platforms that distribute AI-generated content. It is the consumer-facing companion to SB 53 in California's 2025 AI legislative package. The statute requires that AI-generated images, video, and audio carry provenance metadata or visible disclosures, with platform-level obligations to preserve and surface that provenance to end users. The law closes long-standing gaps in California's prior AI deepfake statutes by extending provenance obligations to the upstream model developer rather than only the downstream publisher.

Key Provisions:

  • Provenance Metadata: Generative AI developers must attach durable provenance metadata to outputs (images, video, audio).
  • Platform Preservation: Large online platforms must preserve provenance metadata and provide consumer-facing disclosure mechanisms.
  • Visible Disclosure Option: Developers may pair metadata with visible labels for high-risk content categories.
  • Enforcement: Attorney General authority with civil penalties; coordinated with California Privacy Protection Agency.

Affected Entities: Developers of generative AI systems made available to California users; large online platforms hosting or distributing AI-generated content; ad networks; news and social media platforms.

Compliance Outlook: Platforms operating in California should expect rapid integration of C2PA or equivalent content credentials in 2026, with rulemaking guidance from the California Privacy Protection Agency. Enterprise communications and marketing teams should audit any AI-generated visual or audio assets distributed through California-facing channels.

IntelliSee Relevance: Adjacent. AB 853 does not regulate physical security AI directly, but it is part of the broader California AI accountability stack that customers in education, healthcare, and government will reference when evaluating any AI-driven system. IntelliSee tracks the cumulative regulatory burden in its State-by-State AI Security Legislation Tracker.

Related Legislation & Resources

Stay Ahead of Compliance

Turn Legislative Mandates Into a Working Safety Program

IntelliSee tracks the safety legislation that affects your facilities and maps each mandate to practical AI threat detection on the cameras you already own.

See All Tracked Legislation Talk to IntelliSee About Compliance