Skip to main content

CA SB 942

California AI Transparency Act (CAITA)

Requires large GenAI providers (1M+ monthly users) to provide free AI detection tools, embed latent disclosures (watermarks/metadata) in AI-generated content, and offer optional manifest (visible) disclosures to users.

Jurisdiction

California

US-CA

Enacted

Sep 19, 2024

Effective

Aug 2, 2026

Enforcement

California AG, city attorneys, or county counsel

Signed September 19, 2024; effective January 1, 2026

Who Must Comply

This law applies to:

  • Covered providers (GenAI systems with >1 million monthly users publicly accessible in California)

Applicability thresholds:

1M US-CA users/month

Subject to AI detection tool and latent disclosure requirements

Who bears obligations:

Exemptions

Entertainment Content

high confidence

Video game, television, streaming, movie, or interactive experiences (non-user-generated)

Conditions:

  • • Entertainment content
  • • Not user-generated

Text-Only Outputs

high confidence

Text-only outputs excluded from disclosure requirements

Conditions:

  • • Text-only output
  • • No image/video/audio

Safety Provisions

  • Free publicly accessible AI detection tool
  • Latent disclosures (imperceptible metadata/watermarks) with system info, provider name, creation date
  • User option for manifest disclosures (visible labels/watermarks)
  • Third-party licensees must maintain disclosure capabilities
  • License cancellation within 96 hours if licensee disables disclosures
  • User feedback collection on detection tool efficacy

Compliance Timeline

Jan 1, 2026

  • All provisions take effect
  • Free AI detection tool must be available
  • Latent disclosure (watermark/metadata) requirements begin

Enforcement

Enforced by

California AG, city attorneys, or county counsel

Penalties

$5K/violation

Per violation: $5,000

Civil penalty up to $5,000 per violation

Quick Facts

Binding
Yes
Mental Health Focus
No
Child Safety Focus
No
Algorithmic Scope
Yes

Why It Matters

Addresses deepfake/synthetic media concerns. Only applies to very large providers (1M+ threshold). Detection tool requirement is novel approach.

Recent Developments

Signed September 2024. Applies to large providers only (1M+ users). Excludes text-only outputs. Excludes video games, streaming, movies.

What You Need to Comply

Must provide free AI detection tool. Must embed latent disclosures in AI-generated image/video/audio. Must offer manifest disclosure option to users.

NOPE can help

Cite This

APA

California. (2024). California AI Transparency Act (CAITA). Retrieved from https://nope.net/regs/us-ca-sb942

BibTeX

@misc{us_ca_sb942,
  title = {California AI Transparency Act (CAITA)},
  author = {California},
  year = {2024},
  url = {https://nope.net/regs/us-ca-sb942}
}

Related Regulations

In Effect US-CA AI Safety

CA CPPA ADMT

California Privacy Protection Agency regulations establishing consumer rights and business obligations for Automated Decision-Making Technology (ADMT) that makes significant decisions including healthcare. Requires pre-use notice, opt-out rights, access rights, appeal rights, and risk assessments.

In Effect US-CA AI Safety

CA AB 2013

Requires GenAI developers to publish documentation about training datasets including sources, data types, copyright status, personal information inclusion, and processing methods.

Enacted US-NY AI Safety

NY RAISE Act

Requires large AI developers of frontier models operating in New York to create safety protocols, report critical incidents within 72 hours, conduct annual reviews, and undergo independent audits. Creates dedicated DFS office funded by developer fees.

Enacted US-AR Child Protection

AR HB 1071

Amends Arkansas publicity rights law to explicitly include AI-generated reproductions of voice and likeness. Covers simulated voices and 3D generation.

In Effect US-NY Online Safety

NY S 7676-B

Protects performers from exploitative digital replica contracts. Contracts for AI-generated digital replicas are void unless they describe use, performer has legal counsel or union representation, and contract doesn't replace work performer would have done.

In Effect US-TN Online Safety

TN ELVIS Act

Protects individuals from unauthorized AI-generated use of their name, photograph, voice, or likeness. Explicitly covers AI-generated voice simulations. Criminal and civil penalties including treble damages for knowing violations.