Citizen Portal
Sign In

Get Full Government Meeting Transcripts, Videos, & Alerts Forever!

Get email alerts on the Ai Copyright Transparency topic

No spam. Unsubscribe anytime.

Creators and unions back AB 412; committee advances bill after debate on feasibility

Senate Committee on Privacy, Digital Technologies, and Consumer Protection · June 8, 2026
AI-Generated Content: All content on this page was generated by AI to highlight key points from the meeting. For complete details and context, we recommend watching the full video. so we can fix them.

Summary

The Senate privacy committee advanced AB 412, the AI Copyright Transparency Act, after testimony from artists, unions and technologists and objections from industry groups about technical feasibility and legal uncertainty. The bill was passed as amended to the Judiciary Committee.

Assemblymember Bauer Kehan introduced AB 412, the AI Copyright Transparency Act, telling the Senate Committee on Privacy, Digital Technologies and Consumer Protection the bill is a transparency measure that would allow rights holders to ask whether their copyrighted works were used to train generative AI models and does not itself change copyright law or require compensation.

Supporters, including Tim Friedlander, president of the National Association of Voice Actors, and representatives from SAG‑AFTRA, IATSE and the Concept Art Association, told the committee the bill would protect creators’ economic value and biometric voice data. "Every day without transparency, that value diminishes," Friedlander said, arguing the measure would help artists learn whether their work has been used in training and reduce costly litigation.

Technical witness Ben Zhao, Neubauer Professor of Computer Science at the University of Chicago, told the committee fingerprinting and duplicate‑detection techniques exist across modalities and that research and prototype tools show the query mechanism the bill proposes is feasible. Zhao said the field has long studied these methods and that implementations exist for images, sound and text.

Industry groups opposing the bill urged the committee to reject it or proceed cautiously. Adam Isgrohe of the Chamber of Progress said federal courts have not settled whether model training is infringement and warned the bill relies on approaches that produce unavoidable false positives and false negatives and face scaling and ownership‑verification challenges. "The bill gets ahead of both the law and the technology," he said, urging a no vote.

Trudy Aribilone, founder of music app Beatmatch, told lawmakers the measure could impose heavy compliance burdens on startups required to cross‑reference training inputs, identify rights holders and notify them within tight timeframes. She said small teams lack the resources to meet what she described as sweeping notification obligations.

Committee members pressed technical and legal questions. Senator McNerney distinguished asking a model where content originated from versus requiring a model author to disclose the dataset composition; witnesses and the author said AB 412 focuses on rights holders’ ability to query whether their material was ingested. The author said she will accept committee amendments narrowing the bill to a query mechanism rather than a broad cataloging requirement.

After discussion, the committee moved to pass AB 412 as amended to the Judiciary Committee. The transcript records the committee action as "passed as amended to judiciary" with a recorded tally reported in committee proceedings and the bill placed on call.