On August 1, California begins implementing the nation’s strongest AI transparency law. The law requires digital watermarks to be embedded in AI-generated content, making it easier to trace its origin. State Senator Josh Becker, a Democrat from Menlo Park, announced this on Thursday.
The law addresses a growing problem: AI-generated content that is becoming nearly indistinguishable from real content. It targets misleading content, including deepfakes and election misinformation. These issues cost Americans over $1 billion a year, Becker said in his statement.
Under the law, AI platforms will need to add traceable metadata to content they generate. This includes images, videos, and audio. The data will help users identify whether the content was AI-made. Becker called this a "new age of transparency" for the internet.
Anyone who consumes AI-generated content, even outside California, will benefit from this new visibility. Becker is also working on SB 1000, a follow-up bill to keep the law updated as AI evolves.
Becker said this law could set the standard for global AI transparency. He's already in talks with the European Union on aligning it with EU regulations, including the Code of Practice and the EU AI Act's Article 50.
The law does not ban AI, but it gives users tools to verify the authenticity of content they see online. It's aimed at reducing the spread of scams and deceptive content, especially in the run-up to elections.