Leading on Common Sense AI Safety

You may have heard that Congress is debating President Trump’s “Big Beautiful Bill” (seriously, that is the official name of the bill), which recently passed out of the House of Representatives by only one vote. A lot of attention has rightly been paid to how this bill will kick millions of Americans off of Medicaid and add trillions of dollars to our national debt. Not good.

Less attention has gone to a smaller provision in the bill that would invalidate any and all state laws that have been passed to regulate the AI industry and place a 10-year moratorium on future state-level AI regulations. In fact, Georgia Rep. Marjorie Taylor Greene recently admitted that she had no idea this provision was in the bill, and now that she actually understands what she voted for, she regretted it. Better late than never, I suppose.

In the face of inaction by Congress, California has stepped up to protect our 40 million residents from potential harms from AI. I am proud to have led those efforts.

Six years ago, in 2019, I was the first legislator in the country to write a law to minimize the potential damage that could be caused by AI generated deepfake images, audio, or video depicting a candidate for public office doing or saying something the candidate did not do or say. AB 730 was the first law of its kind in the country to regulate election-related misinformation and disinformation created using AI, and I am proud that many other states have followed California's lead.

As AI technology has become both cheaper and easier to use, I came back last year to strengthen my original groundbreaking law to regulate abuse of AI in our elections by passing AB 2655, which puts more responsibility on social media platforms to block, remove, or label election related content that they know to be both deceptive and digitally modified.

In addition, I partnered with the California District Attorneys Association to author AB 1831, which provides prosecutors with the tools they need to stop child predators by making it illegal to create, possess, or distribute AI-generated child sexual abuse material (CSAM). It is hard to imagine, but millions of these sexually explicit images of children exist online and on the dark web, the vast majority of which are of young women and girls.

These new laws, and many others authored by my colleagues in the Legislature, have been critical in helping our state keep up with this emerging technology.

We must foster constructive uses of AI while implementing laws that would mitigate the harms from AI’s abuses. Trump’s “Big Beautiful Bill” goes in the opposite direction, invalidating these laws and the progress that we have made to protect 40 million Californians.

California is the proud home of technological innovation, but as technology advances, so too must our laws. I am committed to finding common-sense and effective solutions to continue protecting our communities.

Sincerely,

Marc Berman
Assemblymember, 23rd District