California’s New AI Laws

September 19, 2023

Privacy Plus+

Privacy, Technology and Perspective

This week, let’s highlight five new laws enacted in California that regulate the use of artificial intelligence (AI):

AB 2655: Defending Democracy from Deepfake Deception Act of 2024 – Requiring large online platforms, as defined, to block the posting of materially deceptive content related to elections in California, and label certain additional content inauthentic, fake, or false during certain periods before and after an election in California. A link to the new law follows:

https://digitaldemocracy.calmatters.org/bills/ca_202320240ab2655

AB 2839: Elections: Deceptive Media in Advertisements – Prohibiting the knowing distribution of an advertisement or other election communication, as defined, that contains certain materially deceptive content, as defined, with malice, as defined, subject to specified exemptions during certain periods before and after an election in California. A link to the new law follows:

https://digitaldemocracy.calmatters.org/bills/ca_202320240ab2839

AB 2355: Political Reform Act of 1974: Political Advertisements: Artificial Intelligence – Requiring that political advertisements generated or substantially altered by artificial intelligence include a specified disclosure about AI, and outlining formatting requirements for this disclosure, and enforcement of compliance by the Fair Political Practices Commission. A link to the new law follows:

https://digitaldemocracy.calmatters.org/bills/ca_202320240ab2355

AB 2602: Contracts Against Public Policy: Personal or Professional Services: Digital Replicas – Providing that that any agreement for personal or professional services, including provisions regarding the use of a digital replica of an individual's voice or likeness, created on or after January 1, 2025, is unenforceable under certain conditions.  A link to the new law follows:

https://digitaldemocracy.calmatters.org/bills/ca_202320240ab2602

AB 1836: Use of Likeness: Digital Replica – Establishing liability for $10,000 or actual damages for individuals who produce, distribute, or use a deceased personality's digital replica in audiovisual works or sound recordings without prior consent, defining a digital replica as a highly realistic electronic representation of the individual’s voice or likeness. A link to the new law follows:

https://digitaldemocracy.calmatters.org/bills/ca_202320240ab1836

You can read the California Governor’s Press Release regarding these new laws by clicking on the following link:

https://www.gov.ca.gov/2024/09/17/governor-newsom-signs-bills-to-combat-deepfake-election-content/

Our Thoughts

The Impact of Deepfakes on Trust and Society: Deepfakes cross the line between reality and fiction—truth and lies. They complicate our relationships to each other and our institutions by creating a layer of uncertainty and mistrust from communications to creations to everyday interactions. When people cannot easily distinguish what is genuine from what is fabricated, they lose trust in information and each other.

Concerns in Election Campaigns and Entertainment: Given these challenges, we believe the concerns regarding the use of AI, and particularly deep fakes in election campaigns and the entertainment industry are well-founded.  Further, we believe proactive measures, like the enactment of these new laws in California, are essential to safeguarding our democratic processes and protecting individuals from harm.

Recommended Reading: For a fascinating exploration on how deepfakes can influence elections by misleading voters, often in unexpected ways, we would recommend reading the following article from the May edition of Wired Magazine, entitled, “Indian Voters are Being Bombarded with Millions of Deepfakes. Political Candidates Approve,” a link to which follows:

https://www.wired.com/story/indian-elections-ai-deepfakes/

First Amendment Considerations: As these new laws are introduced, we anticipate significant challenges on First Amendment grounds. We would urge that the First Amendment is not absolute (shouting “Fire!” in a crowded theatre is not protected speech), and election integrity should be considered as a prevailing national security issue.  Additionally, the rights of privacy and publicity have long been protected and must evolve to address technological advances and new risks posed by them.

The Debate on Fair Use: We also question whether there might be a “fair use” application for AI-generated avatars and voices. Some history museums, for example, use these for education. However, there is a sharp difference between the use of AI-generated content for education and fraud.

--- 

Hosch & Morris, PLLC is a boutique law firm dedicated to data privacy and protection, cybersecurity, the Internet and technology. Open the Future℠.