According to the California legislature, audio recordings, video recordings, and still images can be compelling evidence of the truth. However, the proliferation of Artificial Intelligence (AI), specifically, generative AI, has made it drastically easier to create fake content that is almost impossible to distinguish from authentic content. To address this concern, California's Governor signed Senate Bill (SB) 942, which requires businesses that provide generative AI systems to make accessible tools to detect whether content was created by AI.

SB 942 defines "covered provider" as "a person [or business] that creates, codes, or otherwise produces generative artificial intelligence systems[, and] that has over 1,000,000 monthly visitors or users and is publicly accessible within the geographic boundaries of the state." Under SB 942, a covered provider must offer a publicly accessible AI detection tool at no cost. This tool allows users to assess whether the content was created or altered by AI and provides system provenance data (i.e., information explaining where the data originated) without revealing personal information.

Moreover, AI-generated content must include clear and conspicuous disclosures identifying it as such. Latent disclosures must also convey information about the content's origin and authenticity, detectable by the AI detection tool.

While this law will not end the challenges employers face trying to discern deepfakes from reality, it might help to avoid some critical missteps. Recall the disruption experienced by a school community, in particular its high school principal, in Pikesville, Maryland, when a recording suggested the principal made racially insensitive and antisemitic remarks. It took several months for the Baltimore County Police Department to investigate and conclude that the recording was a fake, a "deepfake," generated by AI technology. The increased transparency that SB 942 could bring may have reduced or eliminated the flood of calls to the school, heightened security, and employment actions taken against the principal.

Violations of the act can result in civil penalties of $5,000 per violation, enforceable by the Attorney General, city attorneys, or county counsels. This means that certain California businesses that provide generative AI services should create a plan for implementing an AI detection tool that allows consumers to distinguish between AI versus human-created content.

Fortunately, technologies exist and are being developed to help organizations address these transparency issues. For example, the Coalition for Content Provenance and Authenticity (C2PA) "addresses the prevalence of misleading information online through the development of technical standards for certifying the source and history (or provenance) of media content." C2PA may be used to embed metadata into AI-generated content to help verify its source and other information.

The requirements of SB 942 take effect January 1, 2026.