House Bill 375 would create a new Chapter 170, the “Artificial Intelligence and Synthetic Media Act,” to regulate the use of AI-generated or AI-altered media in North Carolina. The bill focuses first on election-related content: within 90 days before an election, it would prohibit sponsoring or distributing deceptive deepfakes of candidates or political parties unless the content carries specified, conspicuous disclosures. It also requires embedded digital content provenance for certain online political audio or visual ads, and it extends disclosure obligations to other political communications that use “materially deceptive media,” with exceptions for satire, parody, bona fide news reporting, and certain platform or publisher activities.
The bill also creates new criminal and civil rules for non-political uses of synthetic media. It makes it a felony to create or possess generated child pornography, and it criminalizes the nonconsensual disclosure of fabricated intimate images, with enhanced penalties for repeat offenses. In addition, it provides a civil cause of action for victims of fabricated intimate image disclosure, including damages, attorneys’ fees, and injunctive relief. The bill further requires transparency disclosures when generative AI is used in regulated occupations, and it states that criminal liability still applies when an offense is committed with AI assistance.
In terms of state law impact, the bill would add a broad new statutory framework governing AI, deepfakes, synthetic audio and video, and digitally fabricated intimate imagery. It would create new definitions, new disclosure duties, new enforcement procedures, and new criminal penalties, while also giving courts expedited authority to issue injunctions in election-related disputes. It would affect candidates, campaigns, political advertisers, publishers, online platforms, regulated professionals, law enforcement, prosecutors, and individuals who create or share synthetic media.
The general sentiment reflected by the bill’s structure is strongly protective and regulatory: it is aimed at preventing voter deception, protecting minors, and limiting harms from AI-generated sexual imagery and impersonation. Although no committee transcripts or votes were provided, the bill’s title, findings, and enforcement provisions suggest a public-safety and election-integrity rationale rather than a permissive approach to AI use. The bill also includes explicit First Amendment and statutory carveouts, indicating an effort to balance regulation with protected speech and news reporting.
The main points of contention likely concern free speech, press protections, platform liability, and the breadth of the new disclosure and criminal provisions. The bill reaches political speech, news media, and online communications, but it contains exceptions for satire, parody, and bona fide journalism, which suggests lawmakers anticipated constitutional concerns. Another likely issue is whether the definitions of “deceptive,” “materially deceptive,” and “synthetic media” are broad enough to capture harmful conduct without chilling legitimate AI use, especially for publishers, interactive computer services, and regulated professionals.
The bill would amend North Carolina law by creating a new Chapter 170 governing artificial intelligence and synthetic media. It would impose election-season disclosure requirements for AI-generated political ads, establish criminal penalties for deceptive deepfakes, generated child pornography, and nonconsensual fabricated intimate images, and create civil remedies for victims. It would also require transparency disclosures for AI use in regulated occupations and clarify that AI-assisted conduct does not avoid criminal liability under existing statutes.
No committee debate or recorded votes were provided, so there is no direct evidence of legislative support or opposition in the materials. Based on the bill text, the overall posture is strongly protective of voters, minors, and privacy, with a clear intent to curb harmful AI misuse. The inclusion of exceptions for satire, parody, news reporting, and First Amendment-protected speech suggests the sponsor anticipated concerns about overbreadth and constitutional limits.
Likely areas of contention include whether the bill’s definitions of deepfake and materially deceptive media are too broad, whether the election-related disclosure rules burden political speech, and whether the criminal penalties are too severe for AI-related conduct. Media organizations, online platforms, and technology providers may also object to compliance burdens or uncertainty about liability, even though the bill includes some exemptions. Civil liberties advocates may focus on the potential chilling effect on satire, parody, journalism, and other protected expression, while supporters would emphasize election integrity, child protection, and privacy harms.