"AI Likeness Protection Act"; concerns distributing realistic representation of individual's image, likeness, or voice created using generative artificial intelligence.
A5088, the “AI Likeness Protection Act,” would create a new civil cause of action for the unauthorized publication, distribution, transmission, or public availability of a highly realistic digital representation of a person’s image, likeness, or voice created with generative artificial intelligence. The bill defines both “generative artificial intelligence system” and “realistic representation,” and it focuses on synthetic media that a reasonable person would believe is the real individual, including altered audio, images, video, or transmissions. It also excludes certain ordinary editing and remixing activities, as well as AI-generated content that is clearly labeled or watermarked as such.
Before making such content public, a person or entity would need prior authorization or consent from the person depicted, or from a parent or legal guardian if the person is a minor. If consent is not obtained, the affected individual or minor’s parent/guardian could sue for injunctive relief, damages, profits attributable to the unauthorized use, attorneys’ fees, or a civil penalty of up to $1,000 per violation. The bill also provides special treatment for minors who violate the law, limiting the remedy for a first violation to injunctive relief, with full remedies available for repeat violations.
The bill would supplement Title 56 of the New Jersey Statutes and create a new state-level framework aimed at regulating deepfakes and other AI-generated likenesses. It would affect creators, publishers, distributors, platforms, and others who share synthetic media, while also carving out defenses for labeled AI content, content that would obviously be understood as AI-generated, and satire or parody involving public figures. The act would take effect eight months after enactment.
The overall sentiment reflected in the bill text is strongly protective of individuals’ identity, privacy, and emotional well-being, with the Legislature emphasizing the deceptive realism and potential harms of generative AI. Because there are no committee transcripts or recorded votes provided, there is no documented public debate in the supplied materials; however, the structure of the bill suggests a policy approach focused on preventing misuse rather than broadly restricting all AI-generated content. Likely points of contention would include free speech concerns, the scope of the consent requirement, the parody/satire exception, and whether the bill could be difficult to enforce against rapidly evolving AI tools and online distribution channels.
The bill would add a new statutory cause of action in New Jersey for unauthorized AI-generated realistic depictions of a person, effectively extending state law into the area commonly known as deepfakes or synthetic media. It would give individuals, and parents or guardians of minors, the right to seek injunctions and monetary remedies against those who distribute realistic AI likenesses without consent, while also creating defenses and exclusions for clearly labeled AI content, obvious AI use, and parody or satire of public figures. In practical terms, it would impose compliance obligations on content creators, publishers, and distributors who use generative AI to depict real people.
The bill’s tone and findings show strong concern about the harms posed by generative AI, especially deception, reputational injury, emotional harm, and misuse of a person’s identity. The available materials do not include committee testimony or vote results, so there is no recorded legislative opposition or support to summarize from the provided context. Based on the text alone, the bill appears to be framed as a consumer- and privacy-protection measure with a generally protective, pro-regulation sentiment.
No committee transcript or voting record was provided, so no specific lawmakers, witnesses, or organized groups can be identified as having raised objections. Based on the bill’s design, the most likely areas of contention are whether the consent requirement is too broad, whether the definition of “realistic representation” could sweep in legitimate creative or editorial uses, and whether the parody/satire and labeling defenses are sufficient to protect First Amendment interests. Enforcement against online platforms and the practical difficulty of distinguishing harmful deepfakes from protected or clearly disclosed AI content are also likely points of debate.