Appropriated Likeness

Mitigation strategy

1. Implement robust technical safeguards, such as enhanced output filtering and adversarial training, to prevent Generative AI models from producing or disseminating unauthorized and realistic depictions of human likenesses (deepfakes, voice clones), aligning with a 'safety-by-design' paradigm. 2. Mandate and deploy digital provenance tracking mechanisms, such as cryptographic content credentials (e.g., C2PA) and imperceptible watermarking, on all Generative AI outputs to ensure the traceability and rapid identification of the source of appropriated likenesses. 3. Utilize and advocate for the strengthening of legal frameworks, including statutory and common law rights of publicity and appropriation torts, in conjunction with formal legal measures (e.g., cease and desist orders and civil litigation) to deter the non-consensual exploitation and commercial use of an individual's digital persona.