Exposed deepfake database shows how users manipulated celeb pics
2025-03-31
The Daily Dot
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
Security researcher Jeremiah Fowler exposed an unsecured database from South Korea’s GenNomis, revealing over 90,000 AI-generated explicit images, including deepfakes of celebrities and depictions of minors. The incident highlights severe privacy breaches and the potential weaponization of AI in generating nonconsensual harmful content.[AI generated]
Why's our monitor labelling this an incident or hazard?
The event describes an AI system (an AI 'nudify' app) generating explicit deepfake images, including non-consensual use of real individuals' faces and celebrities portrayed as children. The exposure of this database constitutes a direct violation of privacy and ethical norms, which falls under violations of human rights and breaches of applicable laws protecting personal rights. The AI system's use has directly led to realized harm, qualifying this as an AI Incident rather than a hazard or complementary information. The harm is clear and ongoing, involving serious privacy and legal concerns.[AI generated]