Answers are grounded in this paper and cite their sources.
Pornography has long driven technological innovation and now appears to be a major force in the AI landscape. New tools are freely accessible, highly sophisticated, and require no technical expertise, enabling anyone, including minors, to instantly create realistic sexualised imagery. The rise of sexualised deepfakes presents serious legal challenges, particularly when real persons are depicted without consent. While parallels can be drawn with non-consensual intimate imagery (i.e. ‘revenge pornography’), the involvement of new actors and technologies raises novel regulatory questions. Structured around the key actors involved in the creation and dissemination of deepfake pornography, this paper identifies an ‘under-regulation’ of the phenomenon in the EU both as regards the deployment of AI technology and user prompting. A possible avenue pointing towards ways of mitigating risks and addressing regulatory gaps would be the classification of deepfake pornography as ‘high-risk’ under the AI Act.
No takes yet. Share an insight, caveat, or question.
Schmitz-Berndt et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: