Abstract

Amidst the general excitement about the opportunities afforded by artificial intelligence (AI), the tech industry must confront the uncomfortable reality that generative AI also facilitates child sexual exploitation and abuse (CSEA). This issue remains under-addressed in the literature. Aiming to deepen the understanding of online CSEA and the misuse of generative AI, we report empirical insights from semi-structured interviews with seven UK law enforcement practitioners. The topics covered ranged from AI facilitating grooming and lowering barriers in image generation to its impacts on society and investigative policing. Interviewees highlighted significant inadequacies and gaps in current safeguards associated with multi-turn AI tools, including interaction-centred misuse patterns and a “testing gap” between existing AI safety evaluation methods and the real-world practices of offenders. We use insights from these interviews to expand the scope of AI safety evaluation methods and propose changes to regulatory guidance. Our contribution foregrounds perspectives of experts on technology-facilitated CSEA crimes and specifies actionable directions for responsible AI and public safety.