New Research Shows AI-Generated Child Sexual Abuse Material Rising Sharply
Guardii Analysis
The Internet Watch Foundation reported on 21 August 2024 that analysts confirmed a sharp increase in AI-generated child sexual abuse imagery during 2023, with a single dark-web forum alone hosting over 20,000 AI-created images assessed as criminal under UK law. The report noted that generative-AI tools are lowering the technical barrier to producing realistic abuse imagery, with offenders using text-to-image models and face-swapping software to create and share material that meets the legal threshold for child sexual abuse material without requiring direct contact with a child victim.
AI-generated abuse material is legally indistinct from photographed content and carries the same potential for harm through distribution, normalisation and use in grooming. Guardii's anti-CSAM detection does not depend on image provenance: the system identifies sharing, solicitation and possession of child sexual abuse material—whether photographed, rendered or synthetically generated—within direct-message conversations across platforms in real time, alerting safeguarding teams and law enforcement to criminal contact as it occurs and ensuring that detection keeps pace with the tools offenders adopt.