AI-Generated Child Abuse Material Reports Surge 8,400% in 18 Months

Synthetic abuse imagery overwhelms reporting systems

Reports of artificial intelligence-generated child sexual abuse material have increased by more than 8,400% in just 18 months, creating an unprecedented challenge for law enforcement and child protection organizations.

The National Center for Missing and Exploited Children documented approximately 4,700 reports of AI-generated CSAM in 2023. That number exploded to more than 400,000 reports in the first six months of 2025 alone, according to data first reported by InvestigateTV.

Cybercriminals are extracting innocent photographs and videos of children from social media platforms and using generative AI tools to create abusive imagery. The material ranges from manipulated real photos to fully synthetic creations that never involved an actual child.

Why it matters

Even fully synthetic AI-generated abuse material causes measurable harm to real children and communities. These tools are trained on datasets containing images of actual victims, perpetuating their exploitation. The surge also threatens to overwhelm reporting systems and law enforcement resources, potentially allowing more dangerous offenders to operate undetected. Parents face new risks when posting children's photos online, as any image can become source material for AI manipulation.

Training data ties synthetic images to real victims

Experts emphasize that the "synthetic" label does not eliminate real-world harm. McNulty, quoted in the InvestigateTV report, explained that even completely artificial files were created using AI models trained on images or videos of actual children.

This training process means every AI-generated piece of abuse material has roots in the exploitation of real victims whose images were used to teach the algorithms. The technology effectively multiplies the harm from original abuse imagery by enabling the creation of unlimited new variations.

The accessibility of generative AI tools has lowered technical barriers for offenders. What previously required sophisticated image manipulation skills can now be accomplished with consumer-grade AI applications, some of which have inadequate safeguards against misuse.

Law enforcement agencies are struggling to adapt detection and investigation methods to this new threat landscape. Traditional approaches focused on tracking the distribution of known abuse imagery become less effective when offenders can generate novel material on demand.

The National Center for Missing and Exploited Children operates the CyberTipline, which receives reports from technology companies and the public about suspected child exploitation. The dramatic increase in AI-generated material reports suggests both growing criminal activity and improved detection by platforms.

Details of the surge in AI-generated child abuse material were first reported by InvestigateTV.