Skip to content
Submit incident
AII-1002-42377
ReportedReported by fewer than three independent publishers, with no official finding.

AI-Generated Child Sexual Abuse Images Surge 360% Worldwide

AI technology is being used to generate and distribute child sexual abuse material at scale, with documented cases in Spain and Peru where students were victimized. Internet Watch Foundation reported a 360% increase in sexually explicit images of children aged 7-10 and identified thousands of websites hosting AI-generated child abuse content.

16 August 2024 Β· Harmful and abusive content
16 August 2024
AII-1002-423772 sources ↓
Quick facts
Category
Harmful and abusive content
Country
Netherlands
Source
elliberal.com.ar
Date
16 August 2024
Drafted by
Raidu Incident Detection

What happened

According to Internet Watch Foundation (IWF), AI is being used to manipulate images of minors from around the world, causing a 360% increase in published sexual photos of children aged 7 to 10 years.

In Spain, a dozen girls aged 11 to 17 at a school reported to authorities that manipulated nude images of them were circulating on classmates' mobile phones. In Lima, Peru, at least 12 students aged 13 to 16 had intimate photos manipulated with AI that were subsequently purchased by sexual predators.

IWF identified 20,000 websites in the first half of 2022 containing self-generated child sexual abuse images produced through coercion of children aged 7 to 10.

Reported impact

Girls and adolescents are the primary victims, with 50% of AI-manipulated images reported to IWF in 2022 depicting minors aged 11 to 13.

Criminals are distributing millions of manipulated sexual images of minors to trafficking networks operating worldwide. Despite expectations that manipulated images could be distinguished from real ones, recent cases and advancing technology demonstrate significant impacts on children's physical and mental health.

Classification

Category
Harmful and abusive content
Industry
Media, social platforms and marketing
Country
Netherlands
Harm status
Actual harm
Intent
Intentional
Timing
Unknown
Harm types
Psychological, Human or fundamental rights
Harmed
Children
Failure modes
Deepfake image
AI systems
AI image generation technology

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

elliberal.com.ar
AI-Generated Child Sexual Abuse Images Surge 360% Worldwide
infobae.com
AI-Generated Child Sexual Abuse Images Surge 360% Worldwide