AI-Generated Child Abuse Reports Surge as Global Index Urges Stronger Safeguards

November 16, 2026
AI-Generated Child Abuse Reports Surge as Global Index Urges Stronger Safeguards
  • Childlight’s 2026 Into the Light Index, unveiled at a global ministerial summit in the Philippines on 16 November, warns that AI is expanding how child sexual abuse material can be created and manipulated, and calls for stronger reporting systems, legal protections and safeguards.

  • The report estimates that more than 300 million children worldwide experience at least one form of technology-facilitated sexual exploitation or abuse each year, while around one in four may experience online abuse before age 18; representative data covering online and offline abuse were available in only 22 of the 103 countries and territories examined.

  • Reports involving AI-generated child sexual abuse material submitted to the US-based National Center for Missing and Exploited Children rose from about 4,700 in 2023 to more than 400,000 in 2025, according to Childlight.

  • That increase cannot be directly compared with the Internet Watch Foundation’s figures: the IWF assessed 6,310 AI-generated images that legally qualify as child sexual abuse material in the first half of 2026, 40% more than in all of 2025, because it counts images its analysts identified and verified while Childlight tracks company reports to NCMEC.

  • These figures count reports or detected material, not convictions, victims, offenders or distinct files, and cannot establish how much of the rise reflects more abuse, improved detection or more companies reporting.

  • The reported material includes deepfakes and images made with nudify applications, but the sources do not identify a particular tool, company or individual as responsible for the increase.

  • AI can create fabricated sexual images from ordinary photographs, including by placing real children’s faces onto explicit images; offenders may use them to groom, threaten or blackmail children into providing money or more images.

  • Among AI-generated images with recorded age and sex, 98% depicted girls, while girls accounted for 77% of children in the IWF’s broader assessment of abuse images during 2025.

  • Childlight’s research covers East Asia and the Pacific, Latin America and the Caribbean, and North America, and its global director of data, Debi Fry, said better detection and reporting may explain part of the increase but urged governments and technology companies to act quickly.

  • Researchers and child-safety advocates say safeguards are neither mandatory nor universally adopted across generative AI services, and are calling for better data, stronger laws and preventive tools that cannot be abused.

  • OpenAI, Google, Meta, Microsoft, Anthropic and other developers adopted Safety by Design commitments in 2024, including screening training data, testing models against abusive requests and preventing the generation or distribution of exploitative material.

  • OpenAI announced a free image classifier for vetted organisations on 16 November 2026; internal tests found it could reduce images sent for human review by 48% while detecting approximately the same number of abusive images as its existing technology.

Summary based on 11 sources


Get a daily email with more World News stories

More Stories