A major online-safety watchdog is warning that generative AI is making one of the internet’s darkest problems worse.
The Internet Watch Foundation says its analysts identified more photorealistic AI-generated child sexual-abuse imagery in the first six months of 2026 than in all of 2025.
The organization says increasingly realistic generation tools are being misused to create abusive material, including content based on the likenesses of real children.
This is not a story that needs sensationalism. The significance is the speed of the increase and the pressure it puts on technology companies, lawmakers and online platforms to build stronger safeguards before tools are released at scale.
What safeguards should AI companies be required to build before releasing powerful image and video models?
