There is no protection. The only defense is prevention.
In an age when a single photograph can be algorithmically transformed into evidence of abuse, British authorities are asking parents to reconsider one of the most ordinary acts of modern life: sharing images of their children online. The National Crime Agency and the Internet Watch Foundation have issued guidance acknowledging a threat most families do not yet know exists — that AI tools now allow criminals to generate child sexual abuse material from innocent public photos, without ever approaching a child directly. It is a moment that forces a reckoning between the human impulse to share joy and the cold arithmetic of technological harm.
- AI tools have made it possible to convert any publicly visible photo of a child into explicit abuse material within minutes, requiring no contact with the victim whatsoever.
- Reported cases of AI-generated child sexual abuse material rose 14 percent last year, with blackmail and extortion targeting children as young as 15 whose clothed selfies were turned into fake nudes.
- Schools have become particular targets, with extortionists scraping pupil photos from websites and threatening to publish AI-generated abuse material unless paid.
- Authorities are urging parents to make accounts private, audit and delete old posts, and revisit photo consent agreements signed with schools before AI capabilities were widely understood.
- Experts admit there is currently no technical defense against this threat — the only available protection is keeping children's images out of public view entirely.
The National Crime Agency and the Internet Watch Foundation have issued new guidance asking parents to think carefully before posting photographs of their children online. The warning confronts a troubling reality: AI tools now allow criminals to convert ordinary, innocent images into sexually explicit material without ever contacting the victims directly.
The scale of the problem is growing. Last year, the IWF identified over 8,000 AI-generated images and videos depicting child sexual abuse — a 14 percent rise on the previous year. A photograph scraped from a public Instagram account or a school website can be fed into freely available software and transformed into extreme pornography within minutes. Most parents remain entirely unaware this is possible.
The cases are real and disturbing. Minors have been blackmailed after their images were converted into fake nudes without consent. A 15-year-old girl contacted Childline after her face, sourced from her own Instagram account, was used to create a convincing fake nude. School websites have been targeted by extortionists who scraped pupil photographs, generated abuse material, and demanded payment to suppress it.
The guidance is practical: make social media accounts private, audit existing posts for visible faces or school uniforms, speak to friends and family about old images they may have shared, and review — and potentially withdraw — photo consent agreements signed with schools and clubs before AI capabilities were widely understood.
Dan Sexton, the IWF's chief technology officer, admitted he was deeply uncomfortable advising parents to restrict sharing images of their children, but felt there was no alternative. "There is no protection," he said plainly. With no technical defense currently available, prevention — keeping images offline or tightly controlled — remains the only mitigation. Authorities are now recommending that schools remove identifiable photographs of pupils from their websites altogether, a step that would have seemed extraordinary just a few years ago.
The National Crime Agency and the Internet Watch Foundation have issued new guidance asking parents to think carefully before posting photographs of their children online. The warning comes as authorities confront a troubling reality: artificial intelligence tools have made it possible for criminals to convert ordinary, innocent images of children into sexually explicit material without ever needing to contact the victims directly.
The rise of this threat has been sharp. Last year, the IWF identified 8,029 AI-generated images and videos depicting child sexual abuse material. That represents a 14 percent increase from the year before. The technology has democratized a form of abuse that once required direct predatory contact. Now, a photograph scraped from a public Instagram account, a school website, or a parent's Facebook page can be fed into freely available software and transformed into extreme pornography in minutes.
Most parents have no idea this is happening. Lorna Sinclair, a child sexual abuse education manager at the NCA, put it plainly: the average parent does not post a picture of their child thinking it might be stolen and converted into abuse material. There are countless parents and carers who do not know the problem exists at all. The guidance from the NCA and IWF is an attempt to change that, not by telling parents how to behave, but by making them aware of what is possible and what they can do about it.
The cases are real and they are disturbing. The IWF has been contacted by minors who were blackmailed after their images were "nudified" by AI—converted into fake nudes without consent. A service called Report Remove, which helps remove explicit images of under-18s taken without their consent, has documented cases where fully clothed selfies were turned into extreme pornography. Childline heard from a 15-year-old girl whose face had been used to create a convincing fake nude; the source material had apparently come from her own Instagram account. In other instances, school websites have been targeted by extortionists who scraped photographs of children, used AI tools to create abuse material, and then threatened to publish the results unless paid.
The guidance itself is straightforward. Parents should make their social media accounts private or share pictures only with a close friends group. They should audit their existing posts, looking for images where their child's face, body, or school uniform is visible, and consider whether those images should be deleted or made private. They should check whether friends or family members have posted pictures of their children, including old posts, and have conversations about what should happen to those images. They should also review any consent forms they signed with schools, nurseries, or sports clubs that gave permission to use their child's image, and consider withdrawing that consent if they are uncomfortable.
Dan Sexton, the IWF's chief technology officer, acknowledged the discomfort in this message. He said he was "very uncomfortable" telling parents not to share pictures of children online, but felt there was no other option. "I don't know what else to say to parents," he said. "I would be very cautious because there is no protection." That last phrase carries weight. There is currently no technical defense against this threat. The tools exist, they are accessible, and the only mitigation available is prevention—keeping the images offline or limiting who can see them.
The NCA and IWF have released videos showing fictional scenarios of everyday moments—a child playing sport, standing at school gates—to remind parents of the risks. The message is not alarmist, but it is clear: if you are uncomfortable sharing a photo, you should say no. The early warning working group, which includes both the NCA and IWF, has recommended that schools remove identifiable pictures of pupils' faces from their websites and social media accounts altogether. What was once considered a normal part of school life—publishing photos of children in newsletters, on websites, in annual reports—is now being reconsidered in light of what is technically possible.
Notable Quotes
The average parent does not post a picture of a child online thinking that it might be scraped to be turned into child sexual abuse material. There are lots of parents and carers who do not know that this problem exists.— Lorna Sinclair, child sexual abuse education manager at the National Crime Agency
I would be very cautious about putting pictures of children online because there is no protection.— Dan Sexton, chief technology officer at the Internet Watch Foundation
The Hearth Conversation Another angle on the story
Why is this happening now? AI image generation has existed for a few years. What changed?
The tools became free and easy to use. What used to require technical skill and expensive software is now available to anyone with a web browser. The barrier to entry collapsed.
So the guidance is essentially telling parents to assume their child's photo could be stolen and abused?
Yes. Not to frighten them, but to make them aware. Most parents post photos without thinking about it. The guidance is saying: think about it now, before you post.
What about photos that are already online? Can they be removed?
Yes, but it requires effort. Parents need to go back through years of posts, identify images of their children, and delete or make them private. And they need to ask friends and family to do the same. It's not automatic.
The article mentions schools being targeted. How does that work?
Blackmailers scrape photos from school websites, convert them to abuse material using AI, then demand payment from the school or threaten to publish. It's extortion. Schools are now being told to remove identifiable photos of children from their sites entirely.
Is there any technical solution being developed?
Not that's mentioned here. The experts are saying there is no protection. The only defense is prevention—keeping the images offline or limiting access to them.
How are children themselves being victimized?
Some have been blackmailed after their own photos were converted to fake nudes. Others have discovered convincing fake nudes of themselves circulating online, created from their social media photos. It's a form of abuse that doesn't require the child to be physically harmed.