Online Sexual Abuse Of Children Has Become A Pandemic. We Need To Act Now.

In three and a half decades working in policing, I have seen the landscape of child abuse change many times.

I have watched technology transform both the scale of the threat and the complexity of the response. I have sat with survivors and with their families, and I know and understand too well what this harm does.

And I have rarely seen anything escalate as fast as this.

The online sexual abuse of children is, in the most clinical and precise sense of the word, a pandemic.

The Global Child Safety Institute at the University of Edinburgh – conducting the first global estimate of its kind – found that 300 million young people are victims of technology-facilitated sexual abuse and exploitation every 12 months.

Those numbers are not projections. They are what is happening to children right now, in every country, in every community.

And AI is accelerating it, exponentially. The Internet Watch Foundation recorded a 26,000% rise in AI-generated child sexual abuse videos between 2024 and 2025. There are hundreds of millions of child sexual abuse images in circulation, and AI is fuelling the numbers as well as the demand.

What once required physical access to a child can now be generated from a photograph – perhaps a photo taken at sports day, or published by a school on its website or social media channels, with nothing but pride and good intention.

And this makes schools the latest and most troubling frontier in this crisis.

Without knowing it, without doing anything wrong, schools have become a target and a source. Every publicly accessible image of a child is now potential raw material for criminals.

The technology to exploit it is freely available, requires no expertise and costs nothing. The only thing standing between a child’s school photograph and its misuse is whether anyone has thought to protect it.

New national research published today lays bare the scale of what is happening. The In Plain Sight report from Aidos, which polled 500 UK headteachers, has found that almost one in two (49%) say their school has already experienced an incident involving pupil image misuse, manipulation or a blackmail attempt.

More than half (54%) say they have personal concerns that their school’s published images are vulnerable, and 45% say their school has already received a direct complaint from a parent about image misuse.

What makes the findings even more alarming is the gap between what schools know and what they are prepared to do about it. The majority of headteachers are aware that AI criminals are targeting school photographs, yet fewer than half – just 43% – would feel very confident knowing what steps to take if their school received a ransom demand involving manipulated pupil images tomorrow.

But schools that publish images are not doing anything wrong. They are doing what schools have always done: celebrating their children and communicating with their communities. In doing so, they are creating an opportunity they had no reason to anticipate.

The threat is anonymous, arrives without warning and its potential to cause severe and lasting trauma to children, their parents, and the staff who care for them is profound. School leaders are only now beginning to understand what they are dealing with, and are understandably struggling to know exactly what to do.

Every time AI generates a new image from a publicly available photograph, another child enters a cycle of harm they did not consent to and cannot escape. They are images of real children, and the psychological consequences for those children and their families, when they discover what has been done with their image, are devastating.

So, what can schools do? They have a duty of care to protect the children in their charge and that duty now extends to every image they publish online. That might mean reviewing what images are currently publicly accessible and for how long.

It also means ensuring safeguarding policies address the risk of pupils using nudification apps and creating deepfakes of their peers, and it means considering tools that can anonymise children’s identities in published photographs, so those images can never be misused, no matter who finds them or what technology they have access to.

This school year, I am asking every headteacher, every governor, every parent and every policymaker to take this seriously. Not next term, not when a case hits closer to home, but now.

The frameworks we have relied on to protect children were not built for this threat. That is not a criticism, it is the sad reality we are faced with, and the only responsible response to reality is to act on it.

The numbers are extraordinary, but behind every number is a child – and that is the only number that matters.

Simon Bailey served for 35 years in UK policing, including as National Policing Lead for Child Protection. He now also advises Aidos, the UK’s first AI safeguarding platform that makes pupils in school photographs permanently unidentifiable before they are shared online.