Horror rise in ‘most extreme’ AI-generated child sex abuse images as 40% are given most-severe rating
CRIMINALS are using AI to create the “most extreme” child sexual abuse scenarios imaginable, a bombshell report will reveal.
Around 40 per cent of AI images created by paedophiles last year were given the most severe grading known as Category A.
In comparison, 21 per cent of all child sexual abuse material found online fell into that category.
The shock findings will be revealed in the annual report of charity Internet Watch Foundation this week.
They have special permission to hunt down and remove sick child abuse images on the internet.
Dan Sexton, IWF’s Chief Technology Officer, said: “AI has given criminals access to their own DIY child sexual abuse tools and they are using those tools to create the most extreme sexual abuse scenarios imaginable.
“That nearly 40 per cent of the AI material we find is of this most severe kind is a disturbing reflection of the sort of abuse criminals want to see inflicted on children.”
Jess Phillips, Minister for Safeguarding, said: “The findings in the IWF’s upcoming report are appalling and show that our children – especially young girls – are being abused in the worst ways imaginable, and that concerningly the scale of this offending continues to rise.
“Much of this increase is being driven by the misuse of AI programmes, so it is vital that our laws keep pace with the threats posed by the latest technology.”
Paedos use AI models to create images on demand like “nudeifying” photos of real life children.
Some criminals are earning £1 million salaries by selling AI tools and guides on how to create images.
KIDS ABUSED IN 'WORST WAYS IMAGINABLE'
By Jess Phillips, Minister for Safeguarding and Violence against Women and Girls
“The findings in the IWF’s upcoming report are appalling and show that our children – especially young girls – are being abused in the worst ways imaginable, and that concerningly the scale of this offending continues to rise.
“Much of this increase is being driven by the misuse of AI programmes, with these sick predators able to manipulate a collection of images to produce vast quantities of abusive content.
“The images they produce then go into circulation online, which leads to children being revictimized over and over again.
“I have personally spoken to victims who tell me that years after their abuse, even as grown women, they still worry that those images are circulating.
“For too long we have been told online safety conflicts with user privacy. The protection of victims – especially children – comes first.
“Online safety and privacy do not have to be a binary choice; we can and must have both.
“As the IWF have repeatedly said, the technology exists which could be used to take down illicit images and prevent this re-victimisation.
“I applaud the IWF for the exceptional work they have already done in identifying and removing this type of vile content, but much more must be done.”
“All tech companies must do more to innovate in this space and work with us and our partners to put a stop to this.
“And the law must continue to keep pace with developments in technology.
“That’s why we are committed not just to the robust implementation of the Online Safety Act but also the introduction of new laws to crack down on online child sexual abuse.
“We are going further so vile predators who possess an AI tool designed to create illicit images, or own manuals teaching them how to do so, will rightly face time behind bars.
“Make no mistake, this government will use every tool available to ensure that our children are kept safe online, offenders are punished and we deliver our mission to halve violence against women and girls within a decade.”
New laws will mean they face time behind bars. But the IWF says the Government should look at more regulation to stop the material being created in the first place.
The IWF discovered a record number of child sexual abuse images last year on 291,270 webpages.
Each page can contain thousands of images and videos. And the majority of all reports – 97 per cent – depicted only girls.
245 reports contained AI-generated images – a 380 per cent increase on 2023.
Although AI images make up a small proportion of abuse material, authorities are really worried about how fast it is growing.