UK aims to curb AI sex abuse to prompt more testing

The UK government will allow tech workers and child safety charities to activate artificial intelligence tools to ensure they don’t visualize sexual abuse.

A reform of the crime and policing bill announced on Wednesday will enable “authorized testers” to identify models for their ability to sexually abuse children before they are released.

Technology Secretary Liz Kendall said the measures would “ensure AI systems are secure at source” – although some campaigns still need to be done.

This comes as the Internet Watch Foundation (IWF) says that the number of AI-related reports has doubled in the past year.

The charity, one of only a few licensees in the world to actively look for child abuse content online, said it took down 426 pieces of reported material between January and October 2025.

It rose from 199 in the same period to 2024, it said.

Its Chief Executive, Kerry Smith, welcomed the government’s proposals, saying they would build on long-standing efforts to combat online CSAM.

“AI tools make it so that survivors can be victimized in just a few clicks, giving criminals the ability to create a limited amount of sophisticated,” Photorealistic sexuised material,” he said.

“Today’s announcement will be an important step to ensure that AI products are safe before they are released.”

Rani Govender, Policy Manager for Child Safety Online at the NSPCC, welcomed the move to encourage companies to have more flexibility and scrutiny of their child safety models.

“But to make a real difference for kids, it’s not going to be an option,” he said.

“The government should ensure that there is a mandatory duty for AI developers to use this provision so that protection against child sexual abuse is an important product design.”

The government said the proposed changes to the Act will also prepare developers and staff to ensure there are adequate safeguards around extreme pornography and inappropriate images.

Child safety experts and organizations often warn against AI tools, in part, using too much online content to open to children or not allow adults.

the others, including the IWF And the Charity Charity Charity Thorn, said that this risk harming the efforts of the police on such material by making it difficult to identify whether such content is real or fabricated content.

Researchers suggest that there is an increasing demand for these images online, especially on the dark weband that’s the rest children do.

Last year, the UK said the UK would become the first country in the world to make it illegal to possess, manufacture or distribute AI sexual abuse tools with up to five years in prison.

Ms Kendall said on Wednesday that “by empowering trusted organizations to scrutinize their AI models, we are ensuring child safety is designed into AI systems, not bolted on as an afterthought”.

“We cannot allow the advancement of technology to compromise our ability to keep children safe,” he said.

Protection minister Jess Phillips said the measures “also mean legitimate AI tools cannot be manipulated to produce malicious material and more children will be protected from predators as a result”.