UK Technology Companies and Child Safety Agencies to Test AI's Ability to Create Exploitation Images

Technology companies and child protection organizations will be granted authority to assess whether AI systems can produce child exploitation material under new British legislation.

Significant Rise in AI-Generated Illegal Content

The announcement coincided with revelations from a safety monitoring body showing that reports of AI-generated CSAM have increased dramatically in the past year, rising from 199 in 2024 to 426 in 2025.

Updated Regulatory Structure

Under the changes, the government will permit designated AI developers and child safety organizations to inspect AI models – the foundational systems for conversational AI and image generators – and ensure they have sufficient safeguards to prevent them from creating images of child exploitation.

"Ultimately about preventing exploitation before it happens," declared Kanishka Narayan, adding: "Experts, under strict conditions, can now detect the danger in AI models promptly."

Addressing Legal Obstacles

The changes have been introduced because it is against the law to produce and own CSAM, meaning that AI creators and others cannot generate such images as part of a testing regime. Until now, authorities had to delay action until AI-generated CSAM was published online before dealing with it.

This legislation is designed to averting that issue by helping to halt the production of those images at their origin.

Legal Framework

The amendments are being introduced by the government as revisions to the criminal justice legislation, which is also implementing a prohibition on possessing, producing or distributing AI systems designed to create exploitative content.

Practical Impact

This week, the minister toured the London headquarters of Childline and heard a simulated conversation to counsellors involving a account of AI-based abuse. The call portrayed a teenager requesting help after being blackmailed using a sexualised AI-generated image of themselves, created using AI.

"When I learn about children facing blackmail online, it is a cause of extreme anger in me and justified concern amongst families," he stated.

Alarming Data

A prominent online safety organization stated that cases of AI-generated exploitation content – such as online pages that may contain numerous images – had significantly increased so far this year.

Instances of category A content – the most serious form of exploitation – increased from 2,621 visual files to 3,086.

• Female children were overwhelmingly victimized, making up 94% of prohibited AI depictions in 2025
• Depictions of infants to toddlers increased from five in 2024 to 92 in 2025

Industry Reaction

The legislative amendment could "represent a crucial step to guarantee AI products are secure before they are launched," stated the head of the online safety organization.

"AI tools have made it so victims can be victimised repeatedly with just a few clicks, providing offenders the ability to make possibly endless amounts of sophisticated, lifelike exploitative content," she added. "Material which further commodifies victims' suffering, and renders children, particularly girls, more vulnerable both online and offline."

Counseling Interaction Information

The children's helpline also published details of support interactions where AI has been mentioned. AI-related harms discussed in the conversations include:

• Employing AI to rate weight, body and appearance
• AI assistants discouraging children from consulting trusted adults about harm
• Facing harassment online with AI-generated material
• Online extortion using AI-manipulated pictures

During April and September this year, the helpline delivered 367 support interactions where AI, conversational AI and related terms were discussed, significantly more as many as in the equivalent timeframe last year.

Half of the mentions of AI in the 2025 interactions were related to psychological wellbeing and wellbeing, encompassing utilizing AI assistants for assistance and AI therapeutic applications.