AI recreation art
Britain is introducing tougher online safety rules requiring tech platforms to tackle AI-generated child abuse material (for illustration purposes only) Andres Siimon/Unsplash

Britain is introducing tougher online safety rules that could force technology giants to crack down on child sexual abuse material or face fines worth up to 10% of their global revenue.

Announced by media regulator Ofcom on 16 November, the measures will require social media platforms, file-sharing services, pornography websites and other online services to take stronger action against child exploitation and grooming.

The crackdown is a response to artificial intelligence making it increasingly easy to generate realistic abusive images, creating new challenges for companies responsible for moderating online content.

Under the proposed rules, services will have to ban users who generate, upload, or share child sexual abuse material. They will also face stricter requirements to identify and remove illegal media, with some platforms expected to deploy automation to detect previously unknown material.

The measures are expected to take effect in early 2027, subject to parliamentary approval.

'While we have seen some progress from industry on tackling child sexual abuse on their platforms, we are not satisfied that enough is being done,' Oliver Griffiths, director of Ofcom's Online Safety Group, said.

AI-Generated Abuse Is Rising at Alarming Rates

The new requirements come amid mounting evidence that artificial intelligence is accelerating the spread of child sexual abuse material.

Research by Edinburgh-based Childlight – Global Child Safety Institute found that reports of AI-related child sexual exploitation increased by more than 8,400% in just two years.

The number of reports involving AI-generated abuse material, including deepfakes created using so-called 'nudify apps', rose from approximately 4,700 in 2023 to more than 400,000 in 2025.

Although some images are fabricated, researchers warned that they can still cause serious harm. Criminals can use manipulated images to blackmail children, demanding money or additional photographs.

The Internet Watch Foundation (IWF) also reported a sharp increase in AI-generated abuse material. Its analysts assessed 6,310 AI-generated images that met the legal definition of child sexual abuse during the first half of 2026, already exceeding the more than 4,500 identified throughout 2025.

Prof Debi Fry, Childlight's global director of data, described the findings as evidence of the 'shocking' scale of online child exploitation.

She also cautioned that improved detection and reporting might explain some of the increase, while stressing the need for urgent action from governments and technology companies.

Platforms Face Tougher Detection Requirements

One of Ofcom's central requirements involves expanding the use of 'hash-matching' technology.

The system compares uploaded images and videos against databases of previously identified illegal material, allowing platforms to detect and remove known content automatically. Relevant services will also have to use the technology to identify known terrorist material.

Ofcom is introducing additional protections against online grooming, requiring platforms covered by these duties to protect children's accounts by default.

Certain services will also have to take greater precautions before recommending content to children through their algorithms.

The crackdown forms part of Britain's Online Safety Act, which gives Ofcom the power to fine companies up to 10% of their worldwide revenue.

'The online world has been left to police itself for too long, and it is children and families who have paid the price,' online safety minister Ian Murray said.

The measures arrive alongside Britain's separate plan to prevent children under 16 from accessing platforms including TikTok, Instagram, Snapchat and X.

Ofcom has already launched investigations into several technology companies, including Meta's Instagram services, as the government moves towards stricter online child protection.