The UK bans strangulation and child roleplay porn. See how Tuteliq helps adult platforms detect illegal content accurately, without blocking lawful uploads.

Key takeaways

  • Accuracy protects compliance and revenue. Missed illegal content exposes platforms to enforcement, while false positives block lawful creators.
  • Payment processing is at risk. Card networks can withdraw support when platforms host unlawful content.
  • Context matters. Images, video, audio and text need to be read together to distinguish prohibited material from lawful adult content.
  • Platforms remain in control. Detection supports platform-set thresholds, human review and clear evidence for decisions.

The rules for adult platforms have changed

Since 6 November 2026, it has been illegal in England and Wales to possess or publish pornography showing strangulation or suffocation, sex between people who are related or pretending to be, or adults roleplaying as children under 16. The BBC has been covering what this means in practice, and the message for adult platforms is clear: content that was legal last month may now be a criminal offence to host.

The law did not arrive from nowhere. Baroness Bertin's independent review found that strangulation had become "totally normalised" in mainstream pornography. Research cited alongside the ban found that 43% of sexually active 16 and 17 year olds said they had been strangled or choked during sex. As a criminologist, I see this as a textbook case of harm that is learned, rehearsed and repeated. As a mother, I see why the law had to change.

The government has also said that depicting strangulation in pornography will be a priority offence under the Online Safety Act. That means platforms are expected to take proactive steps, such as automated systems that detect and hide images before users see them. Waiting for a user report is no longer enough.

Payment providers are the enforcers you cannot appeal to

Ofcom can fine a platform. A card network can switch off its income overnight. For adult services, that second risk is often the more immediate one.

It has happened before. In December 2020, Mastercard terminated acceptance of its cards on Pornhub after its own review "confirmed violations of our standards prohibiting unlawful content". Visa suspended acceptance pending its own investigation. Mastercard said it was looking at other websites too. Creators who had done nothing wrong lost a major source of income in the same week.

The lesson for every adult platform is that compliance is not only a legal question. It is a commercial one. Payment providers require their partners to keep illegal content off their services, and the UK has just widened what counts as illegal. A small number of non-compliant uploads, found by a journalist, a campaigner or a card network, can put the whole business at risk.

Accuracy is a compliance issue and a revenue issue

Adult platforms face a problem most safety tools were never designed for. Almost everything on the platform is sexual, and almost all of it is lawful. A blunt filter tuned for a children's app would flag nearly every upload.

That leaves two ways to get it wrong, and both cost money:

  • Missed content (false negatives). Illegal material stays live. The platform is exposed to criminal liability, Ofcom enforcement and the loss of payment processing.
  • Wrong flags (false positives). Lawful content is blocked or delayed. Creators lose earnings, trust in the platform drops, and moderation teams drown in reviews that should never have reached them.

Platforms often respond to the first risk by over-blocking, then quietly loosen the filters when revenue suffers. Neither position holds up. What adult platforms need is precision: detection that can tell the difference between consensual adult content and content the law now prohibits, and explain why.

How Tuteliq works with adult platforms

Tuteliq is the detection layer that sits underneath a platform's moderation. For adult services, it does three things in a single pass over every upload.

Detecting previously unseen CSAM

Hash matching only finds images that have already been identified and catalogued. It cannot see new material. Tuteliq's patent-pending approach detects previously unseen child sexual abuse material without ever training on victims' data. This matters because Ofcom has just extended hash-matching duties to pornography sites and is requiring automated detection of unknown CSAM on file-sharing services and the largest social media platforms. Its Online Safety Group Director put it plainly: "It is now for industry to step up."

Detecting strangulation and suffocation

This is built into the same image and video analysis, not bolted on afterwards. A platform does not need a separate vendor or a separate integration to meet the new UK offence.

Reading context across formats

Adult roleplay involving someone presented as under 16 is now illegal, but it is often signalled through titles, descriptions, captions and dialogue rather than the images themselves. Tuteliq analyses text, voice, image and video together, in 32 languages, so these signals are caught across formats.

The point is not to block more. It is to flag the right things, with a clear reason, so that lawful content keeps flowing and creators keep earning.

What this looks like in practice

A typical adult platform integration follows four steps:

  1. Upload. A creator submits a video with a title and description. The platform sends it to Tuteliq through our API or SDK, or runs Tuteliq inside its own infrastructure, air-gapped if required.
  2. Analysis. Tuteliq checks image, video, audio and text together for novel CSAM, strangulation or suffocation, and roleplay signals, and returns a risk score.
  3. Decision. The platform sets its own thresholds. Clear risks can be held automatically before publication. Borderline cases go to a human reviewer. Lawful content publishes without delay.
  4. Evidence. Every flag comes with an encrypted explanation of why it was raised, which only the platform's moderator can decrypt. That record shows regulators and payment partners that the platform acted proactively.

Privacy and independent evaluation

Tuteliq never retains content and never trains on it. Moderators are told why something was flagged, without Tuteliq storing or displaying the material itself. For an industry built on the privacy of its users and creators, that matters as much as the detection.

Our detection model has also been independently evaluated, and the current version includes the improvements that evaluation identified. Accuracy is not something we ask platforms to take on trust.

Safe, compliant and open for business

Most adult platforms want to do the right thing. They want to protect the people in their content, keep their payment partners, and stay on the right side of a law that has just moved. What they cannot afford is a tool that treats every upload as a threat.

Tuteliq was built by criminologists and engineers to see harm precisely. For adult platforms, that means catching what the law prohibits and letting everything else through.

If you run or work for an adult platform and want to see how this would work on your content, visit tuteliq.ai to book a demo.

Dr Nicola Harding is a criminologist and Co-Founder and Chief Scientific Officer of Tuteliq.

Sources

Get Started Free · Read Documentation · View Pricing