A new study says a major video platform is still steering young users to risky content, despite fresh safeguards meant to protect minors in the United Kingdom. Researchers reported that the service kept promoting material flagged as harmful to teenagers. The finding raises questions about how well new child-safety rules are working, and whether recommendation systems are being checked.
The site is continuing to recommend harmful videos despite new rules meant to better protect teenagers in the UK, a study finds.
The review comes as the UK expands online safety requirements. Lawmakers and regulators have pushed platforms to limit violent, self-harm, eating disorder, and age-inappropriate content. The timing matters, since parents are heading into school holidays and screen time often rises.
New Rules, Old Concerns
The UK has moved to protect children online through several measures. The Age Appropriate Design Code requires services likely used by children to apply stricter data and safety controls. The Online Safety Act, passed in 2023, gives Ofcom new enforcement powers. It also expects platforms to assess risks and reduce exposure to harmful content.
Many large platforms have said they updated policies and tools in response. They cite teen-specific defaults, more content filters, and tougher limits on how recommendation engines work for young users. The study’s finding suggests gaps remain between policy and practice.
What the Researchers Reported
The study describes continued promotion of videos that could harm teenagers. It suggests recommendation systems remain tuned for engagement, even when that may deepen exposure to risky themes. The research points to a core tension between safety and attention-based ranking.
While the full methods were not released in the summary, the conclusion is clear. The platform’s safety changes did not stop the flow of flagged material to teen accounts. That result will likely prompt closer review by regulators and child-safety groups.
Why Recommendations Matter
Recommendation engines drive what users see next. For teens, a few clicks can quickly shape a feed. That can lead to repeated exposure to content about self-harm, extreme dieting, or other sensitive topics.
- Repeated exposure can normalize risky behavior.
- Algorithms can create feedback loops after a single interaction.
- Age signals are not always accurate, which can weaken safeguards.
Experts have long warned that moderation alone may not be enough. How the ranking system is tuned may decide what young users watch most of the time.
Industry Response and Accountability
Platforms often say they remove or limit content that violates rules. They promote features like restricted modes, time limits, and keyword filters for parents. They also publish transparency reports and invest in moderation staff and machine learning tools.
Child-safety advocates argue for independent audits of recommendation systems. They want regular risk assessments and clearer explanations of how content is ranked for minors. They also push for safer defaults, such as turning off autoplay or tightening age checks.
Regulators now have more leverage. Ofcom can set codes of practice under the Online Safety Act and pursue penalties for noncompliance. The study’s finding could feed into those actions, especially if replicated by other research teams.
What To Watch Next
The key test will be whether platforms can align incentives for engagement with safety for minors. That could include shifting ranking goals, reducing the weight of watch time, and interrupting harmful viewing paths.
Researchers may expand testing to track how slight user actions change a teen’s feed over days and weeks. Audits that simulate teen behavior can show whether the same risks appear across languages, regions, and topics. Public results can help parents and schools plan safeguards.
Lawmakers will watch how fast platforms respond. If harmful recommendations persist, enforcement could accelerate. That might mean deadlines for safer settings, more transparent reporting, or requirements for independent system audits.
The study’s conclusion is simple but urgent. If a platform continues to promote harmful videos to teenagers in the UK despite new rules, then policy, design, or both need adjustment. The next phase will test whether companies, regulators, and researchers can close the gap between public promises and what children actually see.
