Ofcom is investigating whether Meta checked the risks to children before it launched Instagram Instants. The case tests a UK duty to assess new features before release, while Washington leans on industry self-policing.
Ofcom opened a formal investigation into Meta on 6 October over Instagram Instants. The feature, launched in May, lets users share photos that disappear once viewed. Ofcom wants to know whether Meta assessed the risk of illegal content and harm to children before launch. If Meta broke the law, the regulator can fine it £18m or 10% of qualifying worldwide revenue, whichever is greater.
Under the Online Safety Act, platforms must update their risk assessments before they make a significant change to a service. Ofcom warned major platforms earlier this year that unassessed changes could bring enforcement. “Our online safety laws are clear – significant changes to platforms must be risk assessed before they’re launched,” said George Lusty, Ofcom’s director of enforcement.
Meta says it ran a risk analysis and briefed Ofcom several times before launch. Instants blocks screenshots and applies Teen Account protections by default, the company adds. In January, Ofcom closed a case against Snap without penalty after Snap submitted a revised risk assessment for illegal content. The Instants case will show whether a platform can launch first and assess later.
Ofcom’s move is the latest in a run of clashes between regulators and social media and AI companies over safety. Ofcom also has open cases against TikTok, X and 4chan. Since January it has investigated Novi, which runs an AI companion chatbot, over its age checks. In New York, OpenAI, Anthropic, Google and Meta testified under oath to the City Council on 5 October. None would guarantee that their systems always follow safety guardrails.
A $17.1bn settlement in the US
In the United States, Meta settled with 51 attorneys general on 26 August for up to $17.1bn. The deal guarantees $12.1bn and rises if rival platforms settle on similar terms. Meta must verify users’ ages, cap under-18s at two hours a day and block their access between midnight and 6am. It admitted no wrongdoing.
On 25 September, TikTok agreed to pay Alabama at least $100m. It will also bring in a two-hour daily limit and stronger age checks.
Brussels targets design
The European Commission issued its first fine under the Digital Services Act in December 2025. It fined X €120m over its paid blue checkmarks, its advertising archive and limits on researchers’ access to data. On 10 July, the Commission’s preliminary findings said infinite scroll, autoplay and push notifications on Instagram and Facebook breach the same law. In its view, Meta failed to assess how these features and personalised recommender systems affect minors’ mental health. DSA fines can reach 6% of global turnover. Brussels and London are now testing the same duty: assess the risk before users meet a feature.
In the High Court and the Upper Tribunal, Meta is contesting how Ofcom applies the Act. Its legal challenge targets the decision to place Instagram and WhatsApp in Category 1, the tier with the heaviest duties. It also disputes the use of global turnover to set fines. A hearing is due this month.
Washington prefers self-policing
On 29 September, President Donald Trump announced a voluntary accord signed by AI leaders including Meta’s Mark Zuckerberg. The Joint Commitment on Frontier Responsibilities promises independent auditors and board-level oversight of the most powerful AI models. Its commitments focus on frontier risks such as cyberattacks and biological weapons, with no specific terms on children or social media. “I think I’m seeing tremendous self-policing. And they understand that they have to self-police,” Trump said.
State attorneys general and courts have applied most of the US pressure on platforms, using consumer law. Some states are acting too. California’s governor ordered an AI kill switch in September. Britain and the EU have put the risk check into law, and it applies before a feature ships.
Ofcom’s announcement gives no date for a decision. A finding against Meta would apply the pre-launch duty to one of the world’s largest platforms. A fine set on global revenue would also test the turnover rule Meta is fighting in court.
Subscribe
Sign-up to receive our newsletter
