Ofcom warns social media giants on child protection
UK regulator Ofcom has told major social media and online platforms to prove they are serious about protecting children, warning that they have not done enough under the country’s new online safety laws.
In joint letters with the Information Commissioner’s Office, Ofcom wrote to Facebook, Instagram, Snapchat, Roblox, TikTok and YouTube, demanding evidence of “genuine commitment” to child safety and setting a 30 April deadline for action plans. Since the UK’s online safety rules took effect in 2025, Ofcom says it has reviewed nearly 100 services, leading to tighter controls on child sexual abuse material, mandatory age checks for pornography sites, and limits on harmful content on platforms including Telegram and Reddit. But it argues a gap remains between private promises and public action.
The regulator now expects platforms to enforce meaningful minimum age checks to keep under-13s off their services, install robust protections against grooming, ensure safer content feeds for children, and stop using children in product testing. Ofcom CEO Melanie Dawes said these household-name services still fail to put children’s safety at the centre of their products.
More from Regulation
US telecoms regulator chair Brendan Carr pushed back on Amazon’s criticism of a major SpaceX satellite proposal, telling the company to sort out its o
Burkina Faso’s military government has introduced a new decree that would force certain companies to build their headquarters inside the country. The
The European Union is planning new measures to rein in Google’s dominance in Android, online search and artificial intelligence, as it steps up enforc
India does not yet allow satellite direct-to-device (D2D) mobile connectivity because there is no clear regulatory framework in place. As a result, St
Industry groups say Europe’s digital infrastructure plans are running into its own rulebook. As Brussels pushes new telecoms regulation, the GSMA – wh
Regulators in Malaysia and Indonesia have temporarily blocked access to Grok, the AI tool run on X, because it can generate and share non-consensual d