UK Regulator Takes Aim at Scam Ads on Social Media—But Action Is Still Over a Year Away

Ofcom, the UK's media regulator, announced Friday that it wants to require major social media platforms—including Facebook, Instagram, Snapchat, X, YouTube, Google and ChatGPT—to ban advertisers who are running scams. The regulator also wants these platforms to block fraudsters from creating new accounts once they've been caught, and to give police a direct way to report suspicious ads The Guardian.

These rules would operate under the Online Safety Act, a law that gives Ofcom power to oversee how tech companies handle illegal and harmful content. The proposals also require platforms to better protect accounts from being hijacked for fraud, and to make sure financial services ads have proper approval before they go live. Oliver Griffiths, who leads Ofcom's online safety team, said the tech companies "had not done enough" to stop scammers The Guardian.

The public comment period runs until October 2026. Once these rules are finalized, platforms that refuse to follow them could face fines as high as 10% of their global revenue. But Ofcom won't make final decisions until 2027, meaning actual compliance is still at least a year away The Guardian.

The wait has already drawn complaints. Which?, a consumer rights group, welcomed the proposals but said the delay is problematic. Rocio Concha, the group's head of policy, pointed out that AI is making scams smarter and faster while regulation moves slowly. Leaving people unprotected for another year or more means scammers will have time to evolve their techniques while rules are still being written The Guardian.

The Real Problem Behind the Rules

The cost of online fraud in the UK is enormous. The government's own estimate puts it at £18.9 billion a year—a figure that includes both money lost and the wider damage to society Ofcom consultation response.

Platforms have existing legal duties to take action against scam content, but that hasn't stopped the problem. Meta, which owns Facebook and Instagram, publicly committed to stopping illegal financial ads in Britain. Yet a Reuters investigation in March 2026 found such ads running roughly 1,000 times a week on Meta's platforms Reuters. At that time, Ofcom acknowledged it had no power to act on paid scam ads because the relevant law hadn't been activated—a gap Friday's proposal aims to fill.

A Widening Tension Between Tech Companies and the Regulator

The relationship between Meta and Ofcom has become increasingly strained. In May 2026, Meta filed a legal challenge against Ofcom over online safety fees and fines, signaling the company's willingness to fight the regulator in court Reuters.

Ofcom has also criticized other platforms. In May 2026, the regulator reported that TikTok and YouTube were lagging behind competitors on protecting children from harmful content Reuters. The pattern here is clear: Ofcom is finding problems faster than it can enforce solutions.

The AI Factor Makes the Delay Riskier

Concha's warning about the timeline takes on extra weight when you consider artificial intelligence. In June 2026, the Bank of England warned the public about AI-generated scams after deepfake videos—fake videos made to look real—appeared online showing prominent figures in compromising situations The Guardian.

Financial commentator Martin Lewis has spent years fighting scammers who use his image and name in fraudulent ads to make schemes look trustworthy. This new proposal directly targets that kind of abuse. But if the rules don't take effect for over a year, scammers will have more time to use AI tools to impersonate real people and build convincing frauds.

What Comes Next

Alongside the scam-ad rules, Ofcom is proposing separate protections for news content. Platforms would have to notify publishers before removing or downranking news articles, giving outlets a chance to respond. The regulator also wants major platforms to limit how much content about suicide, self-harm, eating disorders, hate speech and abuse users see, and to give people better tools to manage their own exposure—like the ability to block unverified accounts The Guardian.

These rules are being introduced in stages rather than all at once. Each part has its own timeline and faces its own industry resistance, which is part of why critics worry the phased approach cannot keep pace with fraud that evolves by the week. Whether the 10% revenue fine actually deters bad behavior will depend on how seriously Ofcom uses it once 2027 arrives, and whether tech companies decide that compliance is cheaper than going to court—something Meta has shown it's willing to do.