London mayor Sir Sadiq Khan has accused social media companies of failing to tackle a “dark blizzard of disinformation” about crime rates and racial integration in the city.
Khan published research by the Greater London Authority (GLA) that showed a surge in social media posts, many of which spread fake information, and some of which were promoted by accounts aligned with the extreme right, the Kremlin, Beijing and US President Donald Trump’s “Maga” movement.
Speaking at the Cambridge Disinformation summit on Thursday, Khan claimed lies were being disseminated to damage London’s position as “one of the most successful multicultural societies on Earth”. He called for the government to take action, should social media companies fail to take action voluntarily.
The research, conducted by the GLA’s City Intelligence Unit, found London is increasingly the target of false or misleading online narratives that amplify fears around knife crime, violence against women and girls, migration and policing.
The report was based on data drawn from mainstream platforms including X, Facebook, Instagram, Reddit and YouTube, as well as smaller platforms such as Telegram, VK and Weibo.
Over the past two years, posts related to London rose about 7 per cent overall while “London in decline” narratives increased up to 200 per cent, researchers found.
Khan said social media companies should act to tackle falsehoods, including fake AI imagery. If they failed to do so, the state should intervene, he said.
“Disinformation has become an industry: an ‘outrage economy’ organised around a ‘division dividend’ which allows people to profit from poison,” he said. “And today, the captains of the ‘outrage economy’ have London in their sights.”
AI-generated videos purporting to offer an insight into what London might look like in the future have gained significant traction online, accruing thousands of likes in some cases.
Khan called for the creation of a new central body with “the agility and authority to protect our democracy from disinformation”.
He also demanded “more aggressive enforcement of the rules we already have,” arguing that regulators like Ofcom needed more power.
Negative tropes about London have been seized on by Trump who has falsely claimed that parts of the capital are no-go zones for police.
A perception has developed in many countries, and within some regions of the UK, that London is unsafe to visit.
In January, Sir Mark Rowley, Metropolitan Police chief, released data showing 97 people were murdered in the capital in 2025, the lowest level since 2014 when London’s population was smaller.
The homicide rate of 1.1 per 100,000 was the lowest since comparable records began in 1997, he told the FT, and compared it with murder rates of 2.8 in New York and 11.7 in Chicago. He said London also compared well with European cities such as Berlin at 3.2 and Milan at 1.6.
In a letter sent today to YouTube, Meta, Snapchat and TikTok, Khan asked the companies to provide a “specific account” of the action they are taking to “identify and disrupt co-ordinated inauthentic behaviour” on their platforms.
He urged them to make data available to “vetted independent researchers” to enable “ongoing independent monitoring of co-ordinated disinformation targeting London and other cities”, and said he would “welcome the opportunity” to meet with them.
Meta, which owns Instagram and Facebook, said it “constantly” works to disrupt “co-ordinated behaviour.” “We are reviewing the research and are in touch directly with the Mayor’s office,” it added.
A spokesperson for TikTok, which was not included in the GLA research, said the company looked forward to engaging with the mayor on the issues he raised in his letter.
Telegram said: “Calls to violence are forbidden and are removed whenever discovered.”
Reddit said its “site-wide rules strictly prohibit manipulated content and misleading behaviour”. It added that while it was not “aware of any Reddit examples in GLA’s research . . . we will continue to take swift action against any content or accounts that violate our policies”.
X, YouTube, Snapchat, VK and Weibo have been approached for comment.
