The European Union has warned Elon Musk to comply with sweeping new laws targeting fake news and Russian propaganda after X, formerly known as Twitter, was found to have the highest proportion of false information of any major social media platform. The report analyzes the proportion of false information, revealing for the first time the scale of fake news on social media in the EU, with TikTok and LinkedIn removing millions of fake accounts.

Some posts will be deemed illegal across the EU under the Digital Services Act (DSA), which comes into force in August this year.

Still, Facebook and other tech giants, including Google, TikTok and Microsoft, have signed up to a code of conduct set out by the EU to ensure they are ready in time to operate within the confines of the new law.

Although Twitter has withdrawn from the code of practice, under the new law it is obliged to comply with the rules or face an EU-wide ban.

Věra Jourová, the EU commissioner responsible for implementing the new anti-disinformation law, said: "Mr Musk knows that he is not off the hook by departing from a code of practice. Hard laws set out our obligations. So my message to Twitter/X is that you must comply. We will pay attention to what you do. X, formerly Twitter... is the platform with the largest proportion of false/disinformation, followed by Facebook."

The 200-page report describes the preparations made by large platforms to comply with the new law in the first six months of 2023 and sheds light on the behind-the-scenes efforts of platforms such as Facebook to combat Russian propaganda, hate speech and other disinformation.

Yurova said: "Russia has launched a war of ideas, polluting our information space with half-truths and lies, and creating the illusion that democracy is no better than autocracy."

Microsoft, the owner of LinkedIn, blocked the creation of 6.7 million fake accounts and removed 24,000 pieces of false content.

Google-owned YouTube told the EU it had removed "more than 400 channels involved in coordinated influence operations linked to Russia's state-sponsored Internet Research Agency." Tiktok removed nearly 6 million fake accounts and 410 unverifiable ads. Google removed ads from nearly 300 websites associated with "state-sponsored propaganda sites" and rejected more than 140,000 political advertisers for "failed authentication procedures."

Meta has expanded its fact-checking to 26 partners, covering 22 languages ​​in the EU, and now also includes Czech and Slovak, the report said.

The report said that 37% of users canceled sharing after receiving fake news notifications, which the EU believes shows that consumers pay attention to identifying false information.

The EU is particularly concerned about Russia's continued propaganda on social media ahead of important elections in Slovakia on Sunday and Poland on October 15.

TikTok, which was recently fined €345m (£300m) for breaching children's data protection rules, is also working to comply with the DSA.

Microsoft Corp, another participant in the code of practice, told the EU that it had either promoted information or downgraded problematic information in 800,000 search queries about the war in Ukraine.

Yurova said the report proved Russia was engaged in a "war of ideas" and that Kremlin disinformation remained widespread on large platforms. The Kremlin chose Slovakia over Poland as "fertile ground" for division and interference in democracy. One of her main messages to the big platforms was to pay attention to elections, including next year's European Parliament elections, and the "risk of disinformation". The Kremlin's propaganda is "a weapon of massive manipulation costing millions of euros, both within Russia and against Europe and the rest of the world. We have to address this. Large platforms have to deal with this risk."

The war in Ukraine was the most common theme of propaganda, but the platforms also reported hate speech about immigrants, the LBGTQ+ community, and the climate crisis. "I think this is one of the advantages of disinformation, because they are very predictable. On Twitter, disinformation actors have significantly more followers than non-disinformation actors and tend to be more recent to the platform than non-disinformation users," Yurova said.