There is a particularly unsettling kind of online lie that no longer looks or sounds like a lie. It might be a video of a famous actor saying something outrageous, or a news report from a trusted global brand that never actually aired. Such is the reality of Russian disinformation today. Take the deepfake of Tom Cruise criticising the Paris Olympics, or the doctored footage with the BBC logo that blamed Ukraine for a missile strike on its own citizens in Kramatorsk. These are not random pranks or glitches in the algorithmic matrix. They are part of an organised, industrial-scale effort to distort public perception and destabilise democratic societies. The strategies are constantly shifting. Sometimes the lie is spread directly through Telegram, the encrypted messaging platform with huge reach. Sometimes the goal is subtler: to flood the internet with false stories, knowing that AI chatbots will eventually absorb them and repeat them as truth. This is the “poisonous tide” that Andy Burnham, the mayor of Greater Manchester, has vowed to push back against with a new body called the National Centre for Information Defence. Burnham described the campaign as reaching into people’s homes and twisting their feelings about their own country and community, creating a narrative of decline, stoking division, and sowing despair. It is an unusually personal framing, and it reflects a growing understanding that disinformation is not simply about bad information, but about emotional manipulation on a mass scale. The question now is whether Western governments, which have often been slow and reactive, can match the flexibility, speed, and sheer creativity of their adversaries.
One of the most important lessons in the article is that trying to debunk every viral lie is a “losers’ game.” Ned Mendez, head of research at Alliance4Europe, a nonprofit that analyses destabilising threats, says chasing individual TikTok videos or X posts is a waste of time. The problem is not any single falsehood but the infrastructure that enables and amplifies them. That means social media platforms, automated accounts, networks of influencers, and the algorithms that feed on outrage. Sophie Williams-Dunning, a cyber analyst at the Royal United Services Institute, a British thinktank, describes a “sophisticated ecosystem” that exists to “exacerbate provokable users.” This is a crucial point to absorb. It is tempting to imagine Russian trolls sitting in a basement inventing brand new divisions from scratch. In reality, the ecosystem is opportunistic. It latches onto real tensions. Far-right influencers immediately jump on flashpoints like the Southport and Belfast riots. There are also people who simply want to make money, creating hyper-engaging content that exploits disruption for clicks and ad revenue. Russia is often happy to amplify this content, but it didn’t necessarily create all of it. Williams-Dunning calls Burnham’s announcement “welcome, if overdue.” She also warns of a tricky side effect: any government effort to counter disinformation will inevitably spark debate about free speech and the proper role of state intervention. And, predictably, malign networks will seize on that debate and amplify it, making the countermeasure itself look authoritarian. This is one of the paradoxes of disinformation: the very act of fighting it can be used against the people fighting it.
The UK’s response has been slower than some of its allies. In May, the Starmer government warned that Britain was confronting “an industrial-scale attack” on the information environment, aimed at undermining security and democratic systems. MPs on the foreign affairs select committee had already called for a national counter-disinformation centre, similar to Sweden’s Psychological Defence Agency or France’s Vigilance and Protection against Foreign Digital Interference service. The new National Centre for Information Defence, announced by Burnham, suggests that the UK is finally catching up. But it isn’t starting from scratch. After the Southport disturbances in 2024, the government launched the National Internet Intelligence Investigations unit, run by the police, which monitors social media to help local forces manage public safety threats. Then, in May 2025, it sanctioned employees of the Social Design Agency and ANO Dialog, two Russian entities described as tools of the Kremlin’s large-scale interference campaigns. These are direct moves to dismantle the machinery of disinformation rather than just counter its output. There is also the notorious Doppelgänger network, so named because it imitates established media organisations like Le Monde, the Guardian, Der Spiegel, and Fox News. These fake mirror sites can look remarkably real, and they are commonly used to spread pro-Russian narratives. Each of these details matters because it shows that the problem is not abstract. It has a structure, actors, and a pattern. But it also reveals a fundamental difficulty: how do you disrupt a network that can appear overnight, in any language, on any platform?
Consider some of the most prominent examples of how these campaigns operate. In 2023, there was a sudden wave of reports about bedbug infestations in France, with stories falsely blaming Ukrainian refugees for bringing the insects into the country. The falsehood leaked into the British press, and the French government ultimately concluded that Russian-linked social media accounts were artificially amplifying the scare. The goal, seemingly, was to inflame anti-refugee sentiment and undermine European unity. Another long-running narrative is that London is “dangerous” and “lawless.” Researchers at King’s College London analysed Reddit posts and found this theme grew from 874 mentions in 2008 to 258,444 in 2024. That is a staggering increase. But, as Robert Topinka from Birkbeck, University of London, cautions, Russia did not necessarily start such storylines. Rather, it found a point of existing tension and enthusiastically amplified it. This is a critical nuance. If we assume that ordinary people are just passive victims being fooled by an outside actor, we miss the deeper truth. Many people already feel that their country is run-down, that things are getting worse, and that politicians are out of touch. Disinformation feeds on that feeling. It doesn’t have to invent a problem from nothing; it just needs to pour fuel on a fire. This is what makes the fight so hard. Counter-disinformation cannot simply be about correcting facts. It needs to address the underlying grievances and anxieties that make people receptive to false stories. Burnham’s language about a “narrative of decline” and “sowing despair” suggests he understands that. But turning that understanding into effective policy is another matter entirely.
Artificial intelligence is making all of this more difficult. Meta, the company that owns Facebook and Instagram, says AI-generated content is now a “near-standard” part of disinformation campaigns. Gone are the days when fake news required a human to write every post. Now, generative tools can produce endless variations of a lie in seconds, with images, voices, and video that are incredibly hard to distinguish from reality. The UK thinktank Demos recently highlighted a “poisoning” campaign in which a discredited Russian organisation published falsehoods online, including the grotesque claim that the bodies of deceased Ukrainian soldiers were being processed into meat products. Later, chatbot models were found repeating these claims as if they were facts. This is “poisoning” in a literal sense: if false stories litter the internet, any algorithm trained on that data will eventually treat them as true. The consequences are severe. Not only do deepfakes undermine trust in real footage, but they also create a sense that nothing can be believed. The very existence of AI-generated content gives bad actors a defence: when a real video emerges showing something embarrassing, they can claim it is a deepfake. It is a classic information-warfare tactic, and it is only going to become more frequent. On the other hand, AI also provides tools for defenders. Automated detection systems can identify patterns of coordinated behaviour, flag suspect accounts, and trace influence operations in ways that humans cannot. But it is an arms race, and right now the offensive side has the advantage.
Burnham’s announcement is a significant step, but it also resembles the opening round of a long and difficult campaign. The National Centre for Information Defence will have to do more than publish rebuttals. It will need to build alliances across governments, tech companies, civil society, and community leaders. It will need to develop genuine transparency about its methods so that concerns about censorship don’t become a weapon against it. And it will need to be humble enough to recognise that disinformation cannot be separated from the real-world conditions that make it thrive. As Robert Topinka noted, people are not simply being fooled; they might be reacting to their actual lived experience of a country in decline. If the new centre treats them only as victims of foreign manipulation, it will fail. If it instead treats them as people with legitimate anxieties who are being cynically exploited, it has a chance. There is no easy solution. The “poisonous tide” that Burnham describes cannot be reversed with a single government agency. But acknowledging that the problem is systemic, naming the infrastructure, and coordinating a response is the right place to start. Whether it can keep up with the evolving tactics, from deepfakes to chatbot poisoning, will determine whether future elections and public debates remain anchored in some sense of shared reality. The article ends with a simple truth: the centre will have a lot of work to do.