When Andy Burnham stepped up to the podium at the United Nations General Assembly in New York, the message he carried was not the usual diplomatic fare of trade deals and climate pledges. It was a warning, delivered with urgency, about a new kind of warfare being waged in plain sight—one that uses falsehoods as ammunition and social media as a battlefield. The Prime Minister announced that the United Kingdom would establish a new National Centre for Information Defence, a dedicated body designed to detect, attribute, and actively disrupt hostile state disinformation campaigns. His words were stark: the UK and its allies were facing an “industrial-scale assault” of false information, orchestrated by states that wanted to erode trust, sow chaos, and undermine the very idea of democratic consensus. He singled out Russia, which he said spends an estimated £1.3 billion every year on information manipulation. That figure alone tells a story. It is not a matter of isolated trolls or amateur propaganda; it is a professional, well-funded, relentless operation that treats truth as a strategic asset to be controlled, twisted, or destroyed. Burnham painted this threat as an attack on democracy itself, and the new centre is meant to be a shield with a sword—not merely observing hostile operations from afar, but actively working to expose them, attribute them, and neutralise them before they take hold. It was a striking admission that in the twenty-first century, national security is no longer only about missiles and borders, but about narratives, algorithms, and the fragile line between what is real and what is manufactured.
The Prime Minister did not leave his audience to wonder what this industrial-scale assault actually looks like. He laid out the Russian playbook in specific, unsettling detail. It includes bots, networks of automated accounts designed to give the illusion of grassroots opinion where none exists. It includes fake websites that mimic legitimate news outlets, fabricated newspaper articles, and false reproductions of branding belonging to at least 28 different organisations, all carefully crafted to confuse readers and erode confidence in reliable sources. Beyond the technical infrastructure, there was a strategy aimed at the ugliest currents in British society. Burnham accused Russia of attempting to interfere in the UK’s 2019 general election by amplifying far-right narratives, seeking to stoke ethnic tensions, and trying to fuel unrest and anger following horrific events. The implication is deeply troubling: hostile actors are not trying to convince people to believe a particular lie so much as they are trying to make people stop believing anything at all. They want to deepen divisions, radicalise the vulnerable, and normalise the idea that every fact is just another opinion. The method is not subtle, but it is effective. It exploits the very openness of democratic societies, using freedom of expression as cover for a coordinated attack on the foundations of informed citizenship. By laying these tactics bare, Burnham was doing more than recounting past grievances; he was issuing a challenge to the international community to recognise the scale and sophistication of what Russiand is doing, and to treat it with the same seriousness as a military threat. The language of “information warfare” can sound abstract, but the examples he cited make it chillingly concrete—this is about real elections, real communities, and real psychological damage.
The response Burnham outlined goes beyond creating a single new agency. He spoke of a 10-year strategy for the United Kingdom, to be set out before the end of the year, that would treat disinformation as a long-term, structural threat rather than a momentary problem. The phrase he used for the rising tide of falsehood was deliberately vivid and ominous: a “poisonous tide.” The new National Centre for Information Defence is a central pillar of that strategy, but it is not meant to work in isolation. The UK will share intelligence and expertise with other countries, recognising that hostile information operations do not respect borders and that the lessons learned in London or Manchester may be just as relevant in Moldova or Macedonia. Indeed, the UK has already been quietly helping Moldova strengthen its defences against Russian interference during elections, a concrete example of the kind of solidarity that Burnham wants to scale up. There was also a deeper philosophical point in his address. Defending against disinformation is not simply a technical problem; it is about defending the idea that citizens should be able to make political choices based on some approximation of the truth. It is about safeguarding the public sphere from manipulation. It is about ensuring that when people go to the polls, they are not, in effect, voting on the basis of scripts written by foreign intelligence agencies. The plan is ambitious, and there are inevitable questions about how an organisation can keep pace with rapidly evolving AI-generated content, deepfakes, and ever-changing platform algorithms. But the recognition that hostile information warfare is a permanent feature of the geopolitical landscape, rather than a passing internet nuisance, is itself a significant shift. Burnham made clear that the UK intends to play a leading role in the international counteroffensive, and that this is not a battle that any single nation can win alone.
Yet for all the focus on foreign enemies, there was another, less comfortable message emerging from Westminster. Chi Onwurah, the Labour MP who chairs the House of Commons Science and Technology Committee, welcomed the announcement of the National Centre for Information Defence. She noted that her committee had already recommended such a body in a report on social media, misinformation, and harmful algorithms. But she was blunt about its limits. Foreign interference, she argued, is only part of a much wider problem. The UK also faces a domestic misinformation crisis, and the regulator Ofcom, in her view, is “ill-equipped” to deal with it. That is a crucial reminder that the threat to truth is not always wearing a Russian flag. Political actors, online influencers, and even ordinary citizens can spread falsehoods that ripple through communities, with or without state sponsorship. The same algorithms that amplify foreign propaganda also amplify homegrown conspiracy theories, medical misinformation, and divisive content that is technically legal but deeply damaging. The Science and Technology Committee has pushed the government to go further, including measures to address algorithmic amplification—the ways in which social media platforms are designed to give the most extreme and emotionally charged content the greatest reach. There is a delicate balance here, one that every democratic government struggles with. Regulating harmful content without sliding into censorship is not easy. But Onwurah’s warning suggests that the new centre, as welcome as it is, cannot be an excuse to ignore the domestic dimension. The “poisonous tide” is not only flowing in from abroad; it is also rising from within. If the government wants to defend democracy from disinformation, it will need to confront uncomfortable questions about the business models of tech companies, the responsibilities of media platforms, and the vulnerabilities of a public that is often left to fend for itself in a sea of conflicting claims.
Burnham, however, was keen to look to the future as well as the present. He used his UN address to put frontier AI at the heart of the UK’s agenda, particularly as the country prepares to assume the presidency of the G20 next year. He described the UK’s position as that of an “honest broker” between the United States and the European Union, with a unique opportunity to shape international discussions on AI safety and governance. The term “frontier AI” refers to the most advanced and capable systems, the kind that are advancing so quickly that even their developers cannot always predict what they will be able to do in a matter of months. Burnham called it one of the biggest challenges and opportunities of our time. On one hand, AI has the potential to drive extraordinary economic growth, transform medicine, education, and science, and help solve problems that have plagued humanity for centuries. On the other hand, it can amplify existing risks in ways we are only beginning to understand. The Prime Minister pointed to the UK’s AI Security Institute, established under Rishi Sunak, which tests advanced AI models for potential safety risks. He made clear that the UK wants to use its G20 presidency to push for global principles and standards covering transparency, access to AI models, and the safe development of the technology. This is not just about technical specifications; it is about setting the rules of the road before a handful of powerful companies and nations make them unilaterally. In an era where AI can generate convincing disinformation at scale, the intersection between artificial intelligence and information defence becomes even more critical. The same technology that carries the promise of a better future could also become a weapon of mass deception, and Burnham’s emphasis on international standards reflects a growing consensus that this cannot be left to chance or to the whims of market forces alone.
The reaction from the technology and security community was cautiously optimistic. Nick Benson, CEO at CREST, a cybersecurity consultancy, welcomed Burnham’s call for robust global AI standards as a promising step. All nations, he said, are seeking to maximise the opportunities AI offers while becoming increasingly thoughtful about the risks. But Benson also struck a note of realism. AI safety is a vast and sprawling field, covering children and online safety, weapons proliferation, fraud, and, critically, cybersecurity. He argued that when actions in the digital world can threaten the physical world, the need for stronger international and national standards becomes absolutely clear. However, he warned that high-level international principles, for all their political importance, are only a starting point. Policy goals and voluntary commitments, no matter how well-intentioned, are not enough to protect AI-enabled systems or the critical infrastructure that societies depend on. This is a technical, complex, and constantly evolving field, and it demands more than speeches and statements. Managing the shift to AI requires actionable technical frameworks, independent assurance, and practical testing. As AI increases the speed and precision of cyber threats, long-term trust will depend on proving that these systems can be secured. Benson’s words serve as a fitting conclusion to Burnham’s ambitious vision. It is easy to announce new centres and call for global standards, and there is real value in doing so. But the hard work of defending democracy and building safe AI is quieter, less glamorous, and never finished. It will require sustained political will, international cooperation, technical expertise, and above all, an honest reckoning with the fact that the threats we face are not just military or technological. They are threats to the very idea of truth, to the trust that holds societies together, and to the human capacity to make deliberate, informed choices. The announcement in New York was a step, an important one, but the road ahead is long and uncertain.