A digital audit of 10,000 social media posts from former US President Donald Trump has revealed a systematic pattern of misrepresentation, prompting British intelligence to issue a formal warning about the risk of disinformation. The analysis, conducted by a coalition of cyber-forensic researchers and government cybersecurity units, maps a consistent strategy of repeating unverified claims until they achieve viral saturation, a technique known as ‘memetic warfare’ in intelligence circles.
The study, reviewed by GCHQ’s National Cyber Security Centre, found that over 70% of Trump’s posts from 2020 to 2024 contained statements that fact-checkers graded as false or misleading. More strikingly, the data shows a clear repetition cycle: a claim is made, debunked, then rephrased and reposted with slight variations, eventually overwhelming the correction algorithm. ‘This is not random venting. It’s a calculated amplification loop,’ said Dr. Eleanor Pike, a computational linguist at Oxford who contributed to the report.
The pattern exploits a cognitive bias called the ‘illusory truth effect’: the more you hear a lie, the truer it feels. By controlling the timing and framing of his posts, Trump effectively weaponised platform algorithms designed for engagement. Each reposted falsehood acted as a wormhole, dragging followers deeper into a reality bubble where traditional media corrections are dismissed as bias.
British intelligence officials privately describe this as a ‘template for digital autocracy’. The warning, circulated to NATO allies this week, highlights how the technique could be replicated by hostile states to destabilise elections, incite public health sabotage, or erode trust in institutions. ‘The vulnerability is not technical. It’s psychological. We are fighting the architecture of attention,’ said a senior GCHQ analyst under condition of anonymity.
The report stops short of accusing Trump of colluding with foreign actors, noting his posts are ‘domestically generated’. However, the pattern directly mirrors strategies used by Russian and Chinese disinformation campaigns. The difference, analysts note, is scale: Trump’s 90 million followers act as a force multiplier, with each lie gaining an organic life of its own through shares and bots.
This raises uncomfortable questions about digital sovereignty. The platforms Twitter, now X, and Truth Social allowed an ex-president to bypass editorial oversight, creating a parallel information stream. The EU’s Digital Services Act attempts to address this, but enforcement remains slow. Meanwhile, AI-generated content and deepfakes promise to accelerate the problem.
One proposed solution involves ‘immunisation therapy’: pre-emptively exposing users to weakened doses of the misinformation to build cognitive resilience. But critics argue this experiments with public perception. Another approach is mandatory fact-checking APIs linked to platform algorithms, though this risks accusations of censorship.
The immediate concern for British intelligence is the coming election cycle. With AI now enabling real-time fabrication of audio and video, the Trump playbook could be the baseline for a far more dangerous era. As one analyst put it, ‘The first casualty of this war is not truth. It is the ability to agree on a shared reality.’
