Health under attack + News/analysis
The Spectrum of Misinformation returns for 2026 with more news, analysis & practical advice for communicators.
In News digest: UK disinformation defence, traffic terrorism, AI influencer networks, follow the chemtrails, political porn, deadly DRC rumours, deepfake docs, LLM factcheck fail, contraception deception, flood video profiteers. Plus: want a chatbot with your health records? India misinformation warning, and football fake news row. Feverish, pulse racing? Maybe you've already read Health under attack, then get out the popcorn for Rocket from the crypto, before recalling the Before times.
News digest
BBC reports the UK will set up a new National Centre for Information Defence to combat what Andy Burnham called a "poisonous tide" of disinformation from hostile states, including Russia.
A 63-year old retiree has been jailed for 15 years for using explosives to blow up a Ulez camera in Sidcup, in a case linked to conspiracy theories and islamophobia.
Devon Live reports two people over 60 have been arrested following damage to traffic lights in Ilfracombe with an online group claiming the lights are “phased array electronic assault weapons”.
New York Times reveals groups from Iran, China and Israel have combined open source Chinese AI models and AI agents to rapidly seed "fake accounts with false posts about politics and current events to sway opinions and inflame issues" in a new type of AI influence campaign.
The Independent covers RFK Jr saying he is "looking at chemtrails", referring to a debunked conspiracy theory about airplane emissions. He also told a conference of anti-vaccine activists they have "a steadfast friend" at the White House.
WIRED highlights research showing that almost 150 European politicians, overwhelmingly women, have been targeted by deepfake porn websites. Meanwhile a deepfake of a Georgia mayor kissing a man has gone viral.
The Guardian reports a DRC politician has been beaten to death after doing a radio interview to raise awareness of the Ebola outbreak in the country, following a series of violent attacks fuelled by misinformation.
CBC covers fears that deepfakes of real doctors peddling fake drugs are eroding public trust and putting people's lives at risk.
A FullFact investigation has shown that LLMs make dozens of major errors when asked to factcheck false claims circulating online.
ABC News looks at misinformation about contraception spreading on TikTok and a tool developed to better inform women about their choices.
The Independent highlights fake GenAI videos shared about the Nepal floods as the disaster unfolded to gain attention and advertising revenue.
TechCrunch reports ChatGPT is being integrated with Epic's electronic health record system [read my article about health's chatbot trap].
Dozens of submissions to Australian government inquiries incorrectly summarise research or wrongly cite sources due to AI, (The Guardian).
In a Times of India interview about disinformation Joyojeet Pal says: "The WEF report identified misinformation as India’s leading short-term risk."
A new set of online resources has been published to help Canadians protect themselves from health misinformation.
Reuters covers FIFA accusing UEFA of a "misinformation campaign" against Gianni Infantino over a plan to transfer World Cup rights.
Health under attack
In 2025 an Edelman survey revealed, globally, 58% of people aged 18-34 said they regret a health decision they made based on misinformation.
This stat has lived rent free in my head ever since. If that was true globally, what was the picture in the UK? And were the consequences of these decisions merely annoying or did they lead to real harm?
It was this 'missing evidence' about health misinformation in the UK that led LSHTM to switch over the summer from a campaign to counter vaccine misinformation that included some polling, to a polling project that included questions about vaccine misinformation.
The upshot is a new report, Health Misinformation: Public Perceptions of Threat to the UK, based on a survey of 2,110 UK adults.
So what did it reveal?
Our news story highlights some of the key findings, I unpack the results below with my thoughts on what they can tell us:
1 in 5 UK 18-34 year-olds say they've made a decision based on health misinformation they later regretted
This is a lot lower than Edelman's 58% global stat. But then in the UK trust in health authorities and science is relatively high, for example a 2024 survey found 72% of Britons trust the NHS while recent polling by Wellcome showed 84% of Britons have at least some trust in science. Maybe the lesson is even in countries with trusted health authorities people will still be influenced by health misinformation, particularly where their questions go unanswered.
38% of 18-24 year-olds know someone who suffered serious harm from a health decision based on misinformation
Ouch. I found this genuinely shocking when the examples of serious harm we gave were contracting a preventable disease, delaying medical treatment, or suffering symptoms for longer than necessary. Think it's just a GenZ issue? It's worth noting 31% of 25-44 year-olds, and a fifth of all adults, say they know someone who has come to serious harm as a result of a health decision based on misinformation.
True, these could be people be in other countries/online acquaintances or some of the same people, but, if this answer was replicated across the wider UK population it would suggest that millions of people may be suffering serious harms.
19% of UK adults were either not very confident or not at all confident at judging if a health claim is accurate
While nearly half of respondents were very or somewhat confident at judging the accuracy of and checking the facts behind a health claim, 1 in 5 were not confident at either. The youngest and oldest age groups, and those with the lowest income and educational attainment, were most lacking in confidence.
It shows while an overall national stat (half are confident) sounds good it can hide a significant proportion who are vulnerable perhaps due to multiple dimensions of disadvantage. It could also explain why educating the public on how to identify false or misleading health information (54%) was one of the most popular choices for government action.
If we're failing to give vulnerable groups the skills, knowledge and tools they need to identify and factcheck health claims with confidence, how can we expect them to resist health misinformation attacks?
Where have people come across health misinformation?
27% said family or friends, 26% on traditional news media
That the majority of those who had come across health misinformation had seen it on social media (61%) didn't surprise me but the fact 1 in 4 cited family or friends or traditional news media as a source did.
What I also found intriguing was how over 50% of those aged 18-44 worried about family or friends sharing health claims without checking them first but only 28% of over-65s did. There is definitely a piece of research to be done into who is sharing what about health claims in family WhatsApps and on Facebook. Also, which news outlets are the culprits? Or is 'news media' being conflated with blogs or socials?
In last 6 months: 21% had seen misinformation on dietary supplements/weight-loss jabs... 9% on cancer
It was a safe bet vaccines would be the most common topic of health misinformation people had seen (24%). At 21% misinformation about dietary supplements/weight-loss injections was the second most cited topic, probably fuelled by wall-to-wall media coverage and intense public debate, but why was cancer misinformation so low at 9%?
It made me think about how dynamic the information environment is now: instead of a static model of people passively being exposed, people will encounter different kinds of misinformation depending on what they're actively looking for.
Unless you or someone you care about has cancer you're unlikely to go looking for information on this topic, so algorithms and chatbots are unlikely to feed you misinformation about it. But of course if you do and cancer misinformation starts cropping up in your feeds at a moment when you're vulnerable and desperate it could be extremely damaging.
Most popular actions: laws to penalise profiteers, education initiatives, labelling false or misleading health content
It's natural people want government to do something about a problem, but the support for all six suggested actions was unexpected.
57% ranked new laws against profiting from health misinformation, 54% education initiatives, and 46% labelling suspect health content in their top three actions. Other options, such as doing more to support organisations producing trustworthy health content (44%), and setting up a dedicated UK body to fight health misinformation (43%), weren't far behind. Even the least popular option, fund more research into health misinformation, ranked in the top three for 27%.
I wonder if this strong public support for action is down to people's fears about the wider impacts health misinformation could have on UK society. While at 62% the top ranked concern was health misinformation could cause people to believe in or buy fake cures, 57% said it was likely to divide people and lead to more extreme views and 55% it could persuade people to ignore government advice.
It's a reminder that health misinformation doesn't exist in a vacuum. Common misinformer tactics across all topics are to fuel extreme views, discredit experts and institutions, scapegoat, and flood the zone with slop. The same networks and sites spreading misinformation about health are often spreading false narratives about climate, immigration, politics. Perhaps we're right to worry that an attack on the integrity of health information is an attack on the integrity of all information.
Report wash-up: lessons learned
Having never run a polling project before, it taught me a lot about the process: for example turning yes/no questions into statements for people to agree or disagree with, the importance of 'don't knows' for data quality, and ordering questions so they don't prejudice later ones.
As I grappled with both a spreadsheet that wouldn't fit on my screen and impostor syndrome (who was I trying to kid, I'm not a researcher or public affairs expert!) it made me empathise with anyone who has written this kind of report. With so much data, what do you include and what do you leave out? How much should you stick to a factual report of the findings and how much context/comment should you add? How can you avoid the default of publishing as an inaccessible PDF?
What I learned is there's no 'one size fits all' template, just a lot of choices to make. Hopefully, in my attempt to deliver a report that's as interesting, accessible, and useful as possible, I made some of the right ones.
Next steps? Well there is some data on regional differences to analyse...
Thanks to everyone at LSHTM who helped with this project but particularly to my co-author Dr Ben Kasstan-Dabush for his support and contributions to both the survey design and writing of the report.
Rocket from the crypto
I recently watched the documentary Everyone is lying to you for money and I think everyone interested in misinformation should too.
Why? It's a brilliant dissection of the crypto bubble that feels startlingly relevant for the cult of AI: a world of wild claims, false narratives, and mysterious money trails. Does anyone who invests in it really understand the black box that is crypto? What about the black box of AI?
Standouts include Ben McKenzie putting on a 'dumb actor' persona so that he can seize on the misinformation spread by crypto salesmen, and his empathetic interviews with ordinary people who've ploughed their life savings into crypto and, despite losing everything, still want to believe.
Before times
In the last edition I extracted lessons for misinformation watchers from the classic 80s horror movie THEY LIVE. No, really.