So-called alternative medicine (SCAM) in the UK is subject to voluntary, publicly funded regulation. SCAM practitioners are known to make misleading health claims. This study used an artificial intelligence (AI) tool to measure the prevalence of such claims. Websites operated by practitioners of SCAM, registered with the UK ‘Complementary and Natural Healthcare Council’, were downloaded and assessed by the AI, which determined whether a website was relevant to the investigation and, if so, identified health-related claims that it judged as false or misleading, supplying a rationale.
Of 6096 registrants, 1326 met the selection criteria, of which 872 clinics had 725 relevant and operational websites. The AI assessed text from 11 771 web pages, identifying false or misleading claims in 704 (97%) of the websites. The AI’s performance was quality-assured by four human assessors, who manually reviewed 23 relevant web pages. Humans identified on average 39.5 claims likely to be judged false or misleading by advertising regulators, the AI identified 36. Humans misidentified an average of 4.8 claims, AI misidentified two.
The authors concluded that the overwhelming majority of practitioners registered with the CNHC who use pseudoscientific modalities are making false and/or misleading claims on their websites. This puts them in breach of their terms of registration. An AI tool can be used to monitor websites of practitioners promoting pseudoscientific modalities for breaches of compliance with regulators’ codes of conduct for advertising, and does so with a level of accuracy comparable with that of human assessors. It presents an opportunity for regulators to offer more effective consumer protection from their members’ online misinformation than at present.
The investigation was an unfunded project conducted entirely by volunteer scientists, motivated by the wish to protect consumers from health misinformation. The researchers acknowledge a grant of £550 from WePlanet to cover the cost of AI tokens used and thank University College London for paying the journal’s publication fee to enable the public to access their research.
The authors of this paper — emeritus professors David Colquhoun and Susan Bewley; retired clinical research consultant and HealthSense trustee Les Rose, and HealthSense newsletter editor Mandy Payne as well as IT scientist Simon Perry — feel that AI could be a game-changer in protecting the public by supporting regulators to quickly and cheaply clamp down on health misinformation so that the public can put their trust in the PSA Quality Mark logo displayed.
I congratulate the authors on their important study and hope they are correct – but I am not holding my breath.
6 Responses to Artificial Intelligence (AI) could be a game-changer in protecting the public from misinformation about so-called alternative medicine (SCAM)
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
This is what ChatGPT said about the covid-19 vaccines:
[quote]
…
But pregnancy takes 9 months – more than “a few months” – that is not long enough to prove “safe and effective” for pregnant women.
ChatGPT said:
ChatGPT
You raise an important point. The duration of pregnancy is indeed longer than the initial clinical trial phases for COVID-19 vaccines. However, the safety and efficacy of vaccines for pregnant women have been assessed through various means:
Observational Studies: After the vaccines were authorized, large-scale studies and data collection from vaccinated pregnant individuals have provided insights into outcomes for both mothers and babies. These studies have shown that the benefits of vaccination outweigh the risks.
So pregnant women were not tested by the clinical trial to see if the vaccine was safe and effective.
ChatGPT said:
ChatGPT
That’s correct. Pregnant women were not specifically included in the initial clinical trials for COVID-19 vaccines. Early studies typically focused on non-pregnant adults. However, once vaccines were authorized, extensive observational studies and real-world data have since been collected to evaluate their safety and effectiveness in pregnant populations.
[end of quote]
What on earth does this have to do with the topic of this post?
never mind OB!
The “logic”?
ChatGPT is an AI.
ChatGPT “endorses” Corona vaccines.
Old Bob has “proven” many times that Corona vaccines are ineffective or even dangerous.
Therefore AI cannot protect the public from misinformation about SCAM.
Does that sound crazy? No more than anything else OldBob constantly spouts.
I think it’s a mistake to take anything OB posts for more than comic relief.
I think that ‘Old Bob’ is simply an earlier, failed and thus abandoned version of ChatGPT. The words he uses are definitely English, but they make no sense in the way they are combined.