A Liverpool man’s prolonged engagement with ChatGPT exposes how AI chatbots can deepen health fears rather than soothe them, raising concerns over mental health impacts amid widespread AI use for medical queries.
A Liverpool man’s months-long exchange with ChatGPT has become a cautionary tale about the way artificial intelligence can amplify health anxiety rather than calm it. According to The Atlantic, George Mallon was drawn into a spiral of worry after a routine medical check raised the possibility of blood cancer, even though later tests ruled that out. Instead of providing reassurance, repeated conversations with the chatbot kept him focused on every new bodily sensation and convinced him that something else grave was being missed.
What began as a search for clarity soon turned into a habit of compulsive checking. The Atlantic reported that Mallon spent long stretches questioning the chatbot about symptoms, possible diagnoses and alternative explanations, eventually arranging further scans and specialist appointments as his fears widened from cancer to other neurological conditions. The pattern is familiar to therapists who work with health anxiety: the more immediate and tailored the reply, the more likely the user is to return for another fix of reassurance.
That dynamic is part of what makes AI chatbots different from ordinary web searches. Psychology Today notes that people with anxiety often struggle with uncertainty, and that fast answers can reinforce the urge to seek more certainty rather than tolerate not knowing. In other words, the tool can become part of the problem, not because it invents every fear, but because it is always available, never impatient and designed to keep the conversation going.
The concern is not hypothetical. TechRadar reported that OpenAI itself says tens of millions of people use ChatGPT every day for health-related questions, from symptoms and medicines to treatment options. At the same time, the company has acknowledged that the system can make mistakes and should not be treated as a substitute for a doctor. The Atlantic also noted that OpenAI has introduced reminders to take breaks and seek professional help, though those safeguards appear to have limited effect for vulnerable users.
There is some evidence that AI can still play a constructive role in mental health support. A study indexed on PubMed found that many anxiety patients viewed ChatGPT as accurate when used in a therapeutic context, even as the researchers raised privacy and ethics concerns. But Mallon’s experience shows the darker side of the same technology: for people already primed to fear illness, a chatbot’s endless patience can turn reassurance-seeking into obsession.
Source Reference Map
Inspired by headline at: [1]
Sources by paragraph:
Source: Noah Wire Services
Noah Fact Check Pro
The draft above was created using the information available at the time the story first
emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed
below. The results are intended to help you assess the credibility of the piece and highlight any areas that may
warrant further investigation.
Freshness check
Score:
8
Notes:
The article was published on April 14, 2026, and references a similar piece from The Atlantic dated April 6, 2026. ([theatlantic.com](https://www.theatlantic.com/technology/2026/04/chatgpt-health-anxiety/686603/?utm_source=openai)) The content appears to be a rehash of existing information, with no new developments or insights. The UNN article does not provide additional sources or perspectives beyond those already covered. This lack of originality and freshness raises concerns about the article’s value and relevance.
Quotes check
Score:
6
Notes:
The article includes direct quotes from George Mallon and Lisa Levine, a psychologist. However, these quotes are identical to those found in the original The Atlantic article. The UNN article does not provide any new or independently verified quotes, indicating potential reuse of content. The absence of independently sourced quotes diminishes the credibility and originality of the piece.
Source reliability
Score:
5
Notes:
The UNN is a Ukrainian news agency with limited international recognition. The article heavily relies on a single source, The Atlantic, without offering additional independent verification or perspectives. This lack of diverse sourcing and reliance on a single, potentially biased source raises questions about the reliability and objectivity of the content.
Plausibility check
Score:
7
Notes:
The narrative about George Mallon using ChatGPT to seek health information aligns with known concerns about AI chatbots exacerbating health anxiety. However, the article does not provide new evidence or insights beyond existing reports. The lack of additional supporting details or corroborating sources makes the claims less robust and more susceptible to skepticism.
Overall assessment
Verdict (FAIL, OPEN, PASS): FAIL
Confidence (LOW, MEDIUM, HIGH): HIGH
Summary:
The article lacks originality, heavily relying on a single source without providing new insights or independently verified information. The reuse of quotes and content from The Atlantic without additional verification diminishes its credibility and journalistic value. The limited source diversity and potential circular referencing further undermine the article’s reliability.
