A man’s journey to ChatGPT for health advice took a dangerous turn, resulting in poisoning. The 60-year-old individual found himself on a psychiatric hold after unintentionally poisoning himself due to a misunderstanding of advice he received from the AI program. Cara Lynn Shultz reports that the man, with no prior psychiatric or medical history, arrived at the emergency room claiming his neighbour was poisoning him. Doctors, prompted by his blood work results, admitted him for further assessment.
:max_bytes(150000):strip_icc():format(jpeg)/ChatGPT-phone-081225-036a3a6a300b492a9312b1f5f20cfe9a.jpg)
During his admission process, the man mentioned adhering to a strict diet and drinking distilled water, showing signs of paranoia towards the water offered to him. Things took a serious turn within a day as he began experiencing paranoia, along with auditory and visual hallucinations. Attempting to escape, he was then placed on involuntary psychiatric hold due to “grave disability”. Medical tests suggested bromism, a condition caused by bromine poisoning, as the root of his deteriorating health.
To the man’s shock, he realised he had poisoned himself by substituting sodium chloride with sodium bromide, guided by ChatGPT’s advice. The man had misinterpreted the AI’s suggestion, believing it to be a suitable dietary swap. Bromine is a chemical element used in various industries such as agriculture and fire suppression. Survivors of bromine poisoning often face long-term consequences due to the lack of a cure, with treatments mainly focusing on supportive care.
Following a three-week-long treatment period, the man was eventually discharged. Researchers attempted to replicate the man’s ChatGPT interactions, generating responses that included bromide without specific health warnings or further inquiries. This incident sheds light on the potential risks posed by AI in disseminating decontextualised information, leading to preventable adverse health outcomes. The man’s case serves as a cautionary tale, emphasising the importance of seeking professional medical advice rather than relying solely on AI recommendations.
As the technology landscape continues to evolve, it’s crucial for individuals to exercise caution when turning to AI for health guidance. The incident involving the man’s poisoning serves as a stark reminder of the limitations and risks associated with relying solely on AI platforms for critical health decisions. Seeking professional medical advice and engaging in informed discussions with healthcare providers remain paramount in ensuring one’s well-being and safety.
The research community has highlighted the dual nature of AI, acknowledging its potential to bridge gaps between experts and the general public while underscoring the risks of disseminating misinformation. The case study involving ChatGPT underscores the importance of context and precision in AI-generated responses, especially in the realm of healthcare. While AI tools can offer valuable insights, human oversight and discernment are indispensable in safeguarding against harmful outcomes.
In a bid to enhance public awareness and safety, researchers caution against complacency when interpreting advice from AI platforms. By prioritising clarity, context, and accuracy in AI-generated information, the healthcare community can mitigate the risk of inadvertent harm caused by misinterpreted guidance. The incident serves as a poignant reminder of the nuanced relationship between AI technology and human decision-making, urging for a balanced approach that integrates AI capabilities with human judgement to foster positive health outcomes.