Using ChatGPT for Health Analysis: Be Careful!

Due to the onset of the AI revolution, it is becoming increasingly common for the common public to start using AI services, even in the health sector. People have been using ChatGPT as a medium to analyse their health data, and this is turning out to be a dangerous trend.

A recent study has raised concerns about people using AI chatbots like ChatGPT as a doctor to diagnose health problems. Researchers warn that relying on such tools for medical advice can lead to dangerous mistakes and delayed treatment.

The research, conducted by experts from the Oxford Internet Institute and the University of Oxford, involved nearly 1,300 participants who were given different medical scenarios and asked to identify the likely condition and decide whether medical care was needed.

Some participants used AI chatbots for help, while others relied on internet searches or personal judgment.

The findings showed that people who used AI tools correctly identified a relevant medical condition in fewer than 34.5 percent of cases and chose the appropriate level of care in fewer than 44 percent. Surprisingly, participants who did not use AI often performed better in identifying the problem.

Researchers said the issue often arises because users provide incomplete information or struggle to interpret the AI’s multiple suggested diagnoses. In some cases, the chatbot even gave conflicting advice for similar symptoms, including one situation where a serious brain condition received opposite recommendations from the system.

Medical experts warn that while AI tools can provide general health information, they are not ready to replace doctors. The study concluded that none of the tested language models are currently safe for direct patient care without professional supervision.