ChatGPT has come up as an important development in the AI world, with experts weighing in on its potential for medical treatment. The latest version, GPT-4, can score 100% on medical licensing exams, finding the most accurate results possible even in medical disputes. It is also capable of image processing, being able to read and provide analyses from MRI scans and CT scans.
The capabilities of GPT-4 reach further. Even tasks requiring human compassion, such as delivering bad news to patients, can be accomplished highly effectively with the help of this AI-driven technology. While the technology is seen as a potentially significant asset to medical practitioners, the extent to which it should be embraced is not yet understood. Andrew Beam, a professor of biomedical informatics at Harvard, sees the potential in GPT-4’s technology, but is aware of the limitations in understanding when and where it would be both practical and ethical to follow its recommendations.
This is made more alarming as people will naturally rely on this technology as a form of self-care, relying on information put out by the AI with limited awareness of the reliability of its sources. With AI systems like this, people may make questionable decisions in terms of their health care, potentially leading to a decreased life expectancy compared to what could have been achieved.
That said, GPT-4 has benefits, too. It could help to reduce the amount of paperwork medical professionals have to do, helping doctors spend more time with patients and avoid burnout. Dr Isaac Kohane has already had the chance to experiment with GPT-4 and was so impressed he made it into a book, which looks at its potential to improve medical practices going forward.
It should be noted that GPT-4 is not to be blindly trusted. In many cases, it does not understand the full context of a situation and can be fed different prompts to get different results, even if the intentions are the same. For this reason, it is never meant to replace human work, instead helping to enhance it. It is up to every medical professional when and where to appropriately utilize this technology in the most level-headed and ethical way possible.
Microsoft’s Peter Lee and former Bloomberg journalist Carey Goldberg co-authored Dr. Kohane’s book, The AI Revolution in Medicine: GPT-4 and Beyond. This book can act as an insight into the best ways to use GPT-4 in the right way, giving an overall better picture of how this technology can improve medicine.
At the end of the day, ChatGPT is a powerful tool but requires great finesse on the part of the medical practitioner to utilize it correctly. The key is understanding when it is appropriate to use the AI in order to give the best possible advice, delivered in the right manner, which gives the patient a better experience.
