Artificial intelligence ChatGPT can clear parts of the US medical licensing exam, researchers found. The findings raised concerns about whether the AI chatbot could one day help students write the exam or prepare for it.
Victor Tseng, MD, along with his colleagues at Ansible Health, wanted to test the abilities of ChatGPT in aggregating the communications of patients, so that they can better coordinate patient care. While doing so, the members asked various multiple choice questions from the US Medical Licensing Examination (USMLE).
Tseng found the results unpredictable, and the findings were published on medRxiv. He was so much impressed with the results that he decided to allow ChatGPT to collaborate as a contributing author. As a result, the ChatGPT developed the abstract sections as well as the results. The report included minimal cosmetic adjustments by the human co-authors. The article is accepted and will soon be published in the peer-reviewed journal PLOS Digital Health.
Microsoft is highly convinced with the chatGPT and OpenAI’s other applications that it is going to invest $10billion into the company.