The artificial intelligence (AI) website, ChatGPT, created by Elon Musk and Sam Altman, has taken the United States Medical Licensing Examination (USMLE) – and passed.
Microsoft invested US$1 billion in OpenAI in 2019 and signed a new multi-million dollar contract with the company.
According to the British newspaper The Times, the tool answered about 350 out of 376 questions.
It scored above the minimum score of 60% in one of the sessions. In two others, it passed close to approval, with 45% and 54%.
“Achieving the passing score for this notoriously difficult specialized exam and doing so without any human reinforcement represents a remarkable milestone in the clinical maturation of artificial intelligence (AI),” said the study’s author.
According to one of the authors, the results show that AI can assist the practitioner during medical consultation.
“These results suggest that large language models may have the potential to aid in medical education and influence clinical decision-making.”
Its performance, reported in the journal Plos Digital Health, will also raise fears that systems like ChatGPT could be used to cheat on exams.
AI has already passed other university exams, including one for a business administration course at the Wharton School at the University of Pennsylvania, where it earned a B grade.
A spokesman for the US General Medical Council said that there is a possibility that AI could be used to cheat on exams.
“We are aware of the possibilities of artificial intelligence being used for exams and will continue to monitor the situation to ensure that the integrity of medical evaluations is protected,” he said.
“Detecting cheating is becoming more complex in an electronic world.”
With information from Revista Oeste
This article was produced by The Rio Times’ automated newsroom system. How we use AI · Report an error · Editorial responsibility: Matthias Camenzind, Editor-in-Chief
LatAm Markets: Live Signals → — real-time movers, turnover leaders and FX across Latin America.