Search
AI-powered search, human-powered content.
scroll to top arrow or icon

{{ subpage.title }}

a robot that is standing on one foot

Hard Numbers: Doctor vs. machine, Pony rides to an IPO, Hot chips, Foxconn’s crazy demand

90: In a recent study designed to evaluate how doctors can work with large language models, ChatGPT alone achieved a 90% accuracy rate in diagnosing medical conditions from case histories, significantly outperforming human doctors. Meanwhile, physicians who used ChatGPT as an assistant scored 76% on average — only slightly better than those not using ChatGPT, who scored 74%.
Read moreShow less

Subscribe to our free newsletter, GZERO Daily

Latest