Complex medical questions? Your doctor might be using AI for that

An artificial intelligence tool designed for doctors, OpenEvidence, has seen viral adoption, with a significant portion of U.S. physicians using it for complex medical questions and diagnostic dialogues. Nicholas Gavin, an emergency physician at Mount Sinai in New York City, noted that younger colleagues quickly turned to the AI app during a challenging patient case.

Executives at the Mount Sinai health system discovered that one-third of their 9,000 physicians were already regular users of OpenEvidence. This rapid uptake has surprised healthcare leaders, highlighting the tool's growing influence. The application, essentially a chatbot for medicine, has become a go-to resource for doctors seeking quick answers or brainstorming diagnostic possibilities.

OpenEvidence's user base has expanded dramatically, with over half of the nation's doctors reportedly using it regularly. The startup claims that last month alone, doctors used the app for 30 million queries, nearly double the volume from six months prior. A separate survey indicated that 45% of doctors use OpenEvidence, significantly more than those using ChatGPT.

This rapid growth has propelled the startup to a $12 billion valuation, a substantial increase from its $3.5 billion valuation last July. However, the swift integration of AI into medical decision-making, particularly for a tool introduced in 2024, has sparked concerns. Healthcare systems are grappling with complex issues of privacy, security, and patient trust, alongside the inherent limitations of AI technology in high-stakes medical scenarios.

Daniel Nadler, founder and CEO of OpenEvidence, emphasizes that the tool is not an oracle but a support system. He stresses the continued importance of human knowledge and expertise in the medical field. While AI has been explored in medicine for decades, recent advancements have made tools like OpenEvidence more effective. The first wave of AI in medicine focused on reducing documentation burdens, but the current wave aims to assist physicians with reliable information for diagnosis and treatment.