Google's Med-PaLM 2: Testing AI LLM in Mayo Clinic

In the high-stakes arena of healthcare technology, Google is making its move with Med-PaLM 2, an AI chat technology currently being tested at the Mayo Clinic and other hospitals. This innovative tool, a derivative of Google's PaLM 2 large language model, is designed to answer health-related questions, summarize documents, and organize research data, leveraging its training on medical licensing exams and expert demonstrations.

The technology has shown promise, demonstrating alignment with medical consensus and reasoning ability. In some instances, it even generated answers preferred by respondents over physician-generated responses. However, it's not without its flaws, as it has exhibited the same accuracy issues seen in other AI chat models.

Google's foray into medical AI is not without competition. Microsoft, in partnership with healthcare software company Epic, is also developing a similar technology based on OpenAI's ChatGPT. Both tech giants are committed to maintaining patient confidentiality, assuring that their models are not trained on patient data.

While Google sees potential for Med-PaLM 2 to provide tremendous value in regions with limited access to doctors, the company acknowledges that the technology is still in its infancy. Google's senior research director, Greg Corrado, candidly expressed his reservations about the technology's current state, stating, "I don’t feel that this kind of technology is yet at a place where I would want it in my family’s healthcare journey." Yet, he remains optimistic about the future, asserting that the tech could expand the beneficial applications of AI in healthcare tenfold.

As the battle for AI supremacy in healthcare continues, it's clear that the industry is on the brink of a revolution. Whether Google's Med-PaLM 2 will lead the charge or follow in the footsteps of others remains to be seen. Stay tuned for more updates on this exciting frontier.