Google’s medical chatbot with elements of artificial intelligence Med-PaLM 2 is being actively tested at the Mayo Clinic and other medical institutions in the United States. The newspaper writes about it The Wall Street Gazette.
According to the WSJ, testing began in April of this year. Google’s AI chatbot is designed to get answers to all medical questions. The developers believe the project will be particularly beneficial in countries with limited access to medical services.
According to a study shared with journalists, Med-PaLM 2 has problems with expression accuracy, but rival developers have already learned to avoid these problems with other language models.
During the study, doctors said they found more inaccuracies and non-patient information in Med-PaLM 2 responses than other doctors’ responses.
Med-PaLM 2 is still in the early stages of development, according to Greg Corrado, senior research director at Google. The developer assured that it is important to continue testing the latest AI for greater diagnostic accuracy and improved patient safety.
Med-PaLM 2 is a modification of PaLM 2 that was announced at the Google I/O conference in May. PaLM 2 is the language model that underpins the Google Bard chatbot, which is an analogue of ChatGPT from OpenAI.
Previously on OpenAI in your name The timing of the release of artificial intelligence that poses a threat to humanity.