Your next doctor visit could involve Google‘s medical AI, Med-PaLM.
Google’s Med-PaLM 2, an AI tool designed to field medical inquiries, has been under trial at Mayo Clinic since April.
Med-PaLM 2, a variant of Google’s own language model PaLM 2, promises a revolutionary approach to healthcare, especially in areas with limited doctor access.
Trained on expert medical demonstrations, Med-PaLM 2 strives to outperform generalized LLMs like Bard, Bing, and ChatGPT in healthcare conversations.
Med-PaLM 2 harnesses the power of Google’s LLMs, aligned to the medical domain to more accurately and safely answer medical questions. As a result, Med-PaLM 2 was the first LLM to perform at an “expert” test-taker level performance on the MedQA dataset of US Medical Licensing Examination (USMLE)-style questions, reaching 85%+ accuracy, and it was the first AI system to reach a passing score on the MedMCQA dataset comprising Indian AIIMS and NEET medical examination questions, scoring 72.3%.
Med-PaLM 2 has shown similar competencies as actual doctors in providing reasoning and consensus-supported answers, despite occasional accuracy issues.
Customers testing Med-PaLM 2 maintain complete control over their data, which is encrypted and inaccessible to Google.
Google believes this AI can immensely benefit healthcare by about 10x.
What are your thoughts on AI helping doctors?
#AI #generativeai #llms #palm2
Data: WSJ, Google
