News & Alerts
Breaking news & key AI updates in healthcare, medicine & pharma/biotech
Open-ended AI diagnosis lags multiple choice in brain scans
Large multimodal models jump from roughly 60% to 80% accuracy on neuroradiology cases when answer choices are provided.
More in News & Alerts
Review catalogs the mental health dangers of LLM chatbots
A scoping review maps five categories of mental health harms tied to LLM-based chatbots.
Chai Discovery scores BMS pact for AI-built antibodies
Chai Discovery adds Bristol Myers Squibb to its pharma roster for AI-built therapeutic antibodies.
Bioptimus folds UK colorectal data into AI tissue atlas
Bioptimus gains a 1,750-patient UK colorectal dataset to train its STELA tissue atlas foundation model.
Clinicians drift away from LLM decision support in the ER
A Nature Medicine study finds clinician engagement, not accuracy, is the biggest hurdle for LLM…
Chatbots dig up buried patient history for hospitals
Hospitals are putting AI chatbots to work on patient records, with Stanford's ChatEHR leading the…