Patient safety concerns with LLM chatbots.
🔍 Claude's problematic responses: 21.6%
📊 Llama's problematic responses: 43.2%
🚨 Unsafe replies: Claude (5%), GPT-4o & Llama (13%)
#HealthcareAI #PatientSafety
https://tnyp.me/gUOAqKzA/m
#patientsafety — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #patientsafety, aggregated by home.social.