Get Started
Home
Topics
Search
Library
Research questionHow can LLM agents make safe primary-care decisions as the action space grows?Clinical agents must select among multiple actions while respecting workflow and safety constraints. Diagnostic accuracy alone may miss unsafe decisions, especially in high-risk cases.
AI
AI Agents
Alignment & Safety
Evaluation & Benchmarks
Health
Reinforcement Learning
Research Paper
Latest papersRecent research connected to this question, newest first.GPAgentBench-2K: Benchmarking Large Language Model Agents in Complex Clinical Action SpaceThe evidence concerns a constrained benchmark built from expert-validated records of real-world general-practice encounters. It evaluates 16 LLMs across six clinical action types with a topological workflow prior and safety-informed abstention, and includes a constrained reinforcement-learning reference; the reported results show that higher diagnostic accuracy does not ensure safety.research paper · Sep 2, 2026
Related questions
How can long-horizon LLM agents learn when to group actions without overcommitting?How can LLM-agent systems prevent safety compromises from propagating across workflow boundaries?How can LLM agents stay safe during multi-step execution when both policy and runtime harness shape behavior?How can LLM agents reconcile conflicting user instructions, stored knowledge, and live tool observations before acting?