Why Free-Form LLM Agents Fail at Anomaly Discovery

We often hear that autonomous AI agents will soon discover new scientific laws, debug distributed systems zero-shot, and autonomously resolve infrastructure incidents. A new rigorous benchmark, LongChang, put this claim to the test by placing 12 state-of-the-art LLMs inside simulated environments with altered physical laws without hints or... Weiterlesen: Why Free-Form LLM Agents Fail at Anomaly Discovery