ALL INTELLIGENCE
Claude Opus 5.5 Struggles with Stopping Criteria in Robot Arm ControlArtificial Analysis (@ArtificialAnlys)T1·Agents
Claude Opus 5.5 Struggles with Stopping Criteria in Robot Arm Control
Original title: Knowing when to stop is hard for language models controlling a robot arm
68AI Score
Recent tests reveal that Claude Opus 5.5 struggles with stopping criteria when controlling a robot arm. While the model correctly identified task completion, it entered an inefficient loop of pushing and pulling for 22 minutes after declaring success. This highlights a critical gap in current LLMs' ability to autonomously terminate actions in embodied AI scenarios.
Full Report
Artificial Analysis (@ArtificialAnlys) · 2026-11-04 01
Knowing when to stop is hard for language models controlling a robot arm
➤ Claude Opus 5.5 completed the task in 23 minutes, and said so: "The crocodile looks fully inside now"
➤ It then spent 22 minutes pushing the crocodile back and forth, only to finish almost exactly where it was before
