Knowing when to stop is hard: Claude Opus 5.5 pushed a crocodile for 22 extra minutes
ArtificialAnlys · x · 2026-11-04
Artificial Analysis found that knowing when to stop is a real weakness for LLMs controlling robot arms.
- Claude Opus 5.5 completed the task in 23 minutes and even said so: "The crocodile looks fully inside now"
- It then spent another 22 minutes pushing the crocodile back and forth, finishing almost exactly where it started
The test highlights how difficult task-completion state judgment remains for models in embodied control settings.
Related event: Robot-arm test finds most LLMs don't know when to stop(2 posts)→
More from Embodied
- Fruit fly's 166,000-neuron circuit map is being used to drive robot behavior — CurieuxExplorer · 2026-11-04
- Six AgiBot G2 humanoids run a Shanghai convenience store for 24 hours, logging 1,730 tasks — yogthos · 2026-11-04
- Corebrum OS: open-source Physical AI OS now installable on NVIDIA Jetson and any Ubuntu machine — chrismatthieu · 2026-11-04
- Artificial Analysis launches AA-Robotics benchmark: GPT-6 Astra tops robot arm control at $16.36 per task — ArtificialAnlys · 2026-11-04
- Waymo expands South Bay coverage with Berryessa and Milpitas BART connections — reed · 2026-11-04
- Diamandis: a household robot saving 10 hours a week — 520 hours a year — is the robot economy people care about — PeterDiamandis · 2026-11-04