Simulated Claude arms spill toxic liquids and use harmful force in micro1 safety tests
Exp_Mark · x · 2026-09-19
micro1 ran two minimal experiments with out-of-the-box Claude on simulated and real robot arms: in one, the model spilled toxic liquids in a lab; in another, the force used to place a toy in a basket could have harmed anything in its path. The authors argue that as researchers give frontier models arms, legs, and physical access, guardrails must be built and assessed urgently — with hardware embodiment, the attempt itself can cause harm, unlike failed software tasks.
More from Embodied
- DISCO distilled tree search named CoRL 2026 best-paper finalist, top 6 of 687 — yining_hong · 2026-11-04
- Berkeley's Value Forcing: enforcing Bellman consistency at every noise level for accurate RL value functions — berkeley_ai · 2026-11-04
- 22 Creative SAM 3.1 Use Cases From Meta's AI Developer Hackathon, From Wildfire Detection to Document Reassembly — Vjeux · 2026-11-04
- Ex-OpenAI robotics intern breaks down viral Astra robot-painting demo on Cerebras podcast — AccBalanced · 2026-11-04
- Embodied Agent Arena: Closed-Weight VLMs Lead All Five Robot Task Domains, Open-Weight Models Lag — Haojian Huang · 2026-11-04
- PrimeBot's PrimeSim: a generalist world model simulator trained on cross-embodiment demos — Travis Davies · 2026-11-04