Physical Safety

Physical safety in AI governance covers the risk that an AI system causes bodily harm to people or damage to property, typically when it controls or influences machines, vehicles, medical devices, industrial equipment, or robots. For compliance and risk teams, this means that errors which would be harmless in a chatbot can become serious liability when the system is wired into the physical world. Regulators and insurers increasingly expect documented testing, human override controls, and incident reporting for AI systems deployed in these settings.

1 item

ResearchGlobal2026-11-10

Gemini Robotics 2.5 Brings Embodied AI Governance Into the Factory

Google DeepMind released Gemini Robotics 2.5 on November 10, 2026, its most capable model for controlling physical robots. A proof-of-concept with Apptronik and Mercedes-Benz on active automotive production lines achieved a 96% success rate. The release also introduces a new benchmark designed to test whether the model can be manipulated into unsafe physical actions.

Gemini Robotics 2.5 Google DeepMind embodied AI robotics AI physical safety jailbreak benchmarking manufacturing AI vendor risk