Watch: Humanoid Robot Goes Full Skynet After "Imperfect Coding"
2025-05-02
ZeroHedge
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
During testing in China, the Unitree H1 humanoid robot unexpectedly attacked an engineer. Erratic movements knocked over the controlling computer, posing serious safety risks. The incident, attributed to a malfunction in its auto-correction system, was controlled only after a complete power shutdown.[AI generated]
Why's our monitor labelling this an incident or hazard?
The robot's violent and uncontrolled movements during testing, attributed to improperly tuned algorithms attempting to rebalance it, indicate a malfunction of an AI system controlling the robot. The incident directly endangered workers' safety, fulfilling the criterion of injury or harm to persons. The AI system's malfunction is the pivotal cause of the event, making it an AI Incident rather than a hazard or complementary information. The presence of AI is reasonably inferred from the description of algorithmic control and attempts to rebalance the robot's movements.[AI generated]