Physical AI Safety Consortium launched by AIM Intelligence

AIM Intelligence has launched the Physical AI Safety Consortium (PASC), a global non-profit research group dedicated to establishing safety standards for AI that interacts with the physical world. Unlike traditional chatbots that focus on filtering harmful text, PASC addresses the risks posed by robots, autonomous systems, and industrial equipment whose actions can directly impact people and physical environments.

The consortium includes approximately 30 experts from major technology companies such as Google, Microsoft, Oracle, and LG Electronics, alongside researchers from academic institutions including UC San Diego, the University of Wisconsin, and the University of Illinois.

PASC’s initial objectives include publishing a position paper to define physical AI safety and establishing evaluation frameworks and benchmarks. Future research will expand to include adversarial testing in simulated and real-world robotic environments. The group aims to develop systems that can continuously monitor, detect signs of danger, and allow for human intervention, ensuring safety remains consistent even as robot forms or operational environments change.