How LabOS AI-powered smart goggles could reduce human error in science

LabOS is an AI operating system, developed by the Stanford-Princeton AI Coscientist Team with NVIDIA, that utilizes smart goggles to monitor laboratory work and reduce human error, which contributes to the scientific replication crisis. The system streams video from AR/XR glasses to an AI powered by vision-language models, allowing it to compare actions against experimental protocols and provide real-time guidance, warnings, or technique correction to the wearer. By recording entire experiments, LabOS can rapidly learn what causes failures, enabling faster iteration than traditional human learning. In pilot tests, junior scientists with minimal LabOS training achieved results comparable to experts in complex procedures. The team also uses this architecture for MedOS to assist surgeons, with the ultimate goal of making all research labs and clinics AI-perceivable and AI-operable environments to improve outcomes and training.

(Source：Scientific American)