Details
The Orbit platform uses vision-language models and cloud-hosted machine learning models to process images taken by the Spot robot during autonomous facility patrols. The system can answer queries such as yes/no questions, numeric readings, or descriptive assessments, and flags issues like spills, missing fire extinguishers, pallet status, and equipment wear. As of April 2026, Boston Dynamics integrated Google DeepMind's Gemini Robotics-ER 1.6 model into the AIVI and AIVI-Learning systems, enabling more complex spatial reasoning and instrument reading.
Have evidence about Boston Dynamics's AI practices? Submit a report.
Report a Sighting →