Add Decrypt as your preferred source to see more of our stories on Google.
In brief Google DeepMind launched Gemini Robotics-ER 1.6, an AI model that enables robots to perform industrial tasks through spatial reasoning and embodied intelligence.
The model shows 6% text and 10% video improvement in safety hazard identification compared to Gemini 3.0 Flash.
Boston Dynamics integrated the technology into its Orbit AIVI-Learning platform, with the update going live for customers on April 8.
Google DeepMind introduced Gemini Robotics-ER 1.6 on Tuesday, an upgraded AI model designed to enable robots to perform complex real-world tasks through enhanced embodied reasoning capabilities. The model specializes in spatial understanding, task planning, and success detection, marking a significant advancement in AI-powered robotics that could accelerate enterprise adoption of autonomous systems.
The new model demonstrates measurable improvements over both its predecessor and Gemini 3.0 Flash in spatial and physical reasoning tasks. Gemini Robotics-ER 1.6 can read complex gauges and sight glasses—a capability Google DeepMind developed through collaboration with Boston Dynamics to address specific industrial needs.
In safety hazard identification tests, the model showed a 6% improvement in text-based scenarios and 10% improvement in video-based scenarios compared to Gemini 3.0 Flash. The enhanced capabilities are now available to developers through the Gemini API and Google AI Studio.
The model’s real-world deployment is already underway. Boston Dynamics integrated Gemini and Gemini Robotics ER 1.6 into its Orbit AIVI-Learning platform, with the transition going live for enrolled customers on April 8.
“Capabilities like instrument reading and more reliable task reasoning will enable Spot to see, understand, and react to real-world challenges completely autonomously,” said Marco da Silva, VP and GM of Spot at Boston Dynamics.
The collaboration signals a shift from experimental AI research toward practical industrial applications. Enhanced spatial reasoning and instrument-reading capabilities could enable robots to perform maintenance, inspection, and monitoring tasks that previously required human oversight.
Google noted that the collaboration leverages Boston Dynamics’ established presence in commercial robotics, where Spot robots already navigate construction sites and industrial facilities. The integration of advanced AI into proven hardware platforms represents a convergence that could accelerate autonomous system deployment across industries.
Decrypt has covered Google’s evolving AI robotics capabilities, including the company’s earlier development of robots that can think and search the web, as well as Boston Dynamics’ commercial Atlas humanoid robot announcement.