Google DeepMind releases Gemini Robotics 2, which combines several different AI models into a single system to control a range of robots, including humanoids (Will Knight/Wired)
DeepMind is pushing its robotics stack from tabletop skills toward whole-body control.
Gemini Robotics 2 is described as three models working together: a vision-language-action model, an embodied reasoning model, and an on-device model. Google says the system can handle humanoid movement, fine dexterity, multi-robot teamwork, and adaptation to new robot bodies within hours. The Techmeme note also flags the risk side of putting more capable AI into physical machines. Techmeme's note
Gemini Robotics 2 is described as three models working together: a vision-language-action model, an embodied reasoning model, and an on-device model. Google says the system can handle humanoid movement, fine dexterity, multi-robot teamwork, and adaptation to new robot bodies within hours. The Techmeme note also flags the risk side of putting more capable AI into physical machines. Techmeme's note
score 8