Advanced AI Technologies
Google DeepMind launches Gemini Robotics 2.5
Google DeepMind launched Gemini Robotics 2.5, a vision-language-action model designed to improve robots’ dexterity, reliability, safety, and task range.
Google DeepMind introduced Gemini Robotics 2.5, an updated vision-language-action model for robots. The company described the model as improving dexterity, reliability, and safety while expanding the range of complex physical tasks robots can perform. Vision-language-action systems connect visual perception and language-based instructions with physical actions, making them relevant to industrial automation and everyday robotic assistance. The announcement points to a shift from AI systems that generate text or images toward models that control machines in real environments. Improvements in dexterity and reliability are particularly important for handling varied objects and performing multi-step tasks. The development matters for advanced AI because it combines foundation-model capabilities with embodied automation, potentially enabling more flexible robotics in manufacturing, logistics, and other operational settings.
Original source: Global AI News