AI in robotics gives machines the ability to see, plan and act in the physical world, increasingly using models that link vision and language to movement.
In plain words
Classic robots follow fixed programs in controlled settings. Learning-based robots use cameras and models to handle new objects and instructions. Vision-language-action models take what a model learned from web images and text and turn it into motor commands, so a robot can follow requests it was never explicitly trained on.
A quick example
A robot running Google DeepMind's RT-2 model was asked to find something to use in place of a hammer, and reasoned that a rock would do. That skill was not in its robot training data; it came from what the model learned from the web.
Where you'll meet it
- Robot arms that follow plain-language instructions.
- Research on humanoid and general-purpose robots.
Related
Sources
- DocsRT-2: New model translates vision and language into action, Google DeepMind · read 29 Sept 2026
- DocsGemini Robotics, Google DeepMind · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.