Glossary · AI & ML
Embodied AI
In brief
Embodied AI is artificial intelligence that operates through and learns from a physical or simulated body. The category covers robots, virtual agents in physics simulators, and any AI whose inputs and outputs include sensorimotor experience rather than pure text.
The premise of embodied AI is that some forms of intelligence — physical reasoning, common-sense object understanding, motor planning — only emerge when an agent has to act in a world. A pure language model can describe how to fold a shirt; an embodied AI has to actually do it, with all the sensor noise, contact dynamics, and partial observability that implies.
In 2026, "embodied AI" is most often used as the umbrella for the humanoid robotics + VLA-model + foundation-model-for-robotics intersection. It also covers research in simulated agents (AI2-THOR, Habitat) where the embodiment is virtual.
Related terms
See it in the wild
Browse robots and brands using these techniques
Glossary entries are upstream. The catalog is where the implementations live.