General Intuition Hits $6B Valuation in Push for Embodied AI Foundation Models
General Intuition is developing foundation models that teach AI agents to navigate physical space and time, securing a $6B valuation. The startup aims to bridge the gap between digital intelligence and embodied action.
The boundary between digital intelligence and physical movement is rapidly dissolving. General Intuition, a startup focused on building foundation models for embodied AI, is reportedly in talks to raise capital at a $6 billion pre-money valuation. Backed by heavyweights like Valor and Point72, the company is tackling one of the hardest problems in modern engineering: teaching AI agents how to move through space and time with the same fluidity that LLMs navigate language.
Unlike traditional robotics software, which often relies on rigid, pre-programmed logic for specific tasks, General Intuition is pursuing a generalized approach. By developing foundation models specifically for physical interactions, they are creating a brain that can be ported across various robotic form factors, from factory arms to humanoid assistants. This "world model" approach allows AI to predict physical outcomes and adjust movements in real-time, a prerequisite for Physical AI to operate in unstructured human environments.
The influx of capital into General Intuition signals a shift in the AI landscape. While the last two years were dominated by generative text and image models, the next frontier is "Physical AI"—the application of massive compute to the tactile world. As these models mature, we expect to see a drastic reduction in the time required to train robots for new, complex manual labor tasks.
Source: TechCrunch
