General Intuition, which applies foundation model training methodology to physical-world intelligence by training on hundreds of millions of hours of gaming footage as a proxy for movement and spatial reasoning, is reportedly in the final stages of closing a round led by Valor Equity Partners and Point72 Ventures at a six billion dollar pre-money valuation. The round follows a three-hundred-and-twenty million dollar financing closed at a 2.3 billion dollar valuation in June — a step-up of 2.6 times in fewer than three months that reflects both the specific enthusiasm for physical AI applications and the broader dynamic of capital concentrating rapidly around perceived category leaders.

The participation of Valor Equity Partners — its first AI lab bet since SpaceX — and a CoreWeave compute arrangement underpinning the company's training infrastructure signal the degree to which physical AI has become a distinct investment thesis rather than a subcategory of foundation model development. The gaming-data-to-robotics pipeline is, among the various approaches to physical AI training data, one of the more creative. Its viability at scale depends on how well the distributional properties of gameplay footage transfer to real-world physical environments — a question that the next twelve months of deployment will begin to answer seriously.