An AI Startup That Learns From Video Games Is Raising Money at a $6 Billion Value

General Intuition is in talks to raise funding at a $6 billion valuation from new investors including Valor Equity Partners, Point72 Ventures, and Seven Seven Six, according to sources familiar with the matter (TechCrunch). The round is oversubscribed, meaning more investors want in than there is room for, and it is still being finalized. Existing investors Khosla Ventures and General Catalyst are also participating.

The New York-based startup, legally incorporated as General Intuition US Inc., is building an AI system that can learn to take actions in both virtual and physical environments. The company describes itself as "the frontier lab for acting in space and time" (General Intuition).

This latest round follows a fast climb. In June 2026, the company raised $320 million at a $2.3 billion valuation, bringing its total disclosed funding to $454 million at that time. Before that, it had raised a $134 million seed round to teach AI agents spatial reasoning using video game clips. TechCrunch's reporting on the current deal confirms earlier coverage by The Wall Street Journal.

The company's approach is unusual. General Intuition uses hundreds of millions of hours of gameplay footage along with "action labels" — records of exactly what buttons a player pressed and when — collected from Medal, a video game clip-sharing platform. CEO Pim de Witte spun General Intuition out of Medal in October 2025. Vinod Khosla told TechCrunch he believes action labels will be a key part of what he calls the "emergence of intuition," meaning the ability for an AI to handle tasks it was not specifically trained on.

The idea is similar to how a flight simulator teaches a pilot. A pilot can practice thousands of scenarios in a simulated cockpit before ever flying a real plane. General Intuition is betting that the decisions players make in video games can teach AI agents how to navigate and act in the real world, including in physical robots.

General Intuition plans to use the new funds to improve its model with a focus on robotics. The money will go toward more computing power and hiring. The company already has a partnership with CoreWeave for computing infrastructure.

The jump from a $2.3 billion valuation in June to a reported $6 billion in August reflects strong investor interest in AI that can work in physical spaces. The core bet is that data from gameplay — a detailed record of decisions made in fast-moving, three-dimensional environments — can train AI agents to act in the real world.

The participation of both early backers like Khosla Ventures and General Catalyst alongside new investors points to a shared strategic vision. Khosla's description of "emergence of intuition" suggests General Intuition is not just building a robotics tool but making a broader bet on AI that learns through action.

The broader context here is that embodied AI — AI that controls physical machines — has attracted heavy investment throughout 2026, with multiple labs working on training agents in simulations and then moving them to real robots. General Intuition's edge is the size and detail of its gameplay dataset, which captures not just what players see but what they do, moment by moment.

The deal, if it closes at the reported terms, would put General Intuition among the most highly valued private AI labs focused on action and embodiment. The speed of the valuation increase — roughly 2.6x in two months — also reflects a competitive funding environment for companies that can claim large, proprietary datasets.

There is an open question that matters here. Training AI in a video game environment and then expecting it to work well in the physical world is not guaranteed. Real environments are messier, less predictable, and full of situations no game can fully capture. General Intuition's bet is that enough of what an agent learns in a virtual world will carry over to make the approach work. If it does, the payoff could extend well beyond gaming into any field where machines need to understand and move through physical space.