General Intuition raises $320M for game-trained AI agents
General Intuition raised $320M at a $2.3B valuation to train agentic AI models using video game data.
Video games may become more than entertainment data. General Intuition is betting that gameplay can help train AI agents that eventually understand and act in the physical world.
What happened
General Intuition raised $320M at a $2.3B valuation, bringing total disclosed funding to $454M.
The company trains agentic AI models using video game data, especially gameplay action labels, and wants to use that data to build agents that can generalise from games and simulations into the physical world.
Why it matters
This is one of the strongest AI agent and embodied-AI signals in this batch.
Real-world robotics data is expensive and hard to collect. Game data could offer a scalable shortcut for training models that understand action, environment and consequence.
The bigger picture
The next AI frontier may come from simulation-rich data. If games can help train more capable agents, the boundary between gaming, robotics and physical AI could become much more important.