Insider Brief
- General Intuition raised $320 million in Series A funding at a $2.3 billion valuation to expand its work on AI models designed to train agents that can act across space and time.
- The round was led by Khosla Ventures, with participation from General Catalyst, Hillspire and Jeff Bezos, according to the company.
- General Intuition trains large action foundation models on action-labeled gameplay clips from Medal’s 17 million monthly active users and is developing world models to generate training environments for AI agents.
General Intuition announced it has raised $320 million in a Series A funding round that values the artificial intelligence startup at $2.3 billion, as the company expands development of AI models designed to train autonomous agents that can act in digital and physical environments.
The round was led by Khosla Ventures and included participation from General Catalyst, Eric Schmidt’s Hillspire and Jeff Bezos, according to the company.
General Intuition describes itself as a frontier AI lab focused on developing systems that can act across space and time rather than simply generate text or images. The company is building large action foundation models trained on billions of action-labeled gameplay clips collected through Medal, its gaming video platform.
The startup said its training data now includes billions of gameplay clips generated by roughly 17 million monthly active Medal users. The first-person gameplay footage is used to teach AI systems spatial and temporal reasoning, which is understanding how actions unfold over time and how they affect changing environments.
The company said that over the past year it has focused on advancing two core technologies: action models that determine what an AI agent should do next and world models that simulate the likely results of those actions. General Intuition said its gaming roots shape its approach, with the goal of developing tools that work alongside game developers and other creators rather than replacing them.
Rather than commercializing those world models directly, the company uses them to create virtually unlimited training environments for its action models.
