General Intuition: $320m Series A to Train AI Agents on Gameplay Data

Views: 0

US-based AI gaming company General Intuition raised $320m in a Series A round at a post-money valuation of $2.3B. The round was led by Khosla Ventures, doubling down on its Seed position, alongside returning backer General Catalyst. Additional investors include Jeff Bezos, Eric Schmidt, Nico Rosberg, FuturePresent, Hexagon, Coalition, NPHard, co-founders of Remote and Cradle, and researchers from Google DeepMind and MIT. The round closed in Jan’26, three months after the Seed round, but was publicly announced on Jun 25, 2026. Proceeds will fund compute scaling through a partnership with CoreWeave, with a portion earmarked to broaden commercial API access by the end of summer 2026. Total disclosed funding stands at $454m.

General Intuition was co-founded in 2025 by Pim de Witte, creator of gaming clip platform Medal, and researchers Eloi Alonso, Adam Jelley, and Vincent Micheli. Their prior work includes the DIAMOND diffusion world model, IRIS and Δ-IRIS world models, and GAIA-2 generative models for automated driving. The company focuses on action models, systems that generate optimal actions given observations of an environment rather than predicting how environments evolve, targeting gaming, simulation, and physical robotics. Training data flows from Medal, which operates under the same parent and whose 17 million+ monthly active users generate billions of action-labeled gameplay clips. A commercial API for gaming, simulation, and robotics partners has launched, with broader access planned by the end of summer 2026.

The $133.7m Seed round in Oct’25, which we covered at InvestGame, was led by Khosla Ventures and General Catalyst, with participation from The Raine Group, and ranked among the largest early-stage investments in AI gaming at the time of closing. The Series A followed three months later, based on research progress, with Khosla re-leading the round for the second consecutive time.

General Intuition’s raise completes a cluster of three $300m+ rounds in the world model and AI simulation category within months of each other, together representing the three largest AI and gaming infrastructure deals of the 2024-2026 period, alongside Decart’s $300m Series B at a ~$4B valuation and Odyssey‘s $310m Series B at $1.45B, both covered in recent digests. The three companies occupy distinct positions: Decart is focused on inference infrastructure and optimization across hardware platforms; Odyssey is focused on general world simulation with a physical AI and robotics angle; and General Intuition is focused on action models trained on proprietary gameplay data, with gaming as both a training ground and a commercial market. What distinguishes General Intuition is the data moat: while Decart and Odyssey rely on compute scale and research breadth, General Intuition’s access to billions of human gameplay decisions through Medal provides a training resource that is difficult to replicate.