US-based AI startup General Intuition, co-founded by Dutch entrepreneur Pim de Witte, raises €114M in one of 2025’s largest AI investments.

US-based General Intuition, an AI gaming company co-founded by Dutch entrepreneur Pim de Witte, originally from Nijmegen, has secured $133.7M (approximately €114M) in a Seed round led by Khosla Ventures and General Catalyst.

The company, which is a spin-off of Dutch platform Medal, is strategically located in New York to remain close to leading capital markets, talent hubs, and the rapidly evolving U.S. technology ecosystem, according to FD.nl.

Contentlockr

This funding is one of the largest investments this year for a Dutch-born founder working overseas, adds the report.

The company will use the funds to accelerate the development of foundation models that require deep spatial and temporal reasoning, transforming AI for both gaming and real-world applications.

It’s worth mentioning now that, in early 2025, reports surfaced that OpenAI expressed interest in acquiring Medal for around $500M, though no transaction was finalised.

General Intuition: Developing AI with advanced spatial reasoning

General Intuition was founded as a spin-off of Medal, a Dutch platform that enables gamers to clip, edit, and share gameplay moments.

The company is developing an AI system with advanced spatial reasoning skills.

This means General Intuition is creating models that can understand space and time concepts, using large interactive video data ( which contains billions of user-generated gameplay clips collected each year) from its platform, Medal.

“When you play video games, you essentially transfer your perception, usually through a first-person view of the camera, to different environments,” Pim de Witte, CEO and co-founder of General Intuition, told TechCrunch. 

The company’s goal is to create systems that can see, predict, and adjust to different environments.

By using gameplay data for training, General Intuition looks at how virtual worlds can help develop AI systems that work well both in simulations and in real life.

According to De Witte, the company’s model can understand environments it wasn’t trained on and correctly predict actions within them.

“It’s able to do this purely through visual input; agents only see what a human player would see, and they move through space by following controller inputs. This approach, the company says, can transfer naturally to physical systems like robotic arms, drones, and autonomous vehicles, which are often manipulated by humans using video game controllers,” says De Witte.   

Using this dataset, the company is developing AI models focused on spatial and temporal reasoning within game environments. Its research spans:

  • Agentic systems with advanced spatial and temporal reasoning
  • World models as environments for AI training
  • Cross-domain video understanding with applications beyond gaming

“As a public-benefit corporation, we will not develop technology that replaces game developers, designers, or artists. Instead, we believe that we can drastically improve the gameplay across existing games—and create entirely new, previously impossible experiences,” says the company.

Moving forward, the company’s commercial focus includes AI-powered NPCs and interactive simulation tools, with deployment targeted for H1, 2026.