A $2.3 Billion Bet: Can Fortnite and Other Video Games Train AI Robots for Our World?

Imagine walking into a tech office and seeing someone intensely playing Fortnite for hours on end. Now imagine that "someone" is actually an advanced artificial intelligence. That is exactly what visitors experienced at General Intuition, a startup that just secured a massive $320 million investment, boosting its valuation to $2.3 billion.

The real surprise is what happens next. The same AI brain powering the virtual Fortnite player is also driving a large, four-legged robot roaming the office. This robot, relying only on a single camera for sight, explores its surroundings much like a curious toddler, occasionally bumping into chairs or trash cans as it learns its physical space. This quick adaptation is remarkable, especially since the AI model only needed eight minutes of outdoor real-world data to fine-tune its navigation skills for indoor spaces.

General Intuition’s core idea is that AI can learn how to move and interact with the world by studying human gameplay. The company believes this method can give AI a deeper understanding of cause and effect, much like human intuition. This innovative approach has clearly convinced some big names, including Khosla Ventures, Jeff Bezos, and Eric Schmidt, who are all part of this new funding round.

The company was spun out of Medal, a platform where gamers upload and share video clips of their gameplay. Medal’s vast collection of millions of hours of gameplay provided the initial training ground for General Intuition’s AI. But the real secret sauce is not just the video itself, it is the hidden "action labels" in these clips that record every button press and movement a player makes.

Unlike other AI methods that try to guess actions from video, General Intuition uses these precise action labels to teach its AI. This detailed data helps the AI understand what it means to be an "agent" in an "environment," giving it a richer sense of how things work. This method helps the AI learn that walls are solid, ladders are for climbing, and shadows change with the sun, all from virtual experiences.

Why should this matter to you? If this technology scales as General Intuition hopes, it could drastically speed up the development of capable, versatile robots for countless applications. Think of robots that can perform complex tasks in factories, assist in search and rescue operations, or even help with everyday chores with a more natural understanding of their surroundings. This approach could also make developing future AI applications much easier and more accessible.

However, the company’s vision also carries a strong ethical commitment. Founder Pim de Witte, who has a background in humanitarian work, has drawn a clear line: General Intuition’s technology will not be used for military purposes or anything that could harm humans. He is also thinking about how AI might impact jobs, launching a platform called Nerve to help gamers earn income through data labeling and robot control.

What happens next for General Intuition involves scaling up its computing power and making its AI available to other companies. They aim to be a foundational AI model provider, making it easier for others to build innovations like self-driving cars, rather than building those products themselves. The big question remains whether this learning from games and simulations can truly hold up in the unpredictable real world on a large scale.

Do you think training AI on video games will lead to smarter, more intuitive robots than traditional methods?

With more powerful AI emerging, how important is it for companies to clearly state their ethical boundaries for its use?

---

Filed under: AI, Robotics, GamingAI, TechFunding, GeneralIntuition