Is the Future of AI in Video Games: General Intuition Proposes a New Approach

Revolutionary changes continue in the field of Artificial Intelligence (AI). While large language models like ChatGPT and Claude are showing high results in text processing today, experts note their limitations in understanding the physical world. New York-based startup General Intuition is proposing the use of video game data to bridge this gap. This is reported by Techcrunch.com reports .

Supported by Jeff Bezos, this company recently raised $320 million, bringing its total valuation to $2.3 billion. As reported on TechCrunch's Equity podcast, Coatue, Eric Schmidt, and researchers from MIT and Google DeepMind are among the project's key investors. This indicates high confidence from industry giants in game data.

Why is internet data not enough?

General Intuition CEO Pim de Witte explains that traditional language models rely on internet text. However, to achieve Artificial General Intelligence (AGI), a system must understand how objects move in time and space. Static internet data falls short in teaching physical world laws, whereas video games serve as ready-made simulations for this purpose.

The physics laws, object interactions, and complex visual scenarios within video games are the best training grounds for AI. Spun off from the Medal TV platform, General Intuition aims to use this database to create "world models." These models are expected to provide unprecedented opportunities for robotics and autonomous systems.

Ethics and defense issues

New technologies always bring unique risks and ethical questions. In his interview, Pim de Witte did not rule out that these models could be used for defense and military purposes in the future. This places the task of setting strict ethical boundaries before the company. Intelligence formed on the basis of game data could influence not only virtual but also real-life strategic decisions.

For countries on the path of technological development, such as Uzbekistan, such innovations are of great importance. For local developers and AI specialists, moving beyond text-only models to study multimodal and simulation-based systems will ensure future competitiveness. The experience of General Intuition shows that the most powerful AI systems of the future are not raised in libraries, but in virtual worlds.

In conclusion, video games are no longer just entertainment, but are becoming a source of data to solve humanity's most complex technological problems. If General Intuition successfully implements its plan, we may soon witness robots that perceive and understand the physical world just like humans.