Google DeepMind Director Says Games Are a Core Pillar of AI Research at NDC26
The lecture frames video games not as a side application of AI but as the primary experimental field that has produced breakthroughs like AlphaFold, reinforcing the symbiotic relationship between game development and general artificial intelligence research.
Reporting from 1 sources: 4Gamer.net.
At Nexon's NDC26 conference, Google DeepMind Inception Director Alexandre Moufarek detailed how video games have been central to the company's AI research, from Atari and StarCraft to AlphaGo and the Gemini project. He emphasized that games provide safe, complex testbeds for training AI agents and that many DeepMind founders, including CEO Demis Hassabis, are former game developers whose iterative design experience informs current research.
Alexandre Moufarek, Director of the Inception team at Google DeepMind, delivered a lecture at Nexon's NDC26 conference titled "How Google DeepMind is Advancing AI Research with Video Games." He stated that games have "always been a core pillar" in DeepMind's research program. Moufarek, who joined DeepMind about 10 years ago after working on AAA titles at Ubisoft and interactive robotics at SoftBank, explained that the company has used game simulations from Atari to StarCraft to train AI agents. He noted that algorithmic breakthroughs from these game studies led to scientific discoveries such as the protein structure prediction AlphaFold.
Moufarek highlighted that many DeepMind members, including CEO Demis Hassabis, are former game developers. Hassabis previously worked on deep simulation games like "Theme Park" and "Black & White," where AI was central to gameplay. The iterative process of creating fun and engaging experiences in game development, Moufarek argued, closely mirrors the research workflow of forming hypotheses, designing experiments, and pursuing breakthroughs. He defined AI agents as "intelligent entities that autonomously perceive their environment and take actions to achieve a goal," and described how DeepMind's projects-from Atari general algorithms to AlphaGo, AlphaStar in StarCraft, and the SIMA project for language-following agents in 3D worlds-have used games as research opportunities.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.