Google DeepMind and Fenris Creations Explore Long-Horizon AI in Games

From Atari to Go to StarCraft, games have driven some of the biggest breakthroughs in AI. Now, we’re partnering with game developers to prototype new gameplay experiences that push the frontiers of both gaming and AI. Since DeepMind’s foundation in 2010, the constrained yet rich worlds of games have played a critical role in understanding intelligence. They have driven some of our biggest AI breakthroughs, from mastering Atari to helping solve protein structure prediction - and they are still at the heart of what we do. Gaming is in GDM’s DNA. Demis Hassabis, one of Google DeepMind's founders, is himself a former game developer, as are many of us in the GDM team. Together, we have decades of hands-on experience in game development and a deep respect for the craft of making games. [...] ## From mastering games to understanding them Our earlier work demonstrated that AI could master any game given a clear objective and enough training. But the real world doesn't come with scores and r

Alexandre and Piermaria present new research on SIMA (Scalable Instructable Multiworld Agent), an AI agent that follows natural-language instructions to perform tasks across diverse video game environments. Video games are a key proving ground for artificial intelligence (AI) systems. They offer rich learning environments, mirroring real-world complexities with responsive, real-time settings and dynamic objectives. From our early work with Atari games to AlphaStar, which achieved human-grandmaster level in StarCraft II, Google DeepMind has a long history in AI and games. SIMA represents a significant advancement, shifting focus from individual games to a general, instructable game-playing AI. This generalist agent for 3D virtual settings was trained in partnership with game developers [...] settings was trained in partnership with game developers across a diverse range of video games. This marks the first instance of an agent demonstrating comprehension of a broad spectrum of gaming wo

Games have been an important testbed for our AI research for over 15 years. 🎮 From mastering Atari to reaching Grandmaster in StarCraft II, they have driven some of our biggest AI breakthroughs. Our work with SIMA taught agents how to understand 3D worlds, but learning to navigate real human dynamics takes a living, persistent universe. Through our research partnership with Fenris Creations, we’re exploring how to tackle open challenges in AI: 🔵 Continual learning to acquire new skills without forgetting past knowledge. 🔵 Deep memory systems that store and retrieve information far beyond today’s context windows. 🔵 Long-horizon planning over weeks, months, or years. 🔵 Multi-agent dynamics spanning cooperation, negotiation, economics, and emergent behaviors. [...] Games have been an important testbed for our AI research for over 15 years. 🎮 From mastering Atari to reaching Grandmaster in StarCraft II, they have driven some of our biggest AI breakthroughs. Our work with SIMA taught

Games serve as uniquely effective testbeds for advancing AI capabilities like continual learning, deep memory systems, long-horizon planning,

@GoogleDeepMind Google DeepMind @GoogleDeepMind [...] Games have been an important testbed for our AI research for over 15 years. 🎮 From mastering Atari to reaching Grandmaster in StarCraft II, they have driven some of our biggest AI breakthroughs. Our work with SIMA taught agents how to understand 3D worlds, but learning to navigate real human dynamics takes a living, persistent universe. Through our research partnership with @FenrisCreations, we’re exploring how to tackle open challenges in AI: 🔵 Continual learning to acquire new skills without forgetting past knowledge. 🔵 Deep memory systems that store and retrieve information far beyond today’s context windows. 🔵 Long-horizon planning over weeks, months, or years. 🔵 Multi-agent dynamics spanning cooperation, negotiation, economics, and emergent behaviors. Our long-term goal is to use [...] Games have been an important testbed for our AI research for over 15 years. 🎮 From mastering Atari to reaching Grandmaster in StarCraft

Games have been an important testbed for our AI research for over 15 years. 🎮 From mastering Atari to reaching Grandmaster in StarCraft II,