By Alexandre Moufarek and Adrian Bolton

From Atari to Go to StarCraft, games have driven some of the biggest breakthroughs in AI. Google DeepMind is now 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 the organization's biggest AI breakthroughs, from mastering Atari to helping solve protein structure prediction — and they are still at the heart of what the team does.

Gaming is in Google DeepMind's (GDM's) DNA. Demis Hassabis, one of Google DeepMind's founders, is himself a former game developer — he got his start at Bullfrog Productions working on the 1994 hit Theme Park before founding his own studio, Elixir Studios — as are many members of the GDM team. Together, they have decades of hands-on experience in game development and a deep respect for the craft of making games.

Google DeepMind has always been clear that doing AI research with games requires deep partnership with game developers — like the major new research partnership with Fenris Creations and the EVE Universe unveiled earlier this year, and the work done together with acclaimed studios like Hello Games, Coffee Stain Studios, Foulball Hangover and others.

Games as the engine of AI research

The journey began when a small team trained a deep neural network to play Atari 2600 games directly from raw pixels. The Deep Q-Network (DQN) learned to play 49 different games — from Pong to Breakout to Space Invaders — without any game-specific engineering. The 2015 Nature paper on DQN helped catalyze the modern era of deep reinforcement learning.

From there, the team attempted to master more complex games, with each milestone producing more capable and general systems. AlphaGo defeated world champion Go player Lee Sae Dol in 2016 — a feat many experts thought was still a decade away. AlphaGo Zero surpassed every previous version by learning entirely from self-play, with no human data at all. AlphaZero generalized this approach to master chess, shogi and Go with one algorithm, while MuZero learned to play without even knowing the rules. In 2019, AlphaStar reached Grandmaster level in StarCraft II, navigating real-time complexity and imperfect information.

For each game, AI enriched the playing experience. AlphaGo's famous Move 37 — played in the second game of the 2016 match — was a play so unexpected that professional commentators initially thought it was a mistake, overturning centuries of received wisdom in Go and inspiring experts to explore new strategies. AlphaZero similarly inspired entirely new lines of play in chess. Crucially, the spirit of exploration that succeeded in games had profound impacts for other AI systems: AlphaFold applied these foundations to help solve the 50-year grand challenge of protein structure prediction, a breakthrough which was recognized with the 2024 Nobel Prize in Chemistry.

From mastering games to understanding them

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 rule books — which led the team to ask a fundamentally different question: can AI understand and interact with any game world the way a person would?

This is the challenge behind SIMA, the Scalable Instructable Multiworld Agent, first introduced by Google DeepMind in 2024. Rather than optimizing for a high score, SIMA is a generalist agent that "sees" what a player would see on screen, understands natural language instructions, and acts through ordinary keyboard and mouse controls — requiring no APIs or source code access.

Powered by Gemini, Google's frontier AI models, SIMA 2 acts as an interactive companion capable of real-time reasoning and conversation. It achieves human-like play across complex 3D research environments and video games including No Man's Sky, Valheim, Hydroneer, and more.

For game developers, a truly general gaming agent would unlock AI capabilities that work with existing games — no modifications to the game code required. This could power entirely new gameplay, from AI companions that genuinely understand the game world to Non-Player Characters (NPCs) that adapt and respond in ways that scripted systems never could.

A general gaming agent could also transform how games are made. During development, when the game changes with every commit, such agents could enable truly robust QA testing. Post-launch, when new content is introduced or players behave unpredictably, they could adapt in real time — generalizing to new situations without needing to be re-scripted.

To develop SIMA agents safely and responsibly, Google DeepMind has partnered with acclaimed game studios and is building a growing portfolio of games for AI research. This allows the team to challenge its agents with ever more complex tasks that may one day transfer to solving problems in the real world.

Exploring new frontiers of AI and games research, with game developers

Game studios bring expert craft, extraordinary game worlds and deep knowledge of their players. Google DeepMind brings frontier AI — from Gemini to research in generative interactive environments and embodied agents — research expertise, and its team's unique game development background and years of experience building AI for interactive environments. Together, the partners focus on discovering breakthrough experiences: never-before-seen gameplay that wouldn't be possible without AI. The partnership also arrives amid a broader industry shift, as studios and platform holders across gaming experiment with AI in areas from prototyping and QA tools to conversational NPCs.

The point is not the technology itself but fun experiences, so Google DeepMind takes a "show, don't tell" approach with its partners. The team works hand in hand with game developers, exploring new ideas and building playable prototypes to find the fun.

The latest research partnership, with Fenris Creations, the independent studio behind the EVE Universe, represents a new chapter in the history of AI research in games at Google DeepMind.

The EVE Universe: a research frontier

Fenris Creations — the studio behind EVE Online, formerly known as CCP Games — has spent more than two decades building one of the most extraordinary persistent worlds in gaming. EVE Online, launched in 2003, is a massively multiplayer space simulation where thousands of players share a single universe that has evolved continuously for more than 20 years. Its player-driven economy features real supply-and-demand dynamics and trade networks spanning thousands of star systems. Its landscape — shaped by alliances, conflicts and diplomacy — is driven by human interaction, and its player wars have produced some of the largest battles in online gaming history.

For AI research, this is a golden opportunity. It is a living, evolving world that demands precisely the capabilities Google DeepMind believes are essential for frontier AI:

• Continual learning: Acquiring new skills without forgetting what came before, in a constantly changing world.
• Memory: Accumulating and retrieving knowledge across timescales that extend far beyond today's model context windows.
• Long-horizon planning: Reasoning over weeks, months, or even years.
• Complex multi-agent dynamics: Navigating cooperation, competition, negotiation, economics, and emergent social behavior at scale.

These challenges sit at the core of Google DeepMind's broader research initiative to create systems that learn continuously from experience, with the rate of learning accelerating over time. The company believes these frontier capabilities could unlock new gameplay experiences in the future.

EVE Online was envisioned from day one as a sandbox of lasting consequences, shaped by its players. This has driven countless stories of human growth for players and employees across the decades. Together with Google DeepMind, we're pushing into uncharted territory where AI must learn, adapt and remember on timescales that no other game environment demands, while helping us understand how humans and AI can coexist in a virtual environment before we have to contend with the same questions in real life.

The research partnership extends across Fenris Creations' expanding universe, offering distinct environments for AI development. While EVE Online offers a large-scale single-shard persistent universe, EVE Vanguard is played from a first-person perspective, bringing ground-level, fast-paced tactical decision-making to the broader persistent world. This creates opportunities to study agents operating across multiple levels of abstraction, from twitch-level tactics to galaxy-spanning strategy.

Furthermore, EVE Frontier — with its programmable "Smart Assemblies" and its open, extensible architecture — offers an open-ended environment where the very rules of the world can change, demanding agents that adapt to entirely new game mechanics.

The collaboration has already delivered real player value: the Aura Guidance system uses Gemini to deliver player-generated knowledge, based on real Rookie Help questions and answers, to help new pilots.

The longer-term research program begins with an offline instance of EVE Online, which is a safe sandbox, separate from live players. It then progresses through EVE Frontier as a space to study how humans and agents can coexist in a persistent and open-ended world. Only when capabilities are mature would Google DeepMind consider bringing them to EVE Online and EVE Vanguard, with the aim of enriching human play.

The road ahead

Games have always been a mirror for intelligence. As they become more complex, more persistent, and more open-ended, so do the AI systems built to navigate them.

Google DeepMind is aiming for the same ultimate goal it always has: AI as a catalyst, not a replacement. Its long-term ambition is to unlock breakthrough gameplay experiences, make games more accessible and more personalized, and — as seen from AlphaGo to AlphaFold — apply what is learned in games to problems in the real world and advance scientific discovery. Per the partnership's stated roadmap, the milestones ahead are research results from the offline EVE Online instance, followed by human-agent coexistence experiments in EVE Frontier — with any live-game integration coming only once capabilities are mature.

The team thanked all of its game partners on this journey, and Fenris Creations for joining the next chapter, adding that it can't wait to share what it discovers.

Acknowledgements

Google DeepMind recognized the many teams across the organization for their contributions over the years to advancing AI research safely and responsibly in games, and offered special thanks to all of the game developers who partnered with the company: Coffee Stain (Valheim, Satisfactory, Goat Simulator 3), Fenris Creations (EVE Online, EVE Vanguard, EVE Frontier), Foulball Hangover (Hydroneer), Hello Games (No Man's Sky), Keen Software House (Space Engineers), RubberbandGames (Wobbly Life), Strange Loop Games (Eco), Thunderful Games (ASKA, The Gunk, Steamworld Build), Digixart (Road 96), and Tuxedo Labs & Saber Interactive (Teardown).

Source: Google DeepMind Blog — https://deepmind.google/blog/from-atari-to-eve-online-building-on-15-years-of-ai-research-in-games/