From Atari to EVE Online: Building on 15 Years of AI Research in Games

From Atari to EVE Online: Building on 15 Years of AI Research in Games Google DeepMind has outlined its 15-year trajectory of using video games to advance artificial intelligence, tracing a path from early reinforcement learning milestones to its current focus on generalist agents and an extensive new partnership with Fenris Creations in the EVE Universe. DeepMind's historical gaming milestones—such as DQN, AlphaGo, AlphaZero, MuZero, and AlphaStar—paved the way for cross-disciplinary applications like AlphaFold, which earned the 2024 Nobel Prize in Chemistry. The team has transitioned from task-specific score optimization to developing generalist, instruction-following agents like SIMA and the Gemini-powered SIMA 2, which interact with 3D games via keyboard and mouse controls without API access. A research partnership with Fenris Creations uses EVE Online, EVE Vanguard, and EVE Frontier to tackle frontier AI challenges including continual learning, extended memory, multi-agent dynamics, and long-horizon planning. Initial deployments from the collaboration include the Gemini-based Aura Guidance onboarding system, followed by planned staged research in offline sandboxes and open-ended environments before any live implementation. Historical Evolution of Game AI Starting in 2010, Google DeepMind utilized the constrained environments of games to advance AI. The 2015 Deep Q-Network (DQN) mastered 49 Atari 2600 titles from raw pixels without game-specific engineering, sparking the modern deep reinforcement learning era. Subsequent systems demonstrated expanding capabilities: AlphaGo beat Lee Sae Dol in 2016, AlphaGo Zero mastered Go via pure self-play, AlphaZero unified chess, shogi, and Go, MuZero operated without pre-programmed rules, and AlphaStar reached Grandmaster rank in StarCraft II. These foundations in exploration and optimization directly contributed to scientific breakthroughs, notably AlphaFold's protein structure predictions recognized with the 2024 Nobel Prize in Chemistry. Generalist Agents and SIMA Moving beyond rule-bound games with explicit scoring, DeepMind developed the Scalable Instructable Multiworld Agent (SIMA). Rather than optimizing scores through source code or APIs, SIMA parses visual screen inputs, interprets natural language commands, and interacts via keyboard and mouse. Powered by Gemini, SIMA 2 serves as an interactive companion providing real-time reasoning and conversation across complex 3D titles like No Man's Sky, Valheim, and Hydroneer. For developers, general agents offer potential applications in dynamic, unscripted NPCs and continuous quality-assurance testing that adapts to game code commits without manual re-scripting. Research Partnership Across the EVE Universe DeepMind has partnered with Fenris Creations, the studio behind the persistent, player-driven universe of EVE Online. The complex economy, diplomacy, and alliance systems of the 20-year-old title present key research challenges for frontier AI: continual learning without forgetting, memory retrieval across vast timescales, long-horizon planning over weeks or months, and large-scale multi-agent social dynamics. The research scope encompasses three titles: EVE Online for single-shard persistent strategy, EVE Vanguard for ground-level first-person tactics, and EVE Frontier for adapting to open-ended, programmable 'Smart Assemblies' and shifting mechanics. Staged Implementation and Player Impact The collaboration has already introduced the Gemini-powered Aura Guidance system, which uses real player help data to assist new players. For broader AI research, DeepMind uses a phased roadmap that begins in an offline, player-isolated sandbox instance of EVE Online. Research then moves into EVE Frontier to observe agent-human coexistence in open-ended settings, with plans to introduce mature agents to live instances of EVE Online and EVE Vanguard only after safety and effectiveness are established to enrich human gameplay.

从雅达利到EVE Online：Google DeepMind总结15年游戏AI研究并布局复杂虚拟世界 Google DeepMind发布博文回顾了团队15年来的游戏AI研究历程，强调游戏是推动智能边界的核心动力。随着技术从针对特定规则的通关算法转向理解开放世界的通用智能体，DeepMind正深化与游戏开发者的合作，包括与Fenris Creations就EVE宇宙展开深度研究。 游戏AI研究历经雅达利DQN、AlphaGo系列、MuZero、AlphaStar，并延伸出获得2024年诺贝尔化学奖的AlphaFold。 研究方向从追求高分转向通用智能体，基于Gemini的SIMA 2能通过键鼠控制和屏幕视觉直接理解多款3D游戏与自然语言指令。 通用游戏智能体不仅可用于智能NPC和游戏伴侣，还可在开发阶段革新QA测试流程并自适应版本更新。 DeepMind与Fenris Creations建立重大研究合作，利用EVE宇宙探索持续学习、超长跨度规划、长期记忆与复杂多智能体动力学。 双方已落地基于Gemini的新手指导系统Aura Guidance，并规划从离线沙盒、EVE Frontier逐步过渡到在线环境的负责任部署路径。 游戏作为AI技术突破的引擎 DeepMind的游戏AI探索始于通过原始像素自学49款雅达利2600游戏的DQN网络，该成果开启了现代深度强化学习。此后，AlphaGo系列击败围棋世界冠军、实现完全零人类数据自学并泛化到国际象棋与将棋；MuZero在未知规则下学会下棋；AlphaStar在即时战略游戏《星际争霸II》中达到特级大师级别。这些算法在博弈中展现的探索能力直接反哺了科学领域，促成了AlphaFold攻克50年蛋白质结构预测难题并获得2024年诺贝尔化学奖。 从精通规则转向通识理解：SIMA智能体 现实世界缺乏明确的得分和规则手册，DeepMind为此研发了可扩展可指令多世界智能体（SIMA）。SIMA不依赖专门的API或源码，直接像人类玩家一样接收屏幕视觉输入，通过常规键盘和鼠标执行自然语言指令。由Gemini大模型驱动的SIMA 2已能在《无人深空》《英灵神殿》《水力工匠》等复杂3D游戏环境中提供人类水准的操作、实时推理与对话陪伴。这类通用智能体可无缝赋能既有游戏，为自适应NPC和游戏伴侣提供技术支持，同时革新开发流程中的自适应QA测试。 携手Fenris Creations探索EVE宇宙 DeepMind与独立工作室Fenris Creations达成最新研究合作。拥有20多年历史的《EVE Online》包含由玩家驱动的复杂经济网络与外交冲突，为AI研究提供了探索前沿能力的理想环境。该合作聚焦于持续学习、超越当前模型上下文窗口的记忆积累、跨越数周至数年的长周期规划，以及大规模多智能体协作与对抗。此外，合作还涵盖提供快节奏战术视角的《EVE Vanguard》，以及规则开放可编程的《EVE Frontier》。 技术落地与渐进式研究路线 双方合作目前已落地应用：借助Gemini驱动的Aura Guidance系统，根据新手频道的真实问答数据为新飞行员提供解答。针对高复杂度与风险，DeepMind制定了渐进式的研究计划：首先在与在线玩家隔离的离线版《EVE Online》沙盒中验证模型，随后在《EVE Frontier》中探索人机共存与适应机制，只有在技术充分成熟后，才会考虑将智能体引入《EVE Online》和《EVE Vanguard》的实际运行环境中。

August 21, 2026 Research Alexandre Moufarek and Adrian Bolton 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. We’ve always been clear that doing AI research with games requires deep partnership with game developers - like our major new research partnership with Fenris Creations and the EVE Universe that we unveiled earlier this year, and the work we’ve done together with acclaimed studios like Hello Games , Coffee Stain Studios , Foulball Hangover and others. Games as the engine of AI research Our 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, we 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 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 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 rule books — which led us 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 , our Scalable Instructable Multiworld Agent. 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, our 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 — generalising to new situations without needing to be re-scripted. To develop SIMA agents safely and responsibly, we've partnered with acclaimed game studios and we are building a growing portfolio of games for AI research. This allows us to challenge our 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, in partnership with game developers Game studios bring expert craft, extraordinary game worlds, and deep knowledge of their players. We bring frontier AI — from Gemini to research in generative interactive environments and embodied agents — research expertise, and our team’s unique game development background and years of experience building AI for interactive environments. Together, we focus on discovering breakthrough experiences — never-before-seen gameplay that wouldn't be possible without AI. It's not about the tech, it's about fun experiences. So we take a ‘show, don't tell’ approach with our partners. Our team works hand in hand with game developers, exploring new ideas and building playable prototypes to find the fun. Our latest research partnership with Fenris Creations , the independent studio behind the EVE Universe, represents a new chapter in our history of AI research in games. Fenris Creations 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. For AI research, this is a golden opportunity. It's a living, evolving world that demands precisely the capabilities we believe 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 our broader research initiative to create systems that learn continuously from experience, with the rate of learning accelerating over time. We believe these frontier capabilities could unlock new gameplay experiences in the future. Our 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. A “show, don’t tell” session during a workshop with Google DeepMind and Fenris Creations where our teams share playable prototypes, feedback and discuss ideas for new gameplay experiences Our 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. Our 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 we 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 we build to navigate them. We are aiming for the same ultimate goal we always have: AI as a catalyst, not a replacement. Our long-term ambition is to unlock breakthrough gameplay experiences, make games more accessible and more personalized, and — as we've seen from AlphaGo to AlphaFold — apply what we learn in games to problems in the real-world and to advance scientific discovery. We're grateful to all of our game partners on this journey, and to Fenris Creations for joining us on this next chapter. We can't wait to share what we discover. Acknowledgements We’d like to recognize the many teams across Google DeepMind for their contributions over the years to advancing AI research safely and responsibly in games. Special thanks to all of the game developers who partnered with us: 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).