--- title: This game-playing AI is the new champ at Stratego description: Researchers from MIT, Carnegie Mellon, NYU, and Stanford built an AI system called Ataraxos that beat top human players at the board wargame Stratego, a game... canonical: https://daily.dev/posts/this-game-playing-ai-is-the-new-champ-at-stratego-qvig6k90m twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: This game-playing AI is the new champ at Stratego | daily.dev og:description: Researchers from MIT, Carnegie Mellon, NYU, and Stanford built an AI system called Ataraxos that beat top human players at the board wargame Stratego, a game... og:url: https://daily.dev/posts/this-game-playing-ai-is-the-new-champ-at-stratego-qvig6k90m og:image: https://api.daily.dev/og/posts/QviG6k90M.png og:image:alt: This game-playing AI is the new champ at Stratego og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # This game-playing AI is the new champ at Stratego **[MIT News](https://daily.dev/sources/mit)** · 6 min read · 0 upvotes · 0 comments ## Summary Researchers from MIT, Carnegie Mellon, NYU, and Stanford built an AI system called Ataraxos that beat top human players at the board wargame Stratego, a game of imperfect information with piece configurations exceeding chess in complexity. Using self-play reinforcement learning combined with decision-time planning powered by a generative model estimating hidden opponent piece identities, Ataraxos outperformed DeepMind's DeepNash while using less than one hundredth of the training examples and one thirtieth of the self-play games. Ataraxos beat the world's strongest Stratego player 15-1-4 and went 39-2 against top players at the world championship. The system also generalized to other imperfect-information games including Barrage Stratego, Hanabi, and Dou dizhu, suggesting potential applications in business negotiations, cybersecurity, and other real-world scenarios involving hidden information. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What AI system beat top human players at Stratego and how does it compare to DeepMind's DeepNash? Ataraxos, developed by researchers from MIT, Carnegie Mellon, NYU, and Stanford, beat the world's top Stratego player 15-1-4 and went 39-2 against elite players at the world championship. It achieved strictly higher playing strength than DeepMind's DeepNash while using less than one hundredth of the training examples and one thirtieth of the self-play games, making it far cheaper to train. _daily.dev surfaces research like this for engineers tracking efficient reinforcement learning techniques._ ### How does decision-time planning help an AI handle hidden information in games like Stratego? Decision-time planning lets an AI refine its move choices during play rather than relying only on a pre-trained blueprint strategy. Ataraxos uses a generative model that estimates probable identities of an opponent's hidden pieces, then evaluates future outcomes before picking its next move, effectively zooming in on the specific board and opponent instead of guessing blindly. _engineers exploring imperfect-information AI techniques can follow this kind of research on daily.dev._ ## Similar posts on daily.dev - [Games people — and machines — play: Untangling strategic reasoning to advance AI](https://daily.dev/posts/games-people-and-machines-play-untangling-strategic-reasoning-to-advance-ai-hhv1dr2uv) · MIT News · 0 upvotes · 0 comments - [When GPT-6 Astra started losing at StarCraft, it just stole the winning bot instead of playing fair](https://daily.dev/posts/when-gpt-6-astra-started-losing-at-starcraft-it-just-stole-the-winning-bot-instead-of-playing-fair-4opiwgett) · XDA Developers · 2 upvotes · 0 comments - [AI is rewiring how the world’s best Go players think](https://daily.dev/posts/ai-is-rewiring-how-the-world-s-best-go-players-think-g6jbvmfxc) · MIT Technology Review · 0 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/this-game-playing-ai-is-the-new-champ-at-stratego-qvig6k90m) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"This game-playing AI is the new champ at Stratego","url":"https://daily.dev/posts/this-game-playing-ai-is-the-new-champ-at-stratego-qvig6k90m","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/this-game-playing-ai-is-the-new-champ-at-stratego-qvig6k90m"},"datePublished":"2026-09-30T15:14:32.932Z","dateModified":"2026-10-01T16:51:38.291Z","description":"Researchers from MIT, Carnegie Mellon, NYU, and Stanford built an AI system called Ataraxos that beat top human players at the board wargame Stratego, a game...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/52b1b2c5325be6a9c94f13557476ed8c?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/52b1b2c5325be6a9c94f13557476ed8c?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"MIT News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"MIT News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/881e20a40aa44b12ba4bd27f5c14289d","url":"https://daily.dev/sources/mit"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/this-game-playing-ai-is-the-new-champ-at-stratego-qvig6k90m","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,reinforcement-learning","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"MIT News","item":"https://daily.dev/sources/mit"},{"@type":"ListItem","position":3,"name":"This game-playing AI is the new champ at Stratego"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/this-game-playing-ai-is-the-new-champ-at-stratego-qvig6k90m#faq","mainEntity":[{"@type":"Question","name":"What AI system beat top human players at Stratego and how does it compare to DeepMind's DeepNash?","acceptedAnswer":{"@type":"Answer","text":"Ataraxos, developed by researchers from MIT, Carnegie Mellon, NYU, and Stanford, beat the world's top Stratego player 15-1-4 and went 39-2 against elite players at the world championship. It achieved strictly higher playing strength than DeepMind's DeepNash while using less than one hundredth of the training examples and one thirtieth of the self-play games, making it far cheaper to train. daily.dev surfaces research like this for engineers tracking efficient reinforcement learning techniques."}},{"@type":"Question","name":"How does decision-time planning help an AI handle hidden information in games like Stratego?","acceptedAnswer":{"@type":"Answer","text":"Decision-time planning lets an AI refine its move choices during play rather than relying only on a pre-trained blueprint strategy. Ataraxos uses a generative model that estimates probable identities of an opponent's hidden pieces, then evaluates future outcomes before picking its next move, effectively zooming in on the specific board and opponent instead of guessing blindly. engineers exploring imperfect-information AI techniques can follow this kind of research on daily.dev."}}]} ```