Arstechnica AI Ataraxos Defeats Top Stratego Player, Marking AI Milestone
Article Content
- •Ataraxos AI defeated top Stratego player Pim Niemeijer in a 20-game series.
- •The AI's training cost was under $8,000, a fraction of previous attempts.
- •Stratego's complexity with hidden information posed challenges for AI until now.
A team from Carnegie Mellon, NYU, Stanford, and MIT developed Ataraxos, an AI that defeated Pim Niemeijer, the best Stratego player, in a 20-game series with 15 wins, one loss, and four draws. This achievement marks the first instance of AI reaching superhuman performance in Stratego, a game with over 10^33 possible setups. Ataraxos was trained on a budget of less than $8,000, significantly less than the $3 million to $4.5 million spent by DeepMind on its DeepNash system, which previously struggled against top human players. The AI's success is attributed to its efficient training methods and a custom GPU simulator. The game of Stratego presents unique challenges due to its imperfect information and bluffing elements, which had previously stumped AI systems. This breakthrough highlights advancements in AI capabilities in complex strategic games.
Ask AI about this cluster
Answers cite the sources they use
Timeline
More articles in this cluster (3)
Following this threat?
Track Carnegie Mellon University in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.
Free account · no card needed
Common questions
What is Ataraxos?
How did Ataraxos perform against Niemeijer?
What makes Stratego challenging for AI?
Continue Reading
Critical Zero-Day Exploited in BeyondTrust Remote Support by China-linked Group A critical vulnerability in BeyondTrust Remote Support, tracked as CVE-2026-102764 with a CVSS score of 9.3, has been exploited by the China-linked espionage group Armory Castle since August 29, 2026. This flaw allows unauthenticated remote attackers to perform file operations, gaining root control and access to…
Silent Lynx Targets CIS Governments via WinRAR Exploit The threat actor Silent Lynx has launched a series of attacks against government organizations in the Commonwealth of Independent States (CIS) from June 2025 to August 2026. These attacks exploit a vulnerability in WinRAR (CVE-2025-8088) to deploy reverse shells, including FoalShell and PickleShell, facilitating…