The price of beating a human world champion just hit $8,000. ataraxos, from mit, cmu, nyu and…
the price of beating a human world champion just hit $8,000. ataraxos, from mit, cmu, nyu and stanford, beat stratego's greatest player 15-1 with four draws, published in nature on sept 30.
it plays a game where half the board is hidden.
Context
MIT News of September 30, 2026 says the system, called Ataraxos, beat the strongest Stratego player in the world by a record margin of 15-1-4 and went 39-2 against top human players at the Stratego world championship. It says the work appears in Nature, with researchers from Carnegie Mellon, MIT, New York University and Stanford.
Ars Technica says Ataraxos beat Pim Niemeijer 15 games to one with four draws, and that it took 16 GPUs and a few thousand dollars to train. It says the Ataraxos team estimates DeepNash's training would cost $3 million to $4.5 million at 2025 prices, against 16 GPUs for a week plus four GPUs for four days for Ataraxos.
The 15-1 with four draws, the four universities and the Nature publication on September 30 match the sources. The note's 'half the board is hidden' is loose: the sources say the opponent's piece identities remain hidden.
The note says the price was $8,000. Ars Technica says a few thousand dollars, and the dollar figure of $8,000 was not found in the pages read, so it is unsupported here, not refuted. 'Few thousand' is the only figure the sources give.
The headline that this is the price of beating a world champion rests on the training cost only. The sources do not count the researchers' time or the earlier development of the algorithms.
Related work
- This game-playing AI is the new champ at Stratego (MIT News, September 30, 2026) ↗Primary source for the match record, the team and the Nature publication.
- With most information hidden, the game Stratego had stumped AI until now (Ars Technica, October 1, 2026) ↗Source for the opponent's name, the GPU counts and the cost comparison with DeepNash.
- Scalable decision-making for games of imperfect information (Nature, September 30, 2026) ↗The Nature paper; surfaced in search and not read in full here.
Watch next
- Read the Nature paper for the training compute and cost estimate. Check how the $3 million to $4.5 million DeepNash estimate was derived.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 9 October 2026 at 01:47 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →