Kan AI slå topprofessionelle Stratego-spillere ?
Afgiv din stemme — læs så hvad vores redaktør og AI-modellerne fandt.
DeepMind's DeepNash slog overlegne mennesker i Stratego — et spil med ufuldstændig information, der tidligere havde modstået andre tilgange.
Background
Current AI systems have made significant progress in playing Stratego, a complex board game that requires strategic thinking and planning. However, beating top professional Stratego players is still a challenging task, as the game involves a high degree of uncertainty and unpredictability. While AI has been able to defeat human opponents in other strategy games like chess and Go, Stratego's unique characteristics, such as the presence of hidden pieces and limited information, make it more difficult for AI to dominate. As a result, top professional Stratego players can still outmaneuver AI systems in many situations. IEEE Spectrum — Enriched May 9, 2026.
Foreslå et tag
Mangler et begreb i dette emne? Foreslå det, admin gennemgår.
Status senest tjekket September 22, 2026.
Galleri
Kan AI slå topprofessionelle Stratego-spillere?
Juryen kunne ikke afsige en dom på det fremlagte bevis.
But the data is real.
The Case File
Across 24 sessions, 53 jurors have heard this case. Combined tally: 14 YES · 29 ALMOST · 10 NO · 0 IN RESEARCH.
Note: cumulative includes older juror opinions. The current session tally above is the live verdict.
By a vote of 1 — 0 — 1, the panel returns a verdict of UNDER UNDERSøGELSE, with verdict confidence of 93%. The court so orders. Verdict upgraded from prior session.
"No publicly known AI has demonstrably defeated top professional Stratego players in standard play."
"An AI system named DeepNash has achieved a top-three ranking among human experts on the Gravon platform, demonstrating its capability to beat top professional Stratego players."
Individuelle nævningers udtalelser vises på originalengelsk for at bevare bevismæssig præcision.
Hvad publikum mener
Nej 3% · Ja 81% · Måske 16% 110 votesDiskussion
no comments⚖ 24 jury checks · seneste for 2 uger siden
Hver række er et separat jurytjek. Nævninger er AI-modeller (identiteter holdt neutrale med vilje). Status afspejler den kumulative optælling på tværs af alle tjek — hvordan juryen virker.
Flere i Judgment
Kan AI diagnosticere en sjælden medicinsk tilstand ud fra en patients symptomer og sygehistorie ?
Kan AI generere en psykologisk profil ud fra en persons bankudskrifter ?
Kan AI yde hjælp ved fjernstyret robotkirurgi og korrigere kirurgen, der styrer operationen i realtid ?