3 ms·
On the Limits of Engine Analysis for Cheating Detection in Chess [pdf]
- jruohonen 3y agoPrompted by: https://news.ycombinator.com/item?id=38780394 https://news.ycombinator.com/item?id=38780394
- INGELRII 3y agoElite Chess Players Keep Accusing Each Other of Cheating "Seeds of doubt have grown into a disturbance among some in the game’s top ranks." https://www.nytimes.com/2023/12/25/crosswords/chess-hikaru-vladmir-kramnik-cheating.html https://www.nytimes.com/2023/12/25/crosswords/chess-hikaru-v... The lure of money, where the risk of direct detection remains minimal makes cheating inevitable. Incentives are strongest with chess content creators, such as Hikaru. Securing direct evidence is a near-impossible task. Direct evidence means, tracing through game statistics and behavioral cues. The evidence should be rock solid before starting to accuse any individual. This is a systemic problem. While Kramnik and Carlsen might be right, their approach poisons the game deeper. Perhaps a better tactic would be to abstain from naming culprits. Instead, they could have provided anonymized statistical evidence and explained their methodologies. Public consensus before naming the names could have been a plausible. Systemic problem requires systemic solutions. Top players might have to play significant amount of their games in secure "SCIF" under observation, or something, to keep playing.