[chatmasta]

> In the past, the platform would quietly close the account of any grandmaster caught using computer assistance.

> Now when a high-level player is caught, Chess.com plans to make the ban public.

What could go wrong? There are no risks of false positives in a proprietary and closed system making probabilistic and unfalsifiable judgments against players. It’s not like it could impact their career.

[sterlind]

Case in point: the Viih_Sou incident [0]. Viih_Sou was an anonymous account famous for defeating Daniel Naroditsky and Wesley So with a meme opening: 1.a4/...a5, then sacking the rook on move 2. After Chess.com banned the account, GM Brandon Jacobson voluntarily outed himself as Viih_Sou, and insisted he didn't cheat.

Despite the opening looking Bongcloud-tier and objectively terrible (against a computer), it's surprisingly hard to refute as a human, especially during blitz, for two reasons:

1. Rooks are endgame pieces. Many blitz games don't last to an endgame. Trading your late-game piece for one of your opponent's most crucial opening/midgame pieces gives you an incredible jump on activity.

2. The gambit completely neutralizes the opponent's knowledge of opening theory. On the flip side, Viih_Sou has the advantage of studying dozens of lines for months.

Chess.com hasn't reversed the ban, but I think Brandon was telling the truth. I think using a bizarre opening that sacrifices material and involves questionable lines, followed up by the exacting genius of a GM playing top engine moves, stuck out as a statistical anomaly and tripped their Fair Play system.

0. https://www.reddit.com/r/chess/comments/1claxsm/its_me_viih_...

[cedws]

Not sure why anyone would want to play on a site that uses “yeah looks like cheating so you must’ve cheated” as a heuristic for a ban.

[veidelis]

"Young Alireza Firouzja was auto-banned on chess.com because he was deemed too successful for an untitled player."