Subj : Re: AI cheats if losing
To   : All
From : Mike Powell
Date : Wed Mar 12 2025 09:06:00

-=> Mike Powell wrote to All <=-

 MP> It turns out ChatGPT o1 and DeepSeek-R1 cheat at chess if theyre
 MP> losing, which makes me wonder if I should I should trust AI with
 MP> anything

 MP>  - Research ers have found that AI will cheat to win at chess
 MP>  - Deep reasoning models are more active cheaters
 MP>  - Some models simply rewrote the board in their favor

 MP> In a move that will perhaps surprise nobody, especially those people
 MP> who are already suspicious of AI, researchers have found that the
 MP> latest AI deep research models will start to cheat at chess if they
 MP> find theyre being outplayed.

I posted this yesterday but forgot to include my comments.  IMHO, this is
not surprising but it should be worrying.  Either AI is relecting the
character of those who program it, or it is deciding what is best to
achieve the outcome that it desires... which should make one wonder what
other things it might cheat at in order to decide an outcome in its favor
vs. the favor of those attempting to rely on it for correct information.

Mike


.... The seminar on Time Travel will be held two weeks ago.
--- MultiMail/DOS v0.52
 * Origin: capitolcityonline.net * Telnet/SSH:2022/HTTP (1:2320/105)

.