I reply to everyone, but this post above was apropos of my response. BOTS are not necessarily the problem relative to the much worse problem of COLLUSION.
Think about this. At ACR let's say there are a dozen or so tables at most at every stake level. Well that's actually generous. All you need is a few dozen conspirators to join together and play against the "non-colluders" at every table. If you have friends in different countries it's almost impossible to detect. Perhaps ACR does look for BOTS but they do not have a good method of rooting out colluders. Perhaps there is no way because they actually ask YOU to collect the evidence. Burt I will tell you this:
Colluders CANNOT BE BEAT.
Colluding humans are bad, indeed, but they can't compare with the effectiveness of a colluding bot ring. The sharing of cards, and the precise calculation of odds on the fly based on the extra info, is the bots' strongest point, after all. If the bots can't be detected, then they'll always be a bigger problem than humans trying to do the same.
Collusion alone is really hard to detect, especially when the number of players is low, because the colluding players have a legitimate reason to be on the same table regularly. There is no antidote for that. I mean, what can you possibly do?
For example, you could try to track the players in order to find whether they are consistently running well above the EV. Then you could investigate their playstyle for any suspicious behavior, which doesn't make sense from a GTO poker point of view. How can you prove they are cheating, though? The legit highlevel poker players are pretty good at reading and adapting to the holdings of the other players. Strange, unpredictable plays are part of their arsenal too, and GTO poker itself involves a mixed strategy.
Unfortunately, the best way to avoid collusion is to stay away from cash games and small tournaments. It's safe to play such games only with a trusted ring of players.