It may be difficult, but people are trying. I'm sure you are aware there are very serious efforts to make very good bots. Check
here if you are interested. I don't really know anything about what any of these people, or others, are doing, but I don't really see why they couldn't make a poker deep blue if they devoted enough resources to it.
Thing is, a successful bot on party will be optimized to play against typical opponants. It probably wouldn't be based on really evaluating any particular players patterns, as they are probably not very static anyway.
But, even if a bot isn't that great, it is probably better than most of the players and I hope the web sites do a decent job of detecting them.