My experience is that the higher rated bots tend to follow chess principles more closely - they take time to open, have a middle game, and an endgame. This gives you time to do the same with the methods you've learned. Humans around the 600-1000 rating don't always do that - so while you are working on your opening, getting your pieces set up how you'd like them, they're going willy nilly with the slash and burn method. If you don't pay attention to what your opponent is doing and instead are going through the checklist in your head from what you've seen the bots playing (ie: when I open E4, bot will play E5. I play Nf3, bot will play Nc6) you may run into trouble with your pieces where they haven't fully finished developing to prevent the slash and burn player.
So, you have to keep up with what you're doing, but also keep an eye on what the opponent is doing. This is something I'm still working on myself.
Does someone know how the AI rating compares to human players? For example, I have been able to beat all bots up to and including Nelson, who according to chess.com has a rating of 1300. I still have a little trouble with Nelson, but I seem to be able to beat him more frequently now.
However, when I play against humans, I have been unable to beat anyone over like 750. I generally only win against people around 600-650.
So my question is, do these ratings even compare?