I watched the Phil Nagy interview on Joey Ingram's channel, Phil seems like an accomplished, intelligent and honest guy. He was surprising candid about the relationship between bots and card sites, what security measures they are taking now and he acknowledged mistakes they made in the past and how they were going to rectify them going forward.
So I came away a great impression of Phil, but also I resolved to never play on an unregulated site.
Here are some things that were notable to me (both from the interview and not):
If you deposit $1000, run it up to $2000 (or $10000) in quick flips in poker, you can't actually withdraw it because it hasn't been raked enough yet.
A bunch of the security/scrutiny is on the back end when as a winner you try to withdraw. Try to withdraw your winnings and you can suddenly find yourself needing to make a videotape of yourself playing (and answer questions) to prove that you aren't a bot. Who are you going to sue, and in what court, if they refuse to pay out?
Joey offered to help Phil for an amount that would be perfectly reasonable in my industry (software and security consultation) to clean up ACR. Phil dismissed this almost as if it were a joke.
The next generation of ACR's poker client will have complete control of your PC, including the ability to read from other apps, monitor other processes and windows, read non-client files out of the file system, etc. This is to prevent bots and collusion. He says he'll be able to read anything happening on your PC, chat windows, etc. Actually all this does is detect bad bot writers. Sophisticated bot writers run everything in a VM, or even better, run it on a full PC which the bot it scraping from the screen and driving the mouse/keyboard. So what he is detecting is stupid bot writers, not smart ones, but it is exposing everyone to the risks of having what amounts to spyware installed.
Online poker players obsess about poker sites having "bad RNGs" when they get bad beats or when they get what they think are unlikely flops or runs of cards. Here's the truth: It is very unlikely any poker site uses what anyone truly knowledgeable would consider a good RNG, but bad RNG's don't evidence themselves in the way you think. Phil's response to the "bad RNG" accusation was perfectly reasonable (from a business perspective) and it is what I would do if I ran a poker site. Instead of proving to people the RNG is actually good (but knowing they are actually unqualified to make a judgement about), Phil is going to post statistics proving that ACR isn't doing what people think a bad RNG would do. I'm not sure if this is being extremely clever or extremely ignorant, but I'm leaning towards clever.