Two problems. One is that the fraudster isn't going to issue those warnings. I'm trying to think how it would technically work. What's happening is that the individual is not in fact dealing with legitimate organizations. They think they are, but they're dealing with the fraudsters. So how is the warning going to reach them at the time it needs to reach them? I'm not sure how that would work.
Secondly, as I've already pointed out, even if you were able to warn customers—actually, if you could get all customers to stop responding to all phishing and pharming and all of that, it's only going to deal with a fraction of the problem. We still have, but without the statistical support, what seems to us is that perhaps the majority of the problem here is leaks by businesses, hacking into computer databases, insider theft and such, which consumers have absolutely no power over.