Something.
Anything. To discourage this behavior. To hit them where it hurts. (The wallet.)
Yes, they can. The fact that someone made that claim about the reliability of the process doesn't mean that the person sending the notice knew that any particular notice was in fact false at the time they sent it, even notices made after the claim of unreliability.
It is possible that if it can be proven that the recipient read the complaint, and failed to investigate it because they believed it was likely to be correct (or if, even without such a notice, they knew of the unreliability of the process and failed to investigate the facts of particular notices and just blindly relied on the process), and that decision was motivated by a desire to avoid discovering that the information was false, then they might be considered to have constructive knowledge, but that's a far cry from "a notice was sent claiming that the process was unreliable, so any error resulting from that process automatically will be found to be 'knowing'."
Setting loose a bot that takes real action in the world when you don't know what it is going to do and failing to even review its actions?
That is a waaaay worse crime!
Punishing or preventing irresponsible bot use will only become more important with the rise of machine intelligence/deep learning, IoT, drones, etc.