Should this be “If you can prove that you will cooperate, defect”? As it is, I don’t see how this prevents cooperation with Cooperatebot, unless the agent uses an inconsistent system for proofs.
It kills the Lobian argument, I believe, since this implication “if there’s a proof that you cooperate, then cooperate ” is no longer true. Instead, here’s a Lobian argument for defection:
Suppose there is a proof that you defect. Then either there is a proof of contradiction, or there is no proof that your move is the same as your opponent’s. Either way, you defect.
Should this be “If you can prove that you will cooperate, defect”? As it is, I don’t see how this prevents cooperation with Cooperatebot, unless the agent uses an inconsistent system for proofs.
It kills the Lobian argument, I believe, since this implication “if there’s a proof that you cooperate, then cooperate ” is no longer true. Instead, here’s a Lobian argument for defection:
Suppose there is a proof that you defect. Then either there is a proof of contradiction, or there is no proof that your move is the same as your opponent’s. Either way, you defect.