That is where we differ. It is, ultimately, the victim of a scam who makes the choice of "yes, this person is trustworthy and I will do what they say". The only way to prevent that is to block the user from having the power to make that decision, which is to say protecting them from themselves.