AIs should be globally banned from making final decisions. Whenever an AI makes a decision, it should be guaranteed that you can demand a human to actually revi

AIs should be globally banned from making final decisions. Whenever an AI makes a decision, it should be guaranteed that you can demand a human to actually review the case, evaluating evidence manually. It is also questionable whether lifetieme bans should be allowed to exist, let alone full bans like when Google's AI dislikes something in your app and you loose access to your GMail and everything forever (in my oppinion not even a terrorist deserves this).