AI-202X: a game between humans and AGIs aligned to different futures? — LessWrong