Is it still artificial intelligence when the task is to model human stupidity, or would only preventing its devastating consequences get an AI rating?
Erik Naggum