Are we entering in an infinite AI loop, aren’t we? I’m actually pretty curious to see if and when LLM models will reach their top capacity, especially in reasoning.