LLMs are reinforced through adversarial training - you would essentially be playing a keep-up game with AI generated garbage that would get exponentially more d

LLMs are reinforced through adversarial training - you would essentially be playing a keep-up game with AI generated garbage that would get exponentially more difficult to pull ahead in.