But that's my point. LLMs generate good prose "most of the time", certainly better than most people are capable of doing. Yet we frequently react with disgust w

But that's my point. LLMs generate good prose "most of the time", certainly better than most people are capable of doing. Yet we frequently react with disgust when we see tell-tale signs of LLM-generated text in articles. Why? Because it indicates the person was probably too lazy to write it themselves and are simply chucking a half-formed thought over the wall? Why don't we hold generated code to the same standard?