That makes sense and I think everyone should learn to code regardless if a machine can automate it because it makes the person better at thinking - which is what really differentiates us from other animals.

But I think people object to you saying LLMs are not coding. Using them certainly does not help the person using them think the same way making all the decisions while writing the code. But even if LLMs are a Chinese room, if they could produce the same output as a human (given the same goal and information context), they IMO would be coding.