Is the author forgetting a baby in the bathwater he is throwing out?
Especially on coding. He points out that vibe coding is bad, and then concluding that any program written through the use of an AI is bad.
For example if you already have a theory of your code, and you want to make some stuff that is verbose but trivial. It is just more efficient to explain the theory to an LLM and extract the code.
I do like the idea of storing the underlying prompt in a comment.
Same for writing. If you truly copy paste output, it's obviously bad. But if you workshop a paragraph 5 or 6 times that can really get you unstuck.
Even the euler angles example. That output would be a good starting point for an investigation.
A lot of criticism of LLMs is certainly legit and relevant, but are targeting very naive uses of the tools. Post-processing the output with an improvement prompt isn't even mentioned here. I'd rather read a criticism of agents, systems that produce SEO optimized articles etc. The debates would be more interesting.
For example if you already have a theory of your code, and you want to make some stuff that is verbose but trivial. It is just more efficient to explain the theory to an LLM and extract the code. I do like the idea of storing the underlying prompt in a comment.
Same for writing. If you truly copy paste output, it's obviously bad. But if you workshop a paragraph 5 or 6 times that can really get you unstuck.
Even the euler angles example. That output would be a good starting point for an investigation.