Never have I seen someone so eloquently explain the biggest problem with AI-assisted development.
This is what the AI-coding pitch never confronts. Yes, generate code faster, sure. But the catching — the eye that says no — comes from somewhere AI cannot reach.
@Linux_in_a_Bit the funny part? That experienced intuition is the result of our literal neural network brains. So while it's true that current LLMs don't have that ability, we're directly on the path of changing that, and the reasoning models were already a huge leap in that direction.
That also makes your conclusion very wrong. It's somewhere AI can not only reach, it's what it was invented to do.
@Linux_in_a_Bit I was under the impression that modern LLM stacks internally use a bunch of tools that the language model interacts with and is influenced by, including other neural networks.
