[severusdd]

LLMs are amazing at writing code and terrible at owning it.

Every line you accept without understanding is borrowed comprehension, which you’ll repay during maintenance with high interest. It feels like free velocity. But it's probably more like tech debt at ~40 % annual interest. As a tribe, we have to figure out how to use AI to automate typing and NOT thinking.

[usrbinbash]

> is borrowed comprehension,

Or would be, if the LLM actually understood what it was writing, using the same definition of understanding that applies to human engineers.

Which it doesn't, and by its very MO, cannot.

So, every line from an LLM that is accepted without understanding, is really nonexistent comprehension. It's a line of code, spat out by a stochastic model, and until some entity that actually can comprehend a codebases context, systems and designs (and currently the only known entity that can do that is a human being), it is un-comprehended.

[croes]

Depending on the task they are terrible at writing code.

[ansgri]

This is a very good analogy. And this interest rate can probably be significantly reduced by applying TDD and reducing the size of isolated subsystems. That may start to look like microservices. I generally don’t like both for traditional development, but current LLMs both make them easier and more useful.

And the “rule of three” basically ceases to be applicable between components — either the code has localized impact, or is a part of rock-solid foundational library. Intermediate cases just explode the refactoring complexity.