Here's the problem with Large Language Models

1. They lack context.

It's hard to load in all institutional knowledge and live intent; however, even if you do that, it seems they struggle parsing relevant vs irrelevant, contemporaneous vs. legacy, and relevant vs irrelevant context for any given request.

2. They lack intuition

They can't seem to make simple, intuitive leaps about the questions or problems they're asked about. They can often miss the spirit of your question initially or, even more likely, quickly forget the constraints you provided even a few prompts earlier.

3. They lack precision and rigour

Their answers can often include category errors, incomplete lists, and many more very subtle mistakes that are easy to miss.

4. They lack insight

Unless you prompt very carefully with precise questions and context, they will generally give you generic, milk toast answers that are not always appropriate for your particular problem or situation.

They typically lack critical taste and judgment that would help you avoid critical landmines.

In many ways, if you're dealing with a complicated problem, you basically need to be a domain expert to ask the exact right question to get the exact correct answer.

This essentially mitigates the reason you were using them in the first place (trying to avoid involving a domain expert).

So, in summary:

LLMs don't eliminate the need for experts. They can only accelerate (and increase the value of) the work of experts.