[sebmellen]

Constrain the output to a known set of responses by adding a translational layer where you write the enum and the LLM picks the value.

[nyrikki]

If you have a ground truth function, there is no reason to use an LLM outside of marketing.

[Terr_]

That's like saying search-suggestions are nonsense because the system already has a "ground truth function" in the form of all possible result records.

Helping pick a choice--particularly when the user is using imprecise phrasing or non-exact synonyms--is still a valid workflow.

[nomel]

I don't think this fits the "non trivial user input" of my question, but, in my opinion, your "correct" use disallows most of the interesting/valuable use cases for LLM that have nothing to do with chat, since it requires sanitizing all external/reference text. Wouldn't you be mostly limited to what exists within the LLM? Or, do you think all higher level stuffs should be done elsewhere? For example, the LLM could take pre-determined possible inputs and generate an SQL statement, then the rest would be done elsewhere?