Microsoft's own different versions of Office can't always reliably read/write docx between them.
Is a layer of LLM going to make this better or worse? Could you train a model to be very good at it?
Is a layer of LLM going to make this better or worse? Could you train a model to be very good at it?
We should all know what happened when the US government turned on Colossus (D.F.Jones 1966) and it immediately found there was another. That collaboration was humanities near instant undoing.
To answer your second question, yes, I think it's inevitable that LLMs will become very proficient with all commonly used file formats.