I’m seeing some applications for it as an easy-to-implement context-aware OCR scanner for things that have consistent content in wildly inconsistent formats. Think lists of open invoices from suppliers in a thousand different for-printing layouts, plus some just being screenahots from their systems. The AI system can process those and output a consistent list, which can then be matched against a database the old-fashioned way.
I’m seeing some applications for it as an easy-to-implement context-aware OCR scanner for things that have consistent content in wildly inconsistent formats. Think lists of open invoices from suppliers in a thousand different for-printing layouts, plus some just being screenahots from their systems. The AI system can process those and output a consistent list, which can then be matched against a database the old-fashioned way.
That’s the only practical use I’ve found.
I think that’s a fantastic use with a few caveats. Feel free to take this and start a consulting company on this:
Is an LLM what I want?
is the goal to parse or generate natural language?
does your use case have tolerance for inaccuracy?
If both answers are “yes”, go. Go full steam. Otherwise? No.
Outgoing call center? Go. Reading resumes? Go. High tolerance for mistakes. Cast the widest cheapest net.
Math? Health? Explosions? No.