These outcomes aren't because it's 'not ready'. This issue is fundamental to the nature of LLMs. It will never be ready for this use case.
These outcomes aren't because it's 'not ready'. This issue is fundamental to the nature of LLMs. It will never be ready for this use case.
I think that’s probably true. Increasingly I’m convinced it’s a productivity booster for subject matter pros who are also trained on how to use it. Like Westlaw for lawyers; but that’s useless for non-lawyers whereas AI can be affirmatively misleading.