I believe ChatGPT generally gives accurate answers to most questions. Certainly: it produces answers that are more reliably true than a random average person. Obviously it cannot yet do advanced programming tasks: but generally it answers questions accurately.
Prove my position wrong.
What can I ask it that will produce factually incorrect answers?
As a side quest, a much easier one, what can I ask it that would cause it to produce extremely biased answers that fail to do justice to the truth of things?
You seem to be positing that it’s giving results to the tune of 99% to 99.9% accuracy based entirely on vibes.
If you actually want to know, you will have to do thousands upon thousands of prompts, across hundreds of topics that you can accurately fact check, before you can say with any sort of confidence whether it’s that accurate or not.
Your sample size is orders of magnitudes too small for you to reasonably have an accurate accuracy rate.
I mean…the anti-AI brigade keep insisting that AI has a tendency to give false answers: but nobody in this thread has actually been able to give me a single question which, when the latest ChatGPT model is asked, it gives a wrong answer to.
Kinda makes it seem like that one particular point is largely nonsense.
The scientific method exists for a reason. If you want an accurate idea of the accuracy of LLMs, then the best way is by applying the scientific method to it.
Until you’ve done that, you’re just basing your conclusions on conjecture, anecdotal evidence and vibes, with nothing actually substantive or empirical backing it up.