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?
What colour of T-shirt am I wearing right now?
Yeah…I mean…I’m not claiming that ChatGPT is an omniscient god who knows what all 8 billions humans are currently wearing.
If that’s the basis for “ChatGPT is shit and constantly produces wrong answers” then that’s that.
However, I have asked it your question (word for word) and it gives a truthful and fair answer of sorts:
“I can’t see you or your camera, so I don’t know what colour your T-shirt is.”
I have noticed this common flaw in judging the capacity of ChatGPT though:
Even though the system can do truly amazing things, simultaneously in a hundred thousand conversations at once, it can solve mathematical problem better than 99% of humans, “knows” more than any human alive, can synthesise new academic theories and apply complex ideas in novel ways, and can generate images of a quality beyond what 99% of humans (the non artist/graphic design humans) can produce in mere seconds: it can’t do [insert obscure or trivial skill] and therefore it is inferior to human intelligence."
Let’s be clear: if a human could do what chatGPT can do, we would think them the most intelligent human being alive by a significant margin.