People sometimes do that with children’s questions.

The traditional answer is that these things are just (incredibly) advanced ‘next word’ predictors. If they don’t know the answer — and don’t know that they don’t know it (because there’s no training on it not being true) — they make it up because it’s grammatically allowed.
Crucial key to understanding what’s going on in LLM / ChatGPT hallucination (& logical errors)
219
4
Paul Pallaghy, PhD
Anthony (Tony/Pcunix) Lawrence 👀
·Follow
Jan 18, 2024
--
People sometimes do that with children’s questions.
--
--
Written by Anthony (Tony/Pcunix) Lawrence 👀4.6K Followers
·142 Following
Retired Unix Consultant. I write tech and humor mostly but sometimes other things. See my Lists if your interests are specific.
No responses yet
Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams