In a landmark study, OpenAI researchers reveal that large language models will always produce plausible but false outputs, even with perfect data, due to fundamental statistical and computational limits.
they’re not errors either, because that implies they’re unintended. hallucinations are the program working as designed. they are more like… consequences.
deleted by creator
they’re not errors either, because that implies they’re unintended. hallucinations are the program working as designed. they are more like… consequences.
I like the term “confabulation”.
https://en.wikipedia.org/wiki/Confabulation