Comment by action-k

LLM hallucinations are a well known problem, and accepted as such by OpenAI etc... perhaps, as a group / statistically, that rises to the level of "knowing statements [can be] false"? of course, IANAL... but keen to see how it plays out.