OpenAI admits AI hallucinations are mathematically inevitable, not just engineering flaws
(www.computerworld.com)
What do you mean "admits"? It has been known from the very beginning to everyone who has read about LLMs at least from the output of those very LLMs (the irony, huh?).