OpenAI admits AI hallucinations are mathematically inevitable, not just engineering flaws
(www.computerworld.com)
sucks when the pesky limits of physical reality get in the way of your AI superintelligences from actually being intelligent.