Comment by Ironchefpython

I think it would be more accurate to say that LLMs are pretty human-like in their limitations.