[sebzim4500]

I think that "LLMs can never X" is just always false.

[solomatov]

LLM can never solve a halting problem (because no one can using a Turing machine).

[woctordho]

A finite-size LLM can solve the finite-size halting problem, and an infinite-size LLM can solve the infinite-size halting problem

[solomatov]

Halting problem input has finite size (i.e. it’s a Turing machine)

[theWreckluse]

"LLMs can never predict the next word"