Thread

1. andy_ppp · 1791017358 · focus · HN ↗

   So predicting the next word given all humanity’s knowledge is surely going to max out at slightly less good (we probably can’t get perfect data) than the best human in any specific field. What test does the AI do to be able to understand it is improving? At some point it becomes impossible to know that the output is actually better right?

   1. tucnak · 1791017519 · focus · HN ↗

      > surely

      Citation needed

      1. andy_ppp · 1791018192 · focus · HN ↗

         Have you seen anything produced by LLMs that is better than the best humans?

         1. petesergeant · 1791019554 · focus · HN ↗

            I hadn't seen anything three years ago produced by an LLM that looked better than the most mediocre humans. The argument here isn't about today's output, it's about the potential of LLMs to outperform humans. And it's far from clear that the architecture is bound this way, or that it only repeats stuff it's already heard.

Unofficial Hacker News client; not affiliated with Y Combinator.