logoalt Hacker News

zmjtoday at 12:02 AM1 replyview on HN

It's not contradictory to say that expertise is a multiplier, and that models are systematically underconfident in themselves.


Replies

titzertoday at 12:52 AM

It's actually refreshing when a model is sure about something because it actually tested it and has the receipts. Opus 5 seems really good about testing its own knowledge with experiments. Scientific method ftw.