Garry Tan: Cross-model review of frontier models is no longer useful
Garry Tan said that some change has occurred in the latest frontier models, and cross-model review and code review are now becoming less useful. His post was in response to lauren's view: adversarial review is basically a waste of tokens; it is better to spend those tokens on empirical methods combined with formal verification.
Topics
View original postmaybe a hot take but i now consider adversarial reviews mostly a waste of tokens. far better to use those tokens on empirical combined with formal verification. i have updated pstack to use the interrogate skill less, and eventually i think i may remove it altogether i have found very little value with code review in the abstract, agents tend to just get lost in the nitpicks and yagnis
View original
maybe a hot take but i now consider adversarial reviews mostly a waste of tokens. far better to use those tokens on empirical combined with formal verification. i have updated pstack to use the interrogate skill less, and eventually i think i may remove it altogether i have found very little value with code review in the abstract, agents tend to just get lost in the nitpicks and yagnis
Body
Something changed in the latest frontier models where cross model review and code review is just less useful now