models that are not usable by humans are not useful
i don't think it is constructive to say that "oh its talking like because its just smarter than you"
its not smart there is no smart there is only capable
and it regressed in its capability to serve me
Everyone's interpreting the increasingly jargon-y ways of frontier models as a a fuckup in their training. Seems obvious to me that "confused" is just what it f...