aprilthird2021
Joined 2 years ago
Joined 2 years ago
This bug could not be found by just asking an AI "find bugs in ChatGPT", that's why it was reported through a bug bounty programs and that's why it should be worth more than $5k
They certainly don't use many of the same words, probably don't sail or float on a river (but how many kids could ever do that). What else? The core elements of friendship, loyalty, cunning, and the base satire are all still relevant and visceral today
...what do you think will cause kids to be unable to understand it? They understand things like the Odyssey and the Iliad, Charles Dickens, the Bible/Torah/Quran/Baghvad Gita/etc., Faust, Don Quixote, etc. all of which were written far earlier
I guess my point is these benchmarks unless they are constantly refreshed with new scenarios and not ever having their outcomes written about publicly or available on a public server, will always be prone to being gamed
This is proven by looking at the disastrous Trump admin. Who came in thinking "Uh government is so bloated we need to spend smarter and cut waste" and did DOGE for a bit, then quit, and added more money to our deficit and barely saved anything and spent more money hiring back essential people bc they couldn't distinguish essential spend from frivolous spend