HackerTrans
トップ新着トレンドコメント過去質問紹介求人

camillemolas

no profile record

投稿

Show HN: I built a game where domain experts try to break frontier AI

rusmarterthananllm.com
3 ポイント·投稿者 camillemolas·4 か月前·8 コメント

コメント

camillemolas
·4 か月前·議論
We’re also very much interested in multimodal. Do you take pictures, recordings, videos, or anything along that in your domain? We want to find out if models can fail using those as well!
camillemolas
·4 か月前·議論
If it fails let me know!! That’s exactly what we are looking for.
camillemolas
·4 か月前·議論
Thanks! Hopefully you get to beat it and get paid out $$ but also bragging rights !
camillemolas
·4 か月前·議論
Thanks! Hoping you join as well.
camillemolas
·4 か月前·議論
Yes private equity secondaries is a great domain for this. The valuation edge cases and LP agreement interpretation are exactly where frontier models fail confidently. The data becomes part of a verified record of AI capability gaps and is valuable to labs and enterprises building finance AI.

Payment is coming. Right now we’re building the expert network. Verified failures will be compensated monetarily. Would love to have you as an early finance expert, throw your hardest question at it.