News Open AI: GeneBench-Pro

[Jovialis]

"A research-level benchmark measuring how AI agents navigate ambiguity and make consequential judgments in computational biology."

https://openai.com/index/introducing-genebench-pro/

 

[davef]

Pass rates for the others is very bad compared to OpenAI, single digits vs 2 for those gpts.
I personally found chatgpt to have gotten freakishly smarter these past 2 months, and less agreeable which is an added bonus so it’s more likely to say you’re wrong rather than pat you on the head even if you’re super off track.
Anyway let’s see this thing take off