I think the problem statements themselves, like this is where we're starting to hit the ceiling in the benchmarks. Like what I read was a very controlled vignette and patients are messy. One of the things we found really messy about like diagnosis and decision support and medicine is that it's an evolving thing. So any model you build has to be able to like dynamically reason and change over time. And these benchmarks maybe don't do that, but anecdotally, the language models do really well.

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode