In some really important ways, current systems are superhuman. I'm always kind of interested in the ways that reality kind of diverges from our expectations or our shorthand. And one key way is that like, we're seeing sort of superhuman things in some ways, but not in all ways, right? That's just kind of worth not glossing over in my mind. We actually introduced it a few papers ago called like GPT recall or GPT precision. So you can actually get the model to like ask the model if the state, if the open-ended answer was the same as the reported answer.

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