Post

Conversation

LLMs are very good at extracting information from academic articles. They are much better than even highly trained humans (our grad RAs had hundreds of hours of practice). And of course they're ~1000x cheaper and faster
Quote
Ryan Briggs
@ryancbriggs
Replying to @ryancbriggs
We coded our ~100k articles using LLMs. Should you believe them? To answer this, we benchmarked 4 human RAs against 3 LLMs on their ability to recover ground truth article data. Details in the paper and appendices, but the LLMs did well and handily beat the highly trained humans.
Image
David Watson 🥑
Post your reply

Why did the humans do badly? Honestly, the task is so boring but also requires sustained thought and focus. It's genuinely really hard to do it well for hours on end. When I was piloting I did a lot of this myself and (earlier, worse) LLMs often beat me too.
Running parallel agents to do qualitative analysis of literature is a game changer! (It needs close human review afterwards, but so do most grad students.)
Image
Everyone keeps posting “SaaS is dead” in a post-AI world. I think they’re half right. The last vintage of SaaS was designed for one primary user: humans. Clicks. Dashboards. Busywork disguised as “workflow.” The next vintage of SaaS is designed for a new user: AI agents first,

Discover more

Sourced from across X
“Jon faced a lot of pushback when he said professional mathematicians focus too much on counting and neglect symbolic reasoning. He’s heard your criticism, so stay tuned for this week’s guest, one of the world’s leading mathematicians: The Count from Sesame Street”
Quote
The Weekly Show with Jon Stewart
@weeklyshowpod
Jon noticed some anger from economists in the comments last week. He gets into that and so much more with the Chief Economist of American Compass @oren_cass on a new episode tomorrow. #theweeklyshow #jonstewart #politics
The media could not be played.
Readers added context
Oren Cass does not have any formal training in economics beyond the introductory undergraduate level. Cass obtained his title of "chief economist" by giving it to himself. pinpointpolicyinstitute.org/the-point/oren…
You rated as helpful
Edit
This is a very bad precedent. Regulatory agencies changing their stance after ok ing a study is one of the worst things for innovation.
Quote
Christina Jewett
@By_CJewett
NEW: The F.D.A. refused to accept an application from Moderna for its mRNA flu vaccine. Its reason: The agency did not think Moderna compared the new vaccine to one of the best flu shots available. The company spent $750M+ on a 41,000 person study. nytimes.com/2026/02/10/hea
Inspired by Oren Cass naming himself Chief Economist at his own organization, I’d like to say that I have named myself an economist at my own think tank, More Interesting Thoughts.