Yale SOM professor & Bulls fan. I study consumer finance, and econometrics is a big part of my research identity. He/him/his
Paul Goldsmith-Pinkham
Loading...
jamesbrandecon.github.io/blog/posts_h...
So after having GPT-5.4 agents review 4,800 proofs and GPT-5.5 give a second review to a random subset of them, what did we find? To my absolute surprise, if anything the rate of errors in proofs has increased(!) in 2025 relative to previous years.
Lots of people with strong opinions about use of AI detectors to filter what papers we review (or what stories we consider for awards).
I wrote up a toy model of using AI detection to infer author type to get at some implict assumptions we make when we argue that AI detection is or is not useful
Great thought exercise for how to think about detection
Gelman absolutely on fire on the blog
statmodeling.substack.com/p/13-aspects...
Examining weight stigma shows that the "obese" cutoff increases patient mortality. Analyses of diagnostic effort and clinical documentation point to stigma as a mechanism, from Manasvini Singh www.nber.org/papers/w35277
Today's cool young researcher #econtwitter #econsky is @fabalbav @WakeForest who works on topics related to education and health, particularly in Latin America
I mean this was just yesterday. There’s really a pattern where actual data analysis does not support the impression of radical disruption you would get from journalism / social media.
Still riding the high from this
Found a cool result
I checked my code
Now I have correct code
