Paul Goldsmith-Pinkham

Homepage of the economist Paul Goldsmith-Pinkham

Latest posts

Keeping a list of AI writings and Claude Code posts

Keeping a list of all the writings and things I’ve done with AI, both here on the blog and over on Substack.

Apr 27, 2026

Why learn outside your field?

My friend is teaching to his Economics Ph.D. students and explaining why they should be interested in topics that are not in their subfield.

Apr 27, 2026

An Incomplete List of Things I Want to Build

AI has lowered a huge number of frictions for me with respect to programming. I am now able to code up many tools and ideas, and I really enjoy it. But it’s not uniformly doable across all projects — not because the coding is too hard for LLMs (I am not enough of a computer scientist to think of hard problems) but because humans have created moats and frictions to make things inaccessible or…

Apr 8, 2026

Resampling Asset Prices the Right Way

New monograph from Richard Crump and Nikolay Gospodinov that I think deserves more attention: Resampling Asset Prices: An Identity-Based Approach (Cambridge Elements in Quantitative Finance, 2026). Available free here, with a full MATLAB replication package on GitHub.

Mar 31, 2026

AI one-shot papers

On Twitter/X and LinkedIn, there has been a large discussion of the ability to “one-shot” policy evaluation, which David Yanagizawa-Drott has been pursuing in an interesting way: ape.socialcatalystlab.org

Mar 21, 2026

The Butlerian Jihad Was Never About the Machines

The Butlerian Jihad

Mar 19, 2026

Research in the Time of AI

Academics are worried about AI replacing them and their work. I think we spend insufficient energy appreciating how much we can expand the frontier of research thanks to this new tool. To make this concrete, I find it useful to think about research as a pipeline of stages, from ideation through publication, and to ask where AI actually changes things.

Mar 16, 2026