Testing GPT-Pro: Formalizing Academic Papers End-to-End for $4

RexDouglass · x · 2026-08-11

A developer demonstrated that for about $4, GPT-Pro can append a full public formalization of an empirical paper end-to-end.

The model can automatically extract the paper's DAG, implied estimands, and required assumptions, checking if those assumptions are met by design or evidence. The author noted that observational social science often browbeats readers to overlook fragile assumptions, a vulnerability that LLMs can expose at a very low cost.

Related event: GPT-Pro Automated Paper Verification Sparks Academic Integrity Debate(2 posts)→

Original post →

More from coding & agent

coding & agent channel →