An NBER working paper covering more than 500,000 GitHub developers found that interactive coding agents increased weekly lines changed by 958% and pull requests created by 86% in weeks 21–30 after ado…
Harvard theoretical physics professor Matthew Schwartz wrote on Anthropic's science blog Thursday that he produced 36 manuscripts across 18 fields with 19 co-authors in three months using Claude and a…
A new NBER working paper estimates that from November 2022 to December 2025, AI raised the market's expected present value of software engineering productivity by the equivalent of a permanent 32.6% p…
A three-month randomized trial published by the NBER on September 17 found that 133 patent lawyers at eleven US firms using a custom AI drafting assistant improved work quality by 0.34 SD at ten days …
A three-month field experiment on 133 US patent lawyers by Autor et al. (2026), published as NBER paper w35720, found that lawyers given access to a non-public AI drafting tool produced documents aver…
A pre-registered three-month randomized controlled trial of 133 practicing patent lawyers at eleven U.S. intellectual property law firms found that access to a custom AI drafting assistant raised benc…
A review of empirical evidence on coding agents finds no controlled study establishes the optimal specification length, but four narrower conclusions hold: AI accelerates code production more reliably…
A new NBER working paper by Nicholas Bloom, Gideon Moore, Lisa K. Simon, and Caelan Wilkie-Rogers finds that 19.7 percent of established U.S. LinkedIn users retroactively edit job titles or descriptio…
Challenger, Gray & Christmas recorded 101,743 AI-cited U.S. job cuts through June 2026, nearly double the full-year 2025 total, and the NBER projects roughly 502,000 by year-end, but BCG projects 50-5…
A review of eight controlled studies comparing human, AI, and human-plus-AI performance finds that AI's advantage is not universal: it is large and reproducible on some tasks but reverses or disappear…
A new NBER working paper analyzing over 100,000 GitHub developers finds that AI coding tools boosted commits by up to 180% with autonomous agents, but gains fell to 50% for projects and 30% for releas…
A new analysis by Iain on the blog argues that the divide between AI 'foomers' and 'doomers' in software development is a systems failure, not a cultural one, citing NBER Working Paper No. 35275 which…
A software developer argues that the pressure to keep AI agents constantly coding is a manufactured fear that primarily benefits inference providers, not developers. Citing an NBER working paper from …
A systematic review of 181 documents, including 116 empirical studies, finds that AI-assisted software development accelerates code production more reliably than delivery, with controlled assistant st…
A developer built a data pipeline using DSPy and Claude Code to analyze whether AI discourse changed in academic research and policy discussions during the ChatGPT era. The pipeline processes NBER jou…
A new NBER working paper tracking over 100,000 GitHub developers found that AI coding agents increased lines of code by 741% and pull requests by 65%, but shipped releases rose only 20% and end-user d…
A new edition of Nicolaus Copernicus's work on monetary theory has been released. A study published in *Management Science* found that faster replies to job applications significantly increase a candi…
A new NBER study finds that data center expansion in U.S. counties boosts total employment, construction jobs, house prices, and electricity prices. Researchers used facility-level data and shift-shar…
California's 2019 law requiring high schools to start no earlier than 8:30 a.m. led to students sleeping roughly 46 more minutes per night, with the share of teens awake before 6 a.m. falling 86%, acc…