Rideau River, Canada

Randomized Experiments in the Science of Science: Lessons from Peer Review and the Evaluation of New Knowledge, @nberpubs chapter papers.ssrn.com/sol3/papers.… An analytic synthesis of all journal-published RCTs in #ScientificEvaluation and #PeerReview.

Great exercise and apparently the answer is ‘no.’ Which is interesting given the magnitude of investments taking place. nber.org/papers/w34243

Still the most undervalued AI company

In retrospect it is surprising that OpenAI released o1-preview. As soon as they showed off reasoning, everyone copied it immediately. And if they had held off releasing a reasoning/planning model until o3 (& called that GPT-5) it would have been a startling leap in AI abilities.

🚨New WP🚨 I used to think AI would replace workers. Studying a database with the near universe of commercial AI (including GenAI factory, and office tools) I found the opposite: AI increases employment by giving low-skilled workers expert knowledge.🧵

💥 I’m starting something new inside OpenAI! It’s called OpenAI for Science, and the goal is to build the next great scientific instrument: an AI-powered platform that accelerates scientific discovery.

Bana, S., Boudreau, K. (2025) — AI-ENABLED JOB MARKETS & MARKET PARTICIPATION: A FIELD EXPERIMENT ON HOW AI SHAPES JOBSEEKERS’ EXPECTATIONS OF COMPETITION Field experiment showing that use and disclosure of AI reduces jobseeker participation by about 25%. lnkd.in/dsERsfy3

Best analogy for @karpathy Vibe Coding. Is watching @justinbieber conduct an orchestra. With no sheet music, no conducting training—just vibes, hums and vision. And it's the future of how creators and designers will work. We will think in ''concept level'', orchestrating AI to bring ideas to life. @antonosika on the @20vcFund Listening, iterating, and flowing with it. Bieber is like master Vibe-Designer here. It's hypnotic and awesome to watch this tbh. 😎🥰 This is what 'text-to-x' paradigm will do for behavior. And why the @lovable of this era are creating. (sourced from his IG LIVE)

AI labs have managed to grab a significant piece of the profit generated by SaaS companies. Interesting piece by @mims: wsj.com/tech/ai/ai-costs-exp…

Call for papers: HEC Paris AI & Entrepreneurship Workshop (#3) Topics: Entrepreneurship & AI, and implications of AI for orgs & markets, entrepreneurship, … When: Dec 13 Where: HEC Paris Keynotes: @EmmaBrunskill @lugaricano Deadline: Sept 22 Info: sites.google.com/view/hecpar…

Is AI already impacting the job market? A new paper from me, @erikbryn, and @RuyuChen at @DigEconLab digs into data from ADP. We find some of the ***first large-scale evidence of employment declines for entry-level workers in AI-exposed jobs.*** A thread on our paper:

US BLS Labour Statistics Agency was Teetering even Before Trump ⁦@FT⁩ ft.com/content/2505af1e-dd7a…

Why is AI so slow to spread? Economics can explain Businesses are ignoring the street of hundred-dollar bills @TheEconomist economist.com/finance-and-ec…

#EconomicsofScience Are China’s universities really the best in the world? @TheEconomist shorturl.at/BQBzo

Kevin Boudreau, Lars Bo Jeppesen, Milan Miric "ARTIFICIAL INTELLIGENCE AS A PLATFORM TECHNOLOGY: STRATEGIC IMPLICATIONS OF COMPETING ON TOP OF AN AI PLATFORM". forthcoming chapter in HANDBOOK ON AI & STRATEGY eds. Felipe Csaszar and Nan Jia papers.ssrn.com/sol3/papers.…

Assessing the potential and pitfalls of using Natural Language Processing in social science research on innovation, from @a_bergeaud, Adam B. Jaffe, and @DimitrisPapan20 nber.org/papers/w33821

Between 50 to 60 percent of a scientist's research output is attributable to the institution where they work, and two thirds of this effect is driven by the presence of star researchers, from @amitabhchandra2 and Connie Xu nber.org/papers/w33996

Does professor quality matter for training the next generation of researchers? Or will good PhD students do good research either way? Quality matters

I wish data centers would offer tours to the public and schools could take field trips to them. They are the defining pieces of infrastructure of our generation, but unlike railroads, the grid, or anything else, we never get to see them and experience their scale.

Quite a useful reading list for AI, digital, and antitrust issues