SSRN removed 253 papers from one professor's account in October. Nicholas Polson teaches statistics at the University of Chicago's Booth School of Business. He posted nearly all 253 papers in nine months of 2026.
The platform hosts preprints, papers posted before a journal reviews them, free for anyone to read. It also tells authors to say if they used AI, and Polson skipped that question, The College Fix reported2.
Read the number one way and Polson cheated. A professor used a chatbot, posted hundreds of papers to one site, and got caught. That reading makes a tidy story: one bad actor, a quick suspension, a lesson learned.
Ivan Oransky doubts the tidy story. He co-founded Retraction Watch and teaches journalism at New York University. Auyon Siddiq, a UCLA professor, first flagged Polson's output in August, University Herald reported4. "The question needs to be: do these add anything of value and should we rely on them or not?" Oransky asked, University Herald reported4.
Colleagues called the papers strange and repetitive, Chicago Maroon reported1.
Oransky's question reaches past one professor. Elsevier's roughly 3,000 journals took in 4,200,000 papers in 2025. They published 795,000 of them, the publisher's own count shows7. Fewer than one in five papers clear the bar in an ordinary year, before anyone opens a chatbot.
Elsevier published fewer than one in five of the 4.2 million papers it received in 2025
Share of 2025 submissions to Elsevier's 3,000 journals that reached publication
Source: Elsevier [7]. Chart by The AILately.com Desk.
The numbers behind this chart
| Item | Value |
|---|---|
| Published | 18.9% |
| Whole | 100% |
Polson earned his Ph.D. at the University of Nottingham in 1988. He joined Booth's faculty in 1991 and published at an ordinary pace for decades. The College Fix counted 12 papers in 2016, 14 in 2017, and 10 in 20182. Then 2026 arrived, and the count jumped past 200 in nine months, plus 14 books.
Nicholas Polson's output jumps from 14 papers a year to 253 in nine months
Papers Polson wrote or co-wrote each year, 2016 to 2018, against his 2026 total before SSRN's removal
Each gray column is one year of Polson's prior output; the magenta column is his 2026 count, posted in nine months before SSRN pulled every one.
Sources: The College Fix [2]; Hoodline [3]. Chart by The AILately.com Desk.
The numbers behind this chart
| Item | Value |
|---|---|
| 2016 | 12 |
| 2017 | 14 |
| 2018 | 10 |
| 2026 | 253 |
Polson answered his critics in one line. His output, he said in an email, drew "jealousy and a desire to hijack my writing process and methods," AI Weekly reported5.
Vadim Sokolov defends the work. He teaches systems engineering at George Mason University and co-wrote more than 100 of Polson's 2026 papers. "I co authored some of them. They were not written by a bot," Sokolov wrote on the blog where the dispute played out, Economist Writing Every Day reported6.
Andrew Gelman takes the opposite position. He runs Columbia's statistics center and reads papers like Polson's for a living. "It's not just that a chatbot can produce slop, but that it enables the human production of slop," Gelman told The College Fix2.
Claudine Gartenberg watches the same strain from inside a journal. She edits papers at Organization Science, where volume has doubled since 2022. "There's going to be a flood of poor-quality research that the system is not equipped right now to handle," she said, Hoodline reported3. Peer reviewers work for free, so every new paper asks for an evening already spent.
The crowding predates AI tools by years. Gartenberg's own papers were already climbing before 2022, and the chatbot only added speed to a line already long.
Here is Rhoden's question, asked of a machine age: who profits, who pays, and who carries the risk. Polson profits either way, since his own tenure survived the takedown untouched. He kept both of his courses at Booth, including Bayes, AI and Deep Learning. Reviewers like Gartenberg pay in hours, reading a pile of papers that already strained before anyone drafted a sentence with a chatbot.
Job candidates pay hardest of all. A hiring committee that once trusted a paper count now must weigh which count is real. That doubt lands first on researchers with the thinnest records, while Polson's tenure keeps him clear of it.
The verdict is plain. A paper count once stood for effort; now it mainly measures speed, and a chatbot makes speed cheap. Research itself needs a record of what a study found and who checked it, apart from how fast someone posted it. Readers who want that record can find it every morning in The Signal, AI Lately's daily email that names the people behind the findings.
Sources
- Boris Archipov, "Tenured Booth Professor Criticized for Publishing 'AI Slop' Research Papers," Chicago Maroon, Oct. 12, 2026, https://chicagomaroon.com/53994/news/tenured-booth-professor-criticized-for-publishing-ai-slop-research-papers/
- Gabriel Sison, "Publisher removes 200 papers University of Chicago statistician wrote in under 1 year," The College Fix, Oct. 12, 2026, https://www.thecollegefix.com/publisher-removes-200-papers-university-of-chicago-statistician-wrote-in-under-1-year/
- Richard M. Sullivan, "UChicago Statistician's AI Papers Pulled From SSRN," Hoodline, Oct. 13, 2026, https://hoodline.com/2026/10/uchicago-professor-s-253-ai-written-papers-yanked-from-research-site/
- Madelaine Panganiban, "University of Chicago Professor's AI-Assisted Papers Spark Academic Debate," University Herald, Oct. 12, 2026, https://www.universityherald.com/articles/80513/20261012/university-chicago-professors-ai-assisted-papers-spark-academic-debate.htm
- Staff, "SSRN pulls 257 papers by U. Chicago's Nicholas Polson," AI Weekly, Sept. 24, 2026, https://aiweekly.co/alerts/ssrn-pulls-257-papers-by-u-chicagos-nicholas-polson
- Jeremy Horpedahl, "Nicholas Polson Has Written Over 200 Academic Papers in 2026 (so far)," Economist Writing Every Day, Aug. 26, 2026, https://economistwritingeveryday.com/2026/08/26/nicholas-polson-has-written-over-200-academic-papers-in-2026-so-far/
- Elsevier, "Fast facts about Elsevier," Elsevier, 2026, https://elsevier.shorthandstories.com/fast-facts/index.html