This webpage provides details about individuals affiliated with CITP whose research has recently garnered attention in the media.

November 2024

“Considerations for governing open foundation models” published in the journal Science, authored by Rishi Bommasani, Sayash Kapoor, Kevin Klyman, Shayne Longpre, Ashwin Ramaswami, Daniel Zhang, Marietje Schaake, Daniel E. Ho, Arvind Narayanan, and Percy Liang.

October 2024

CITP Director Arvind Narayanan was on the Wall Street Journal’s Tech News Briefing podcast. Listen to the conversation with WSJ Tech Columnist Christopher Mims “Is Generative AI All It’s Hyped Up to Be?” online.

Peter Henderson collaboration with Stanford University researchers to map racial covenants in Santa Clara County, California. Read about the project featured in SPIA’s Research Record. Read the full paper titled “AI for Scaling Legal Reform: Mapping and Redacting Racial Covenants in Santa Clara County” authored by Faiz Surani, Mirac Suzgun, Vyoma Raman, Christopher D. Manning, Peter Henderson, and Daniel E. Hois.

September 2024

Building Agents to Computationally Reproduce Scientific Papers
In “CORE-Bench: Fostering the Credibility of Published Research Through a Computational Reproducibility Agent Benchmark,” Zachary Siegel, Sayash Kapoor, Nitya Nagdir, Benedikt Ströbl, and Arvind Narayanan release a benchmark that measures the ability of AI agents to computationally reproduce existing scientific papers. The benchmark requires agents to set up the code to produce the results found in scientific papers and analyze the code output to answer questions about the results. Scientific research is facing a reproducibility crisis, and having agents reproduce existing work can automate a tedious yet essential research process.

August 2024

An interdisciplinary team of 19 researchers led by CITP Director and Computer Science Professor Arvind Narayanan and Computer Science Graduate Student Sayash Kapoor release guidelines for ethical machine learning in science. A paper detailing their guidelines appeared recently in the journal Science Advances. Read more about it here.

CITP Director Arvind Narayanan is quoted in the article: Everyone is judging AI by these tests. But experts say they’re close to meaningless (Cal Matters).

April 2024

CITP Associated Faculty Member Aleksandra Korolova was quoted in several publications recently regarding Meta’s IA:
–  Associated Press: Meta’s newest AI model beats some peers. But its amped-up AI agents are confusing Facebook user
–  Yahoo News and Sky News: Meta’s AI tells Facebook user it has disabled, gifted child in response to parent asking for advice
– The Washington Post: Meta AI expands to Facebook and Instagram. Misinformation may grow too
–  VOI – Claiming to Have Children, Meta AI Shocks Parents Group in New York 

March 2024

The Office of Engineering Communications details the March 18 testimony of CITP Executive Committee Member Andrew Appel before the Pennsylvania Senate Committee. The article is entitled “Princeton election expert warned legislators about a commonly used computer voting system.” Appel states “The continued use of computers to mark ballots on behalf of voters, as many counties do, is a “disaster waiting to happen.”

Former CITP Fellow and current OII Researcher Jakob Mökander, along with OII Professor Ralph Schroeder, recently published an article in the Social Science Computer Review entitled: Artificial Intelligence, Rationalization, and the Limits of Control in the Public Sector: The Case of Tax Policy Optimization.

CITP Executive Committee Member Andrew Appel will testify Monday March 18, 2024 before the Pennsylvania State Senate Committee on Government, at the invitation of the chair of the committee, PA State Senator Cris Dush. The hearing will be live-streamed on this web site from 10 a.m. – noon EDT. Also on that web site will appear a statement Monday morning, “Suggested principles for state statutes regarding ballot marking and vote tabulation” signed by Appel and 19 other election cybersecurity experts.

CITP Affiliated Faculty Member Molly Crockett along with coauthor Lisa Messeri published a paper in Nature on artificial intelligence and illusions of understanding in scientific research. Media coverage of this paper includes an editorial at Nature and an interview at Ars Technica. Professor Crockett will be speaking about this work at the CITP Seminar on April 9, 2024.

Graduate Student Sayash Kapoor and Professor Arvind Narayanan, along with other researchers, released a paper on the societal impact of open foundation models. This paper consists of work from 25 authors and 16 organizations. Media coverage of this paper includes Axios, Politico, and OSI.

The International Journal of Communication published a book forum on former CITP Fellow Eszter Hargittai‘s book Connected in Isolation. She received nine responses from her colleagues, and provided a direct response to the book forum’s contributions. Watch the book event discussion held during Hargittai’s time at CITP.

February 2024

CITP Associated Faculty Member Lydia Liu, along with Solon Barocus, Jon Kleinberg, and Karen Levy’s, research paper On the Actionability of Outcome Prediction is coming out in the AAAI conference this month. The work was also featured by the Montreal AI Ethics Institute (an international non-profit research institute with a mission to democratize AI ethics literacy) as a research summary.

Former CITP Emerging Scholar Christelle Tessono (2021-23) virtually testified in parliament on February 14, 2024. The testimony discussed the report she wrote while at CITP on Canada’s proposed AI and Data Act. You can view her testimony here.

CITP Fellow Kevin Munger‘s article in Mother Jones The Algorithm” Does Not Exist explores “From the fullness of our physical being we are reduced to mere “operators,” experiencing life primarily through the apparatus, which ‘programs’ both the producers and consumers of media.”

January 2024

In the current issue of Discovery, Research at Princeton 2023-24: AI hope versus hype, CITP Director Arvind Narayanan “dispels myths around artificial intelligence.”

Scientific America’s article Can this AI Tool Predict Your Death? Maybe, But Don’t Panic, Professor Matthew Salganik speaks to how “The study demonstrates an exciting new approach to predicting and analyzing the trajectory of people’s life” but “It’s hard to determine exactly how well it does relative to reality.”

FedScoop’s emerging tech article Proactive approach from White House, NIST needed for facial recognition technology talks with Ed Felten, CITP’s founding director and current committee member of the National Academies of Sciences, Engineering, and Medicine.