UMD Among Top 5 Doctorate Producers in 5 Critical Tech Areas

The University of Maryland is a top five American producer of doctorates in several technology areas that are key to national security and science-driven innovation, according to a new study.

The working paper, released in June by the National Bureau of Economic Research, found that UMD ranked in the top five universities in the number of Ph.D. graduates trained in each of five critical areas from 2000 to 2022: quantum science, artificial intelligence (AI), space technology, networked sensing, and data privacy and cybersecurity. UMD sits alongside the Massachusetts Institute of Technology, Stanford University, Harvard University and the University of California, Berkeley in these rankings.

"This recognition highlights the University of Maryland’s leadership in preparing the next generation of scientists to innovate in an ever-changing world," said UMD President Darryll J. Pines. "Our top rankings in critical technology fields reflect the strength of our faculty and research enterprise to make meaningful impact in addressing the grand challenges of our time.”

According to the study’s authors—Dror Shvadron of the University of Toronto, Lee Fleming of the University of California, Berkeley, and Hansen Zhang and Daniel P. Gross of Duke University—“doctoral trainees are the feedstock of future science-driven innovation,” so it’s important to examine who enters science, to what effect and how early careers evolve.

They used AI, specifically, large language model-enabled natural language processing, to analyze 1.2 million Ph.D. dissertations completed between 1950 and 2022 in the natural sciences, math and engineering using information from ProQuest (a commercial provider of global dissertation data), OpenAlex (an open-access bibliometric database) and universities’ institutional repositories.

They then filtered the available theses or dissertations—75% in total, and roughly 96% since 2000—to include only those in 12 critical technology areas identified by the White House Office of Science and Technology Policy in 2024 as U.S. national priorities.

UMD ranked No. 4 on the list of the top 15 universities of Ph.D. graduates in AI in 2000-22, with 579.

It counted 258 quantum science dissertations at UMD, for the No. 5 spot, noting that Maryland, like other universities on the list, was linked with a quantum science center. The study named the Joint Quantum Institute, along with the Joint Center for Quantum Information and Computer Science, both longtime partnerships between UMD and the National Institute of Standards and Technology. (UMD now has 10 quantum research centers.) The study also tallied 474 dissertations in networked sensing (No. 4), 201 in space technology (No. 5), and 124 in data privacy and cybersecurity (No. 2).

UMD’s rankings reflect investments that the university and the state have made in turning College Park into the Capital of Quantum and a hub of computer science and AI research.

In January, Maryland Gov. Wes Moore announced the public-private Capital of Quantum initiative, which aims to pour $1 billion into efforts to recruit top quantum scientists and engineers, expand access to the National Quantum Laboratory, construct a new building for UMD’s Applied Research Laboratory for Intelligence and Security and support quantum startup development at UMD.

The university last year launched the Artificial Intelligence Interdisciplinary Institute at Maryland (AIM) as the epicenter of AI at UMD, supporting faculty research and course development with seed awards, creating partnerships among AI experts across schools and colleges, and coordinating efforts to give students in all majors the opportunity learn the principles of AI and how they apply to their field of study. The university recently launched an M.S. in artificial intelligence, while two bachelor’s degrees (one in the arts and one in the sciences) are currently under development.

The researchers write that the U.S. federal government is “by far the largest source of financial and in-kind support for STEM Ph.D. training in America,” providing support to 40% of doctoral trainees, and suggest that increasing that funding increases Ph.D. production roughly one-for-one.

They also hope that their methodology provides a useful alternative to large-scale surveys administered by the National Science Foundation like the Survey of Earned Doctorates and the Survey of Doctorate Recipients, which are “expensive to administer, have gaps in what they measure, and are difficult to link to other data sources.”