Short-Term Researcher Ties May Slip Past U.S. Disclosure Rules, Report Finds

A sweeping new analysis of research security across American federally funded science has uncovered a potential blind spot in the disclosure systems designed to protect the research enterprise: researchers who spend only a short time at U.S. institutions may carry prior or concurrent affiliations with restricted entities that never surface through existing screening processes. The report, authored by Dr Xueying (Shirley) Han, Head of Research Analytics at the research technology company Digital Science, finds that nearly half of the researchers whose records show both a restricted-entity affiliation and a U.S. research institution held the American appointment for 24 months or less. Of the researchers examined over a ten-year window, 14,549 held such short-term affiliations, a population that the report argues deserves particular attention as federal policy tightens.

The study, titled Research Security in U.S. Federally Funded Science: Evidence, Exposure, and a Risk-Calibrated Path Forward, is described as the first ecosystem-wide assessment of research security exposure across U.S. federally funded research. It arrives at a consequential moment. New funding prohibitions and restricted-party collaboration requirements are scheduled to take effect in 2026 and 2027, extending a period in which federal agencies have progressively sharpened the rules governing who may participate in federally supported research and under what conditions. Against that backdrop, the report’s central question is whether the machinery now being built can actually see the exposures that matter most.

To answer it, Dr Han assembled an unusually broad evidence base. Drawing on publication, funding, and researcher affiliation data from Digital Science’s Dimensions database, the analysis cross-referenced affiliations against restricted-entity lists maintained by U.S. and allied government agencies. The scope is formidable: more than 3 million publications acknowledging U.S. federal support, representing approximately 1.9 million researchers who stated U.S. affiliations, spanning the decade from 2016 to 2025. By linking the affiliation histories of individual researchers to the restricted-party designations published by government agencies, the study attempts something that individual institutions, working with only the information disclosed to them, generally cannot do on their own.

The technical problem the report identifies lies in how affiliation is recorded. During a U.S. appointment, a researcher’s affiliation of record is typically the host institution. That convention is sensible for bibliometric purposes, but it means that a visiting scholar’s prior or parallel ties to an entity on a restricted list may not be captured through the disclosure processes that universities and federal agencies currently rely upon. A researcher who arrives for a one-year or two-year appointment, publishes under the host institution’s banner, and departs, may leave behind a publication record that looks unremarkable in isolation. Only by tracing the researcher’s full affiliation history across systems and jurisdictions does the potential exposure become visible, and that is precisely the kind of longitudinal, cross-referenced view that most institutional screening workflows lack.

The scale of the short-term population is what gives the finding its force. Among researchers whose records show both a restricted-entity affiliation and a U.S. research institution, 48.2 percent held the U.S. affiliation for 24 months or less. In raw numbers, 14,549 researchers over the ten-year study period held short-term affiliations. The report does not claim that these individuals posed security risks; rather, it argues that they represent a structural gap. Disclosure frameworks, which tend to assume a stable, long-duration appointment at a home institution, may systematically under-detect ties among the very population, visiting scholars and other short-term researchers, whose movement patterns make continuous institutional oversight hardest to sustain.

Importantly, the report does not call for blanket restrictions on international collaboration. Dr Han’s proposed alternative is a risk-calibrated approach in which oversight scales with the sensitivity of the research and the assessed level of exposure. The report recommends that institutions assess research sensitivity and establish an explicit institutional risk tolerance; screen all researchers consistently and monitor actively for security risks or exposure; and apply cooling-off periods where warranted, based on that sensitivity and risk tolerance. The logic is that a uniform, heavy-handed regime applied to every visitor and every collaboration wastes institutional capacity on low-risk cases while potentially diverting attention from the small number of situations where genuine risk concentrates.

Dr Han frames the tension between openness and security as less stark than much of the public debate suggests. There is a tendency, she notes, to treat openness and security as opposites, but the data suggest that is not quite right. Addressing the oversight gap, in her view, does not require restricting collaboration; it requires institutions and agencies to know where to look and to act appropriately. That distinction matters for a research system whose competitiveness depends on the flow of international talent. Visiting scholars, postdoctoral researchers, and short-term appointees contribute substantially to American science, and policies that inadvertently discourage their participation could impose costs that outweigh any security benefit.

The broader findings of the study temper alarmist readings of the research security landscape. Overall exposure, the report finds, is modest and declining. The share of federally supported publications carrying a restricted-entity affiliation dropped 30 percent from 2018 to 2025, a decline that coincides with the period of intensifying federal attention to research security. Meanwhile, just 2 percent of researchers held both a U.S. and a restricted-entity affiliation in the same year. Those figures suggest that the headline risk is concentrated in a small fraction of the research population, which strengthens the case for targeted rather than universal measures. A policy apparatus calibrated to a 2 percent exposure rate, the report implies, should be designed to find that 2 percent efficiently, not to burden the other 98 percent.

Beyond the short-term affiliation gap, the research identifies additional structural weaknesses in current federal policy. There is, the report notes, no unified list of critical technology areas, leaving institutions and agencies without a common reference point for determining which fields warrant heightened scrutiny. Collaboration thresholds are inconsistent across the policy landscape, and one-size-fits-all mandates may fail to direct scrutiny toward the areas of greatest risk. Each of these deficiencies shares a common thread with the short-term affiliation finding: they reflect policies built on assumptions that do not match the observed structure of the research workforce, in which careers are international, affiliations are fluid, and risk is unevenly distributed across fields, institutions, and appointment types.

For the officials who must operationalize the next generation of research security rules, the report offers a data-driven starting point. Dr Han describes the assessment as giving research security officers, university leaders, and federal program managers data they can use to better understand where risks are concentrated and to focus attention where it is most warranted, an approach she argues can strengthen research security without treating every international collaboration as equally risky. As the 2026 and 2027 requirements come into force, the study’s central message is likely to shape the debate: the most effective security regime may be the one that looks hardest at the short, fluid, and easily overlooked affiliations that current disclosure systems were never designed to catch, while leaving the vast majority of international scientific exchange to proceed unimpeded.

Subject of Research: Research security exposure and short-term researcher affiliations in U.S. federally funded science

Article Title: Digital Science research security report finds short-term researcher ties may fall outside U.S. disclosure rules

Article References: Digital Science research security report finds short-term researcher ties may fall outside U.S. disclosure rules. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: research security, Digital Science, Dimensions database, restricted entities, short-term affiliations, disclosure rules, federally funded research, visiting scholars, risk-calibrated oversight, federal policy, international collaboration, Dr Xueying Han