Large-Scale Study Finds Single-Country DURC Oversight Has Structural Limits

A peer-reviewed paper published Thursday in Science puts hard numbers behind one of biosecurity's most stubborn policy dilemmas: when one country tightens control over dual-use research of concern, the science migrates, the risk does not.

The study, "Dual-use research under scrutiny" (DOI: 10.1126/science.aee2479), is the work of Professor Seokbeom Kwon of KAIST's School of Business and Technology Management. According to a KAIST institutional release reviewed by EurekAlert, Kwon drew on roughly 600,000 research papers, cross-referencing patent and publication citation data to measure what happens to dual-use research output and downstream scientific impact when security oversight intensifies.

The dual-use research of concern category covers work with legitimate public-health value and meaningful misuse potential in the same experiment. <cite index="17-4,17-5">Dual-use research has both legitimate civilian applications -- such as vaccine and treatment development -- and potential security-sensitive applications, such as biological weapons or bioterrorism, with viral transmission mechanisms and pathogen behavior cited as representative examples.</cite>

The policy backdrop is not abstract. <cite index="18-5">Central to the recent intensification of oversight is Executive Order 14292, signed in May 2025, which heightened federal scrutiny over biological research with security implications, particularly gain-of-function studies.</cite> That executive order sits alongside a federal budget picture that CSIS described, in an analysis published two weeks ago, as pulling in the opposite direction: <cite index="7-1">the FY 2026 budget proposal threatens large cuts to biosecurity-related agencies, including the National Institutes of Health (-$18 billion), the Centers for Disease Control and Prevention (-$3.6 billion), and the Administration for Strategic Preparedness and Response (-$240 million).</cite>

Kwon's central finding, as summarized in the KAIST institutional release, is that domestic security policies enacted in isolation demonstrate fundamental structural constraints in their global efficacy. The mechanism is not difficult to reconstruct from open sources: <cite index="18-6">frameworks like National Security Decision Directive 189 apply primarily when federal funding is involved, leaving a regulatory blind spot for research conducted without direct U.S. governmental participation.</cite> Work that loses federal funding or falls outside federal jurisdiction does not disappear -- it moves to venues where the same oversight requirements do not apply.

<cite index="11-3">Strong regulation of research driven by concerns over national security threats could stifle even the core scientific advances humanity needs, according to the study.</cite> That is a genuine trade-off, not a rhetorical one. The finding should be read as an assessment of structural incentive effects, not as an argument against oversight -- and the paper does not make that argument. What it does, on the available evidence, is quantify costs that biosecurity policymakers have historically treated as unquantifiable.

The dataset Kwon used to reach those conclusions is methodologically notable. <cite index="18-2,18-3">The empirical study integrates patent and publication data spanning approximately 600,000 research papers to examine the interplay between scientific innovation and national security concerns in a globalized research environment.</cite> Using U.S. Patent and Trademark Office security-review records as a proxy for oversight exposure gives the analysis a paper trail that case-study approaches lack.

The study's authors note its relevance extends past biology. <cite index="19-5">The findings are expected to serve as a reference for discussions on research security regulation in advanced technology fields connected to security concerns, such as artificial intelligence and quantum technology.</cite> That framing is accurate but also worth scrutinizing: the dual-use problem in AI and quantum is structurally different from biology, where a containment failure can replicate autonomously. Readers should not import the paper's conclusions directly across those domains without accounting for that asymmetry.

The core policy implication -- that single-country controls alone cannot hold the line on globally distributed research -- is not new as a theoretical proposition. What is new is the scale of the empirical support. Whether that evidence changes the calculation for policymakers in Washington or Brussels will depend on whether the oversight conversation shifts from national compliance frameworks to the harder problem of coordinating standards across jurisdictions that have no shared enforcement mechanism.

Sources cited:
- EurekAlert (KAIST institutional release) (https://www.eurekalert.org/news-releases/1131017)
- Science journal (Kwon, "Dual-use research under scrutiny") (https://www.science.org/doi/10.1126/science.aee2479)
- BioEngineer.org (KAIST study summary) (https://bioengineer.org/kaist-study-delivers-first-comprehensive-large-scale-analysis-of-dual-use-research-and-security-oversight/)
- CSIS (AI-enabled bioterrorism policy analysis) (https://www.csis.org/analysis/opportunities-strengthen-us-biosecurity-ai-enabled-bioterrorism-what-policymakers-should)
- Seoul Economic Daily (Jang Hyung-im, AI-translated) (https://en.sedaily.com/technology/2026/06/05/kaist-tougher-research-curbs-over-security-fears-may-stifle)
- KAIST News Center (https://news.kaist.ac.kr/newsen/html/news/)