Sections

Commentary

Does scientific collaboration with China make America stronger or more vulnerable?

The flag of the People's Republic of China flies in the wind above the Consulate General of the People's Republic of China in San Francisco, California on July 23, 2020.
The flag of the People's Republic of China flies in the wind above the Consulate General of the People's Republic of China in San Francisco, California on July 23, 2020. The US Justice Department announced July 23, 2020 the indictments of four Chinese researchers it said lied about their ties to the People's Liberation Army, with one escaping arrest by taking refuge in the country's San Francisco consulate. (Philip Pacheco/AFP via Getty Images)

An Office of Management and Budget rule proposed to take effect this December would ban all funding for scientific research collaboration with China. Proponents of this rule argue it is necessary to protect American national security, including by limiting the risk of America falling victim to Chinese scientific espionage. Opponents of this rule argue that China is ahead of the United States in cutting-edge research in fields such as quantum science and biotechnology, and cutting American researchers off from scientific research in China will slow American discoveries.

In the following written debate, leading experts with differing views on the merits of becoming more open vs. closed to China in scientific discovery share their perspectives. While the contributors differ on whether collaboration with China strengthens American innovation or poses unacceptable security risks, they broadly agree on the importance of protecting sensitive research and attracting scientific talent. Their disagreements center on how best to safeguard American interests without undermining the country’s scientific competitiveness, including whether existing safeguards at universities and federal research funding agencies are effective.

For purposes of this exchange, the authors distinguish between “basic research,” which aims to advance theoretical knowledge and understanding and is generally intended for open publication, and “applied research,” which seeks to address practical problems and develop real-world applications, often involving proprietary technologies with commercial or military value. They refer to “sensitive fields” as those that have direct linkages to national security, such as cryptography and quantum computing, and “non-sensitive fields” as those that do not have any direct or indirect linkage to national security, such as paleontology and wildlife restoration. In the following debate, the contributors differ over where to draw these boundaries and what safeguards different types of research and collaboration warrant.

Question 1

Open or closed to China in scientific research

Will America become safer and more prosperous by becoming more open or closed to China in scientific research?

Kyle Chan

It is in America’s interest to continue collaborating with China on scientific research in nonsensitive fields. Closing ourselves off to working with Chinese researchers and Chinese research institutions would not hold China back from making scientific progress. It would only cause the United States to become more isolated from the global scientific community and risk falling behind.

There is an outdated view that the United States remains far ahead of China in science and has little to gain from working with Chinese researchers. But the reality is very different. China has emerged as a scientific powerhouse across a wide range of fields, from biology and medicine to physics and materials science. In the 2026 Nature Index, Chinese institutions make up over a third of the world’s top 500 academic research institutions. In areas such as renewable energy, nanotechnology, and genomics, China is at the global frontier of scientific research. In the first half of 2026 alone, global pharmaceutical firms signed $110 billion in licensing deals for access to Chinese biotech innovations.

Rather than seeing scientific collaboration with China as a one-way transfer of American knowledge and expertise, the United States must realize that this is a two-way exchange from which the United States has much to gain.

Moreover, scientific research is increasingly conducted through international networks involving researchers and institutions from multiple countries. Many American scientists participate in collaborations that include Chinese counterparts alongside researchers from Europe, Japan, and elsewhere. In 2024, 41% of American science and engineering publications involved international collaboration, with China taking the top place as America’s largest scientific research partner. Excluding Chinese institutions from these networks could backfire, leaving American researchers isolated while the rest of the world continues working together.

None of this means ignoring legitimate national security concerns. The United States already has restrictions on work with certain Chinese institutions and in sensitive fields, such as advanced semiconductors, quantum computing, and hypersonic flight technologies. But there is an enormous difference between protecting sensitive research and imposing blanket restrictions on scientific collaboration with China. America’s scientific strength has always rested on its openness to the world’s best ideas and talent.

Jacqueline Deal

This question should be reframed: How can the United States protect its eroding capacity for innovation in the face of Chinese Communist Party (CCP) policies to exploit the American research establishment for asymmetric, zero-sum gains? Specifically, the CCP has used access here to 1) accrue leverage by rendering the United States increasingly dependent on Chinese personnel; 2) bolster Chinese security; and 3) weaken the U.S. research and development base.

Welcoming STEM students from China for decades has left the United States reliant on Chinese nationals to staff its labs and conduct a substantial portion of its research. This dependence provides the CCP—which surveils and coerces Chinese citizens abroad while claiming even noncitizens’ allegiance—with leverage, and also reduces opportunities for foreign (not from the People’s Republic of China) and American students. Disproportionate Chinese representation in STEM fields has become a kind of self-fulfilling prophecy. U.S. schools argue that they need Chinese nationals, without considering the evidence that some PRC students have been entering partly through non-merit-based “talent program” channels. These channels include contracts compelling participants in CCP academic recruitment schemes “to train or recruit additional students to work for them in the United States,” as disclosed in a bipartisan 2019 Senate Committee on Homeland Security and Governmental Affairs investigative report.

Through such talent programs and other forms of collaboration between U.S.- and PRC-based scholars, U.S. R&D funding has directly benefited the Chinese security establishment. The U.S. House China Select Committee examined two years of publications on dual-use topics—supported by over $9 billion across approximately 700 Pentagon and 730 U.S. Department of Energy awards—that were coauthored by U.S.- and PRC-based researchers. In roughly half the cases, the PRC-based colleagues worked directly for China’s defense research industrial base. A follow-on study examined only coauthors at PRC defense labs and found at least $943.5 million in federal research grants shared since 2019.

To be sure, America benefits from hosting talented, hard-working foreign scholars. But the CCP regime exploits Chinese nationals’ access to expensive U.S.-sponsored fundamental research. By leveraging their know-how, Beijing can prioritize investments in both military and commercial applications. Over time, this division of labor reduces the revenue available to U.S. firms facing subsidized Chinese competitors. Lower profits for these firms, along with lower tax receipts for the U.S. government, in turn reduce the funds available for further investment in research. 

To recover and preserve its capacity for innovation, then, the United States must protect itself from Beijing’s predatory policies. Critical first steps include rebuilding a non-PRC STEM talent pipeline while denying China’s defense and security R&D base access to U.S.-based personnel and research. This would require passing the pending Securing Innovation and Research from Adversaries (SIRA) Act, which bans recipients of federal funds from engaging in collaboration or exchange unless a waiver is granted, with any high-risk entity designated by the Pentagon, Federal Communications Commission, Office of Management and Budget, Treasury, Commerce, State, or Homeland Security Department. It would also require outlawing all PRC-sponsored talent recruitment programs, such that no graduate student, postdoctoral fellow, faculty member, or visitor who is a selectee of such a program at any level could receive U.S. federal funds, and any visas issued to selectees would be rescinded.

At the same time, the U.S. government should increase scholarships for U.S. and trusted foreign students to study STEM. Funding for such study both in the United States and abroad could proceed via a new “trusted scholar” program for Five Eyes members, European partners, key Asian allies such as Japan and South Korea, and perhaps select additional countries that already boast or agree to adopt research security provisions aligned with those recommended here.

Yingyi Ma

Openness is the United States’ most unique competitive advantage in the world. Scientific collaboration with China serves the American national interest and will make America more prosperous.

The United States benefits from scientific collaboration with China, which has already surpassed it as the world’s largest investor in and producer of science and engineering research. Nature index figures show China ranking first in the share of high-quality natural science publications since 2023, with the United States second. Collaboration with a research power of that size makes the United States stronger: the more capable the partner, the greater the potential benefits.

Talent is vital here. America’s competitive edge lies in attracting the world’s best talent, and China remains the largest reservoir of that talent. China now produces more Ph.D.s than any other country, and by several accounts, more top AI talent as well. But years of suspicion toward Chinese scientists have weakened that attraction, to China’s benefit. The share of Chinese-origin top AI researchers working in China rose from 57% in 2022 to 69% in 2025. The United States is still attracting Chinese talent, but it is no longer the default destination.

There are real security concerns surrounding scientific collaboration with China, including but not limited to illicit technology transfer and dual-use research in sensitive fields. But efforts to shut down scientific collaboration with Chinese universities are aimed at the wrong target. Most federally funded U.S.-China collaboration at the university level produces basic research, the kind that becomes public knowledge accessible in peer-reviewed publications. U.S. policy itself has long recognized this distinction: “to the maximum extent possible, the products of fundamental research remain unrestricted.” The Justice Department’s own testimony illustrates that serious risks of technology theft extend well beyond universities, citing cases involving commercial aviation, wind turbines, and semiconductor technologies. These cases underscore the importance of protecting proprietary technologies rather than imposing sweeping restrictions on open academic research. Closing the door on university research does not fix that problem. It just gives up the openness that built America’s scientific lead in the first place.

Miles Yu

More openness to China in scientific research will make America less safe and will be detrimental to America’s prosperity.

America will be safer and more prosperous if we stop treating Chinese research institutions as ordinary academic partners and start recognizing the national security risks they pose. This doesn’t mean abandoning scientific openness. It means understanding that openness works only when both sides play by the same rules. With Communist China, they plainly don’t.

For decades, American universities have helped educate China’s scientific and technological elite. This year, roughly 254,000 Chinese students are attending American universities, while about 2,700 Americans are doing the same in China. According to the People’s Republic of China Ministry of Education, nearly 90% of those sent by China to study abroad have returned. The imbalance, roughly on the same scale for years, is staggering. Beijing has enjoyed extraordinary access to American laboratories, scientific expertise, and research networks, while American scholars face censorship, political restrictions, and limited access in China.

The problem isn’t Chinese scientists; many are brilliant researchers who have contributed enormously to American innovation. The problem is the system they operate under. Chinese universities answer to CCP authorities, and Beijing’s military-civil fusion strategy deliberately channels civilian scientific advances into military capabilities. Research in artificial intelligence, quantum computing, semiconductors, and biotechnology can just as easily strengthen the People’s Liberation Army (PLA).

We should also abandon the comforting illusion that scientific engagement will somehow liberalize China. Decades of cooperation have helped make China technologically stronger and militarily more formidable, but certainly not politically freer.

Washington should prohibit federally funded research partnerships with PLA-linked institutions, restrict sensitive dual-use cooperation, and require full disclosure of Chinese institutional ties, backed by meaningful penalties.

America should continue welcoming independent Chinese scientists who value academic freedom. But there is no reason American taxpayers should subsidize research that could help modernize the military of our principal strategic adversary.

Scientific openness is one of America’s greatest strengths. Letting the CCP exploit it against us is nothing but strategic folly.

Question 2

Screening for Chinese citizens

Would there be greater cost or benefit to strengthening lawful screening for Chinese citizens to study, conduct research, emigrate, and/or launch businesses in the United States?

Kyle Chan

One of America’s greatest competitive advantages is its ability to attract the best talent from around the world, including from China. Restricting the flow of talented Chinese students, researchers, and entrepreneurs to the United States risks undermining a core American strength—one that China has long envied as it seeks to create a globally competitive talent pipeline.

Chinese citizens working and studying in the United States contribute tremendously to American scientific and economic progress. In university and corporate research labs, Chinese graduate students and technical staff often perform the less glamorous but critical benchwork needed to make scientific breakthroughs. China sends more science and engineering Ph.D. students than any other country, and the vast majority—84%—choose to stay and continue contributing to the United States approximately 10 years after graduating. Some of the leading scientists currently advancing American capabilities in technologies such as batteries and robotics originally came from China. And now in AI, Chinese-trained talent is a major source of America’s competitive edge. A study by the Carnegie Endowment for International Peace found that over one-third of the top AI researchers working in the United States earned their undergraduate degrees in China.

Yet there are already signs that America is becoming less welcoming. A 2023 study found that 61% of surveyed Chinese-origin university faculty have thought about leaving the United States, while 45% had considered avoiding federal research grants. Policies intended to protect American science can backfire if they drive away the very researchers who contribute to its success. While there should always be proactive and robust efforts to safeguard sensitive research and prevent illicit technology transfer—such as access restrictions to certain national labs conducting sensitive research—these measures should be narrowly targeted at specific risks rather than broadly applied to researchers of Chinese origin.

The United States should distinguish between genuine national security threats and the overwhelming majority of Chinese students and researchers who come to America to study, work, and build their careers. In the intensifying competition with China over cutting-edge technology, it would be self-defeating for the United States to close off one of the very channels through which it maintains its technological lead.

Jacqueline Deal

It’s clear that existing policies, such as the National Science Foundation’s SECURE program, are not protecting the U.S. innovation base. Recent China Select Committee and independent investigations show that SECURE leaders are themselves collaborating with PRC entities executing Xi Jinping’s zero-sum “military-civil fusion” strategy to leverage foreign know-how for the CCP and the PLA. Enhanced protections are sorely needed: From 2016 to 2025, $6.8 billion in federal funding went to 1,687 researchers affiliated with institutions on the U.S. Commerce Department Entity List at the time of the award, and almost all of these institutions were in China. It’s morally and strategically wrong for American taxpayers to underwrite the technological advancement of Chinese human rights violators and the PLA. Heightened vetting for some visitors is only part of the solution; other necessary steps include better protecting Chinese nationals in the United States, increasing academic transparency requirements, and expanding prohibitions around interactions with high-risk China-based entities and their affiliates.

Given the CCP’s success in leveraging academic access here, lawful screening should be strengthened for Chinese nationals seeking to study or do research at the graduate or postdoctoral level in STEM fields. The United States should be less concerned about prospective English, history, or law students. 

At the same time, Washington should better protect Chinese nationals in the United States from Beijing’s efforts at extraterritorial jurisdiction. This would entail limiting the CCP’s ability to conduct transnational repression via consulates, liaison stations, and Chinese Students and Scholars Associations in the United States.

Additionally, the United States should enact common-sense research security and integrity measures such as expanding the Pentagon’s Section 1286 List to include all State Administration of Science, Technology and Industry for National Defense-administered (i.e., PLA-linked) universities and banning federal funding of collaborations with such China-based institutions or individuals tied to China’s defense and security R&D base. At the same time, U.S. universities should be required to disclose all the who, what, where, and when details of foreign gifts and contracts, and they should be prohibited from accepting funds, directly or indirectly, from any federally listed or sanctioned foreign entity.

Ideally, a new national center would be established to set government-wide standards and work with federal research partners to ensure their compliance while addressing challenges as they arise. A good start would be to implement draft National Science Foundation (NSF) regulations preventing collaborations with restricted entities, as well as passage and implementation of the aforementioned SIRA Act and talent program bans.

Yingyi Ma

Effective screening of Chinese citizens to study, conduct research, emigrate, or establish businesses in the United States requires distinguishing activities that pose fundamentally different security risks. Treating them as equivalent exemplifies the risk of over-securitization.

Consider international education. Most Chinese students pursue undergraduate or graduate education in fields with no apparent connection to national security. My research on Chinese international students finds that their education is primarily financed by their parents, many of whom are members of China’s rising middle class. They include small-business owners, teachers, doctors, and engineers. Others come from working-class families whose parents never attended college but were able to finance their children’s overseas education through the substantial wealth accumulated through rising property values during the first two decades of the 21st century. Beyond enriching campus intellectual life, tuition-paying Chinese students provide important revenue to universities. That revenue helps sustain academic programs and may indirectly support financial aid for domestic students. Treating these students as presumptive security threats based on nationality is discriminatory and defies America’s commitment to equal opportunity.

Research collaboration requires further differentiation. Basic research is defined as experimental or theoretical work undertaken to acquire new knowledge without a specific application. Because its findings are generally intended for open publication, it warrants a different level of scrutiny from applied research, which is directed toward a specific practical objective and often involves proprietary technologies with commercial or military value.

Immigration and entrepreneurship likewise deserve differentiated consideration. Establishing a business in a sensitive technology sector may warrant enhanced scrutiny of ownership, financing, and technology transfer. Outside of sensitive fields, visa and immigration decisions should rest on individual evidence, such as the specific research or business activities involved, rather than on nationality alone.

National security risks are real, and targeted screening is legitimate. For example, foreign nationals who work with controlled technologies in fields such as advanced semiconductors or quantum computing may require export-control licenses. But nationality-based suspicion is not only wrong, but ineffective, because it targets who they are rather than what they do. Excessive screening risks becoming a self-defeating policy by driving away the best and brightest Chinese talent.

Protecting national security and preserving academic openness are not contradictory. Effective policy requires distinguishing genuine security threats from the educational, scientific, and entrepreneurial exchanges that sustain American competitiveness.

Miles Yu

The benefits of stronger screening would far outweigh the costs. We should welcome talented people who want to study, work, and build their lives in America. But we would be foolish to ignore the fact that the CCP has systematically exploited our openness to acquire American technology, scientific expertise, and sensitive information.

This isn’t speculation. In January 2022, FBI Director Christopher Wray disclosed more than 2,000 investigations involving Chinese government efforts to steal American information and technology. Beijing’s military-civil fusion programs, overseas talent recruitment networks, and extensive control over Chinese research institutions make ordinary academic and immigration screening increasingly inadequate.

We need practical safeguards, not blanket discrimination.

For starters, Chinese scholars seeking research visas should be required to sign legally enforceable affidavits disclosing affiliations with Chinese government agencies, state-directed research institutions, military organizations, and sensitive dual-use projects. Deliberately concealing such connections should carry serious immigration and legal consequences.

We should also insist on genuine reciprocity. If American researchers cannot freely access Chinese laboratories, archives, and academic colleagues without political interference, why should researchers sponsored by those same Chinese institutions enjoy unrestricted privileges here? Reciprocity should be demonstrated, not merely promised.

American universities must disclose their Chinese institutional partnerships, funding arrangements, and intellectual property agreements. Investments, business acquisitions, and technology ventures involving sensitive sectors deserve equally rigorous scrutiny.

None of this should become an excuse to target people simply because they are Chinese. Many Chinese immigrants and scientists have made extraordinary contributions to America precisely because they reject the political restrictions imposed by the CCP. We should distinguish those individuals from people concealing obligations to Beijing’s military or intelligence apparatus.

Yes, stronger screening will impose some administrative costs. But compare those costs with the potential loss of critical technologies, compromised research, and the strengthening of the PLA. The choice isn’t difficult.

America should remain open to talent and opportunity, but closed to espionage, coercion, and technological exploitation. That isn’t discrimination but common sense.

Optional

Responses

All participants were given an opportunity to offer responses to the arguments made by other contributors.

Kyle Chan

While the threat of technology theft is real and should be taken seriously, it is important to understand the true scale and nature of this risk according to the data. When FBI Director Christopher Wray stated that there were over 2,000 investigations into Chinese government efforts to steal American technology, he did not specify how many of these investigations involved Chinese scientific researchers as suspects or how many resulted in criminal charges, let alone convictions.

An MIT Technology Review study of the Department of Justice’s China Initiative, a program from 2018 to 2022 aimed at combating Chinese economic espionage and technology theft, found only 19 of 77 cases involving alleged economic espionage or theft of trade secrets. Out of 23 cases brought against academic researchers, nearly all centered on failures to disclose ties to China, and only two researchers were accused of trying to obtain information or technology. For these two researchers, the charges were ultimately dropped.

What is harder to quantify—but arguably far more important—is the immeasurable benefit to American science and industry derived from collaboration with Chinese researchers. Now, with Chinese research institutions reaching the global frontier in many scientific fields, from renewable energy to medicine, the United States has even more to gain from these scientific partnerships.

The United States has adequate guardrails in place today, including targeted investigations on a case-by-case basis and special restrictions on research in sensitive fields. Rather than imposing sweeping restrictions on all collaboration with Chinese researchers, the United States should preserve the openness that has long been one of our greatest sources of scientific and technological strength.

Jacqueline Deal

Opponents of tighter restrictions on U.S. research with high-risk PRC partners argue: 1) existing restrictions suffice; 2) openness is a “two-way” exchange; and 3) Chinese scientists’ stay rates translate to U.S. benefit. I will address these in turn.

First, restrictions on fundamental research are scarce: the Wolf Amendment restricts only bilateral cooperation on NASA grants, and a provision under the National Defense Authorization Act bars Pentagon funding of institutions collaborating with high-risk PRC entities beginning in fiscal year 2026. Even if implemented, they would affect very little U.S. government-funded basic research. A recent study found that the NSF, not the Pentagon, provided 70% of the taxpayer dollars funding collaborations with Chinese defense labs.

Second, assertions of mutual benefits fail to quantify or compare them. As the China Select Committee wrote, to justify cooperating with the PLA or known human rights violators, a handwave that “American researchers ‘learn something too’” is insufficient. The question is whether we “got enough unique benefit to justify the transfer of knowledge, access, legitimacy, and relationship-building that the collaboration simultaneously provided to entities tied to China’s defense ecosystem.”  

No empirical study has analyzed the results of collaboration, quantitatively or qualitatively. What’s obvious is that over decades of virtually unrestricted exchange, the PRC has benefited both absolutely and relatively. 

Some win-win claims posit that the United States benefits because the PRC now leads in fundamental research funding, Ph.D. production, or citations. But raw numbers don’t capture quality. Citation gaming is common, particularly in China. As mentioned above, talent program investigations have also documented that a portion of Chinese researchers who end up in the United States qualify not due to merit but rather as beneficiaries of rigged access. This deprives other talented foreigners and Americans of places in U.S. institutions, compromising the non-PRC pipeline while fostering dependence on Chinese personnel.

Third, and finally, stay rates don’t prove friendship or net benefit. The CCP can pressure Chinese researchers anywhere in the world. As Cheryl Yu writes, Chinese guidance on overseas talent stresses the need to “focus not on location, but on contributions.” A 2019 U.S. Senate investigation pointed out that scientists who stay and become established are precisely those the CCP prioritizes for recruitment.

Yingyi Ma

We are not opposed to tighter restrictions on U.S. research collaboration with China, particularly when legitimate national security concerns are at stake. But we oppose the sweeping, indiscriminate prohibition on federal funding for scientific collaboration with Chinese institutions proposed in the Office of Management and Budget rule.

First, the proposed rule makes no meaningful distinction between sensitive and nonsensitive research. It would apply even to basic research, whether in physics, biology, sociology, economics, or any other discipline. A sociological study on educational inequality, for example, could face the same funding ban as a sensitive project involving advanced technologies with potential military applications.

Second, although the quantity of publications is not necessarily a reliable measure of scientific strength, the rise in the quality and influence of Chinese research is indisputable. According to the 2026 Nature Index rankings, based on 2025 publications in selected high-quality scientific journals, China substantially surpassed the United States in high-quality research output. Having a blanket ban on federal funding for collaboration with these research powerhouses is against the U.S. national interest.

Third, Chinese scientists and students have made enormous, measurable contributions to American science and innovation. Some argue that high stay rates do not necessarily demonstrate friendship toward the United States. That may be true, but the relevant question is not whether they are friends of the United States; it is how much they contribute to American innovation. Their contributions are particularly striking in artificial intelligence. A 2019 MacroPolo report found that nearly half of the world’s top AI talent received their undergraduate degrees from China. Of those researchers, 56% later chose to work in the United States. Among those who went to the United States for graduate school, 88% remained there to work. It is these top AI researchers that help the United States maintain its global AI leadership.

Authors

  • Footnotes
    1. Admissions began in the 1970s, as part of Washington’s late-Cold War rapprochement with Beijing to isolate Moscow, but Chinese students were granted extended visas in 2005, long after the Soviet collapse and as their numbers were starting to surge.
    2. The argument that the United States has grown dependent on nationals from the People’s Republic of China (PRC) reflects data on international students enrolled in STEM graduate degrees at U.S. institutions. For instance, the linked 2026 National Science Board study states that foreign nationals accounted for 42% of the science & engineering (S&E) master’s degrees, and 38% of S&E doctoral degrees awarded by U.S. institutions in 2024. According to the study, within S&E, temporary visa holders accounted for more than half of doctoral degrees in information sciences (61%), engineering (54%), and mathematics and statistics (52%). Further, among international students in S&E in 2024, China was the number one country of origin for doctoral degrees (nearly double the next highest country, India), and number two for master’s degrees behind India. The first such National Science Board report was published in 2020, and that report indicates that China has been a predominant source of foreign S&E Ph.D. students in the United States since the turn of the millennium: “In 2017, temporary visa holders earned one-third (34%) of S&E doctoral degrees, a relatively stable proportion over time. They account for half or more of the doctoral degrees awarded in engineering, mathematics and computer sciences, and economics. Three Asian countries—China, India, and South Korea—are the largest source countries and accounted for just over half (54%) of all international recipients of U.S. S&E research doctoral degrees since 2000.”

The Brookings Institution is committed to quality, independence, and impact.
We are supported by a diverse array of funders. In line with our values and policies, each Brookings publication represents the sole views of its author(s).