International Scientific Collaboration

Over the past half-century, science has shifted from isolated labs to **global networks** of co-authors spanning continents. In OECD countries, the share of internationally co‑authored papers jumped from about 2% in 1970 to 27% in 2023, showing how central cross-border work has become to modern research.

This rise has boosted impact: studies repeatedly find that internationally co‑authored papers receive more citations and tend to tackle more complex questions. But recent data suggest the growth curve is flattening, as politics and security concerns begin to slow the expansion of global teams.

Governments are rapidly expanding **research security** rules to protect sensitive technologies and limit foreign interference. OECD countries report roughly ten times more such policies in 2025 than in 2018, signaling a clear policy shift toward securitizing science.

Concrete measures include tighter grant rules, database access bans and new vetting of partners. The U.S. National Institutes of Health, for example, is ending nested foreign subawards and directly restricting access for researchers in several ‘countries of concern,’ disrupting ongoing collaborations. At the same time, academic boycotts and institutional bans are altering partnership patterns, especially in politically sensitive regions.

Collaboration patterns are moving from **bilateral** (two-country) projects toward **multilateral** networks that connect many nations at once. Countries such as the U.K. and Germany now conduct most of their international work through multilateral teams, while the U.S. and China are slowly shifting in that direction.

Meanwhile, U.S.–China scientific collaboration—once the world’s most dynamic pair—shows signs of decoupling under strategic rivalry and ‘scientific nationalism.’ China is deepening ties with partners across Africa, the Middle East and Asia, where co‑authored output is growing faster than regional research overall, signaling a diversification of global knowledge hubs.

Despite tensions, many of today’s biggest challenges—climate change, pandemics, food security and AI safety—simply cannot be solved within national borders. COVID‑19 illustrated this reality when rapid sharing of the SARS‑CoV‑2 genome enabled worldwide vaccine efforts within weeks.

Large international consortia, sometimes called **big‑team science**, are now standard in fields like neuroscience, climate science and particle physics. These projects pool massive datasets, specialized skills and expensive infrastructure, showing how collaboration is not just idealistic but practically necessary for cutting‑edge discovery.

Looking ahead, experts foresee a tug‑of‑war between fragmentation and **“smart openness”**—a model that protects truly sensitive areas while keeping most research broadly collaborative. Open science practices, preprints and cloud-based labs lower entry barriers and make cross-border work easier, even when travel or politics get in the way.

If funding keeps pace and security rules are applied narrowly rather than broadly, large multilateral networks may become the gold standard, especially for data‑intensive and AI‑driven research. But if nationalism and restrictions continue to spread, the world could see parallel, less connected scientific blocs—slowing progress on problems that affect everyone.