For decades, the United States benefited from one of the most powerful competitive advantages in science: many of the world’s best researchers wanted to work there.
That advantage is not disappearing overnight. The U.S. still has unmatched concentrations of research universities, federal laboratories, private capital and technology companies. But the flow of talent is becoming less one-directional, and Canada is making an unusually aggressive attempt to capitalize on the shift.
The latest example comes from the University of Toronto, which has recruited six prominent researchers currently working at U.S. institutions, including MIT, the National Institutes of Health (NIH), University of California San Francisco (UCSF), Ohio State University, Boston University and Washington University School of Medicine in St. Louis.
The six are part of the new Eddie Goldenberg Research Chairs of Canada program, the centrepiece of Ottawa’s C$1.7 billion Canada Global Impact+ Research Talent Initiative. The broader initiative is designed to attract and support more than 1,000 international and expatriate researchers over 12 years, with C$1 billion specifically allocated to research chairs.
What makes the first group particularly notable is where the talent is coming from. Of the first 64 scholars recruited through the research-chair program, 48 are currently based in the United States.
This is beginning to look less like ordinary academic recruiting and more like a geopolitical competition for intellectual capital.
Why the U.S. Is Suddenly More Vulnerable to a Research Brain Drain
The timing is difficult to separate from the upheaval that has swept through American science since President Donald Trump returned to office in January 2025.
A January 2026 analysis by Nature counted more than 7,800 U.S. research grants that had been terminated or frozen, along with roughly 25,000 scientists and other personnel leaving federal agencies involved in research. The administration also proposed substantial reductions to federal science budgets, including cuts approaching 40% at the NIH, 57% at the National Science Foundation (NSF) and 47% at NASA’s Science Mission Directorate. Not all of those proposed reductions ultimately became law, and some administration actions have faced opposition from Congress and the courts, but the uncertainty itself has affected laboratories, hiring and long-term research planning.
The administration’s position is more nuanced than simply abandoning science. As Unite.AI recently examined, the White House is attempting to restructure approximately $200 billion in annual federal research spending. Its proposals favour more funding that follows individual scientists, faster grant mechanisms, high-risk research, artificial intelligence-driven discovery and strategic technologies such as AI and quantum computing. The administration argues that the existing university-centred funding system has become bureaucratic and insufficiently focused on national priorities.
But restructuring a research ecosystem also creates winners and losers. Grant cancellations, federal workforce reductions and changes to what types of research receive support have made career planning considerably less predictable for many scientists.
The warning signs appeared quickly. In March 2025, three-quarters of the more than 1,600 respondents to a Nature poll said they were considering leaving the United States, with Canada and Europe among the most frequently cited destinations. The poll was self-selecting and should not be interpreted as representative of all American scientists, but subsequent job-search data also showed growing interest in positions outside the country.
Canada appears to have decided that uncertainty in one country can become opportunity in another.
Canada Is Building a Recruitment Machine, Not Just Offering a Few Professorships
Ottawa’s strategy goes beyond paying senior scientists to relocate.
The C$1.7 billion initiative includes research chairs, infrastructure funding, early-career positions and doctoral and postdoctoral awards. In May, the government offered 659 research training awards to candidates from 72 countries, while the Canada Impact+ Emerging Leaders program is designed to help universities recruit promising researchers much earlier in their careers.
Canada has also begun aligning immigration policy with the strategy, creating new pathways for researchers and promising expedited processing for some of the people being recruited.
The approach matters because elite researchers rarely move alone. A senior scientist can bring postdoctoral fellows, graduate students, collaborators and future faculty recruits. Their laboratories can attract industry partnerships, generate intellectual property and ultimately produce startups.
That multiplier effect is what makes U of T’s six recruits worth examining individually.
Adrienne Campbell-Washburn: Rethinking What an MRI Machine Can Be
Adrienne Campbell-Washburn is moving from the NIH’s National Heart, Lung, and Blood Institute, where she became a senior investigator after originally joining as a postdoctoral fellow in 2013.
Her work challenges a basic assumption in magnetic resonance imaging: that stronger magnets are necessarily better.
Campbell-Washburn’s laboratory helped pioneer high-performance 0.55-tesla low-field MRI, combining weaker magnetic fields with modern computing, acquisition and reconstruction techniques. The approach can provide advantages when imaging areas such as the lungs, which traditionally present difficulties for MRI, while reducing problems associated with metal devices and potentially lowering the cost and infrastructure requirements of scanners.
At Sunnybrook Research Institute and U of T’s Department of Medical Biophysics, she plans to push this further through lower-cost MRI systems, portable scanners and AI-enabled imaging techniques.
The broader implication is significant. MRI is extraordinarily powerful, but large conventional scanners are expensive and concentrated in major hospitals. If lower-field machines combined with increasingly sophisticated AI reconstruction can produce clinically useful images with cheaper hardware, MRI could eventually become available in far more settings, including smaller hospitals, community clinics and potentially bedside environments.
Tanya Berger-Wolf: Using AI to Understand the Natural World
Tanya Berger-Wolf arrives from Ohio State University with an unusual combination of expertise spanning theoretical computer science, artificial intelligence, ecology and conservation.
She is one of the pioneers of what has become known as imageomics, using machine learning to extract biological information from the enormous quantities of images being collected across the natural world.
Berger-Wolf was a founding contributor to Wildbook, an open-source platform that can identify individual animals from photographs. Similar to how facial recognition distinguishes humans, these systems can recognize distinctive markings on animals such as zebras and whale sharks, allowing researchers to monitor populations and behaviour without relying exclusively on physical tags or GPS collars.
Her Imageomics Institute has since pushed the concept considerably further. Its BioCLIP model was trained on more than 210 million images covering roughly one million species, allowing AI to identify organisms across a substantial portion of the known tree of life.
At U of T, Berger-Wolf intends to combine imagery with audio, genetic information and sensor data to build richer models of ecosystems and how they respond to climate and human activity.
The project illustrates an increasingly important direction for AI: foundation models are moving beyond text, images and consumer applications into specialized scientific domains where enormous datasets are effectively impossible for humans to analyze manually.
Shu (Joy) Jiang: Moving Medical AI From Algorithms Into Hospitals
Shu (Joy) Jiang is joining U of T from Washington University School of Medicine in St. Louis, where her research combines biostatistics, cancer prediction and artificial intelligence.
One of her most notable projects already demonstrates how academic AI research can cross into commercial medicine.
Jiang and epidemiologist Graham Colditz co-founded Prognosia, which developed an AI system that analyzes mammograms to predict a woman’s five-year risk of developing breast cancer. The software received FDA Breakthrough Device designation in 2025, and Prognosia was subsequently acquired by South Korean medical AI company Lunit.
Rather than simply detecting an existing tumour, the technology looks for patterns in mammograms that can indicate future risk, including signals that may not be visible to human clinicians.
At U of T’s Dalla Lana School of Public Health, Jiang will tackle another problem that may prove just as important as developing the algorithms themselves: getting validated medical AI into real clinical environments.
Her planned work involves building infrastructure connecting hospitals, regulators and industry so medical imaging AI can be evaluated and deployed across Canada.
That addresses one of healthcare AI’s largest bottlenecks. The industry already has an expanding supply of promising models. The harder challenge is proving that they work consistently across hospitals, patient populations and imaging equipment, and then integrating them into workflows where clinicians can actually use them.
Michelle Arkin: Going After Medicine’s “Undruggable” Targets
Michelle Arkin brings considerably more than an academic drug-discovery background to U of T Mississauga.
At UCSF she has served as a professor of pharmaceutical chemistry, executive director of the Small Molecule Discovery Center and vice dean for research technology and entrepreneurship. Earlier in her career at Sunesis Pharmaceuticals, she worked on some of the first potent inhibitors targeting protein-protein interactions.
One compound emerging from that work was lifitegrast, which ultimately became an FDA-approved treatment for dry-eye disease.
Her research now focuses heavily on biological targets historically considered difficult or impossible to drug, including protein-protein interactions, molecular machines and transcription factors. She is also active in Alzheimer’s research and has co-founded several therapeutics companies.
At U of T, Arkin will investigate areas including molecular glues, precision medicine and AI-enabled drug discovery.
Molecular glues are particularly interesting because they work differently from conventional medicines. Instead of merely blocking a protein, they can force biological molecules into interactions that would not ordinarily occur, potentially allowing the cell to eliminate disease-causing proteins.
Combined with AI systems capable of predicting molecular interactions and searching enormous chemical spaces, approaches like these could expand the fraction of human biology that can realistically be targeted by drugs.
Raymond Fisman: Studying What Happens When Money Meets Political Power
Not all of the recruits work in laboratory science.
Economist Raymond Fisman joins U of T’s Rotman School of Management from Boston University, bringing a research program focused on corruption, political influence, governance and the relationship between institutions and economic behaviour.
One of his best-known studies used an unusual dataset: parking violations committed by United Nations diplomats in New York City.
Because diplomats enjoyed immunity from enforcement, Fisman and economist Edward Miguel were able to examine how officials behaved when conventional legal consequences were largely absent. Their research found strong differences correlated with corruption norms in diplomats’ home countries, while later enforcement measures sharply reduced violations.
More recent work has examined political donations, government transparency and how election incentives can affect responses to freedom-of-information requests. A 2026 paper found that agencies in higher-corruption environments were more likely to reject or delay requests ahead of elections.
At Rotman, Fisman will investigate how wealthy individuals and organized interests influence public policy, why voters sometimes tolerate evidence of corruption and how governments can improve accountability.
His inclusion in the program is important. Canada’s research strategy is not limited to technologies that can immediately become products. It also recognizes that the institutions governing technology, markets and public trust can be as consequential as the technologies themselves.
Sara Seager: Bringing the Search for Other Worlds Back to Toronto
Sara Seager is arguably the best-known name among the six.
The Toronto native graduated from U of T in 1994 before earning her PhD at Harvard and eventually becoming a professor at MIT. She went on to become one of the pioneers of exoplanet science, particularly the study of planetary atmospheres and the search for gases that might reveal whether life exists beyond Earth.
Her career has included a MacArthur Fellowship, the Kavli Prize in Astrophysics and appointment as an Officer of the Order of Canada.
Her return to Toronto had actually been announced before the latest federal chair program results. She is scheduled to join U of T on September 1, 2026, making her case somewhat different from researchers recruited specifically after the federal initiative was launched.
At U of T she plans to combine astronomy, chemistry, computational physics and aerospace engineering while continuing to pursue the search for habitable worlds. She is also leading the Morning Star Missions to Venus, which aim to directly investigate the planet’s atmosphere for organic molecules.
The new chair expands that work into planetary and molecular sensing technologies, including instruments capable of detecting chemicals at extremely low concentrations and AI models for interpreting complex planetary and environmental systems.
Those technologies do not have to remain in space science. Ultra-sensitive molecular sensing could eventually have applications in environmental monitoring, industrial detection and human health.
The Bigger Prize Is the Ecosystem That Forms Around the Researchers
Canada should be careful not to declare victory too early.
The United States remains the world’s dominant research ecosystem by most measures, and it possesses advantages Canada cannot reproduce simply by writing larger recruitment cheques. Its universities, national laboratories, venture capital industry, pharmaceutical companies, hyperscalers and defence research infrastructure create an enormous gravitational pull for scientific talent.
The Trump administration is also not uniformly reducing support for advanced technology. AI, quantum computing, defence technologies and AI-enabled scientific research are among the areas it explicitly wants to prioritize.
But research leadership can shift at the margins long before it shows up in national rankings.
A researcher who relocates today may recruit ten students tomorrow. Those students may train another generation of scientists. A laboratory may generate a patent, which becomes a startup, which attracts engineers and investors, which eventually creates an industry cluster.
Toronto has seen that dynamic before. Geoffrey Hinton moved from Carnegie Mellon University to U of T in 1987, at a time when neural networks were far from fashionable. Decades later, his work helped establish the intellectual foundations of the deep-learning revolution and contributed to Toronto becoming one of the world’s major AI research centres.
Canada is now attempting to reproduce that kind of long-term compounding effect across AI, biotechnology, medicine, climate science, quantum technology and other strategic fields. It fits into a broader effort to convert the country’s academic strengths into economic capacity, including the national AI strategy Canada launched earlier this year. That strategy similarly emphasizes keeping more research, companies and intellectual property inside Canada.
Canada Still Has to Prove It Can Keep the Talent It Attracts
There is also a legitimate criticism of the strategy.
Canada cannot build a sustainable research ecosystem entirely through superstar recruitment. Existing Canadian researchers have long complained about limited grant funding, salaries that struggle to compete with the United States and insufficient opportunities for young scientists. Critics have questioned whether heavily funded international chairs could create a two-tier system in which newly recruited researchers receive extraordinary packages while equally capable scientists already working in Canada face tighter resources.
That makes the early-career, infrastructure and training components of the C$1.7 billion initiative particularly important.
If Canada simply imports famous scientists, the program will produce impressive announcements. If those scientists receive the resources to build laboratories, recruit younger researchers, collaborate with industry, commercialize discoveries and create institutions that survive beyond their individual careers, the consequences could be much larger.
For much of the last century, the United States understood something fundamental about innovation: attracting exceptional people was itself a form of economic policy.
Canada appears to have learned the lesson.
The question now is whether the United States is temporarily giving its neighbour an opening, or whether the direction of scientific talent across the border is beginning to change in a more lasting way.
