The Trump administration spent 2025 trying to gut American science. It proposed the deepest cuts to non-defense research in half a century. It froze or killed more than 7,800 grants. It drove more than 25,000 scientists and staff out of federal agencies. It tried to cap the overhead payments that keep university labs running. It shuttered NASA’s largest library and more than 100 labs at Goddard Space Flight Center.
Then Congress, controlled by Trump’s own party, quietly said no.
That is the strange, unfinished story of American science in the second Trump era: an administration that tried to break a system that has powered U.S. innovation since World War II, and a legislature that mostly refused to let it happen. But “mostly” is doing a lot of heavy lifting in that sentence. Because while Congress preserved the budgets, the human damage was already done, and it will take far longer than a budget cycle to repair.
What Trump tried to do
The scale of the proposed destruction was historic. In its fiscal year 2026 budget, the Trump administration requested a 35% cut to non-defense research and development, amounting to $32 billion. Individual agencies faced staggering reductions: the National Science Foundation by 57%, the NIH by 40%, the CDC by 53%, NASA’s science budget by 47%. After adjusting for inflation, these cuts would have reduced federal science spending to 1991 levels.
But the administration did not wait for Congress to approve the cuts. It started dismantling the system on its own. Officials began terminating already-funded grants at the NIH in February 2025, and later at the NSF. By year’s end, 5,844 NIH grants and 1,996 NSF grants had been cancelled or suspended. As of November 2025, the financial toll had already reached roughly $3 billion in remaining funds. The cuts disproportionately targeted research on health disparities, infectious diseases, vaccine hesitancyReluctance or refusal to vaccinate despite vaccine availability, driven by factors such as distrust, safety concerns, or complacency rather than lack of access., and climate science.
The administration also tried to impose a 15% cap on indirect cost reimbursementFederal funding paid to universities on top of direct research costs, covering shared expenses like lab space, utilities, and administration. Rates are individually negotiated and typically exceed 50%., the funds universities use to pay for lab space, equipment, utilities, and support staff. Negotiated rates typically average above 50%. The cap, by the NIH’s own estimate, would have stripped $4 billion annually from the research ecosystem and caused an estimated $16 billion in economic losses and 68,000 lost jobs.
What Congress did about it
Lawmakers in both chambers, including Republicans, rejected the president’s vision. The NIH was set to receive $48.7 billion, a $415 million increase over 2025. Congress boosted cancer research by $128 million, Alzheimer’s funding by $100 million, and ALS research by $15 million. It added language designed to prevent the administration from again attempting to cap indirect costs.
The NSF took a 3.4% cut instead of 57%. NASA’s science directorate got a 1.1% reduction instead of 47%. The Department of Energy’s Office of Science received a roughly 2% bump to $8.4 billion, not the 14% reduction the White House wanted.
“Congress has essentially rejected the president’s very dramatic cuts,” said Joanne Padron Carney, chief government relations officer for the American Association for the Advancement of Science. “In past years, we might not consider flat funding to be a success, but considering how we’re operating this past year, I think we’re quite pleased.”
Courts helped too. A federal judge ruled the NIH’s grant terminations “void and illegal” and “breathtakingly arbitrary and capricious.” An appeals court affirmed the block on the indirect cost cap. These legal victories preserved billions on paper.
Why the damage is already done
Here is the part that budget numbers cannot capture. A nationwide survey of nearly 1,000 NIH-funded researchers by STAT found that more than a quarter had laid off lab members. More than two in five had cancelled planned research. Two-thirds had counseled students to consider careers outside academia.
And the court victories? They look better on paper than in practice. Only 35% of researchers whose grants were cut or delayed said their funding had been fully restored by the end of 2025. Courts can order grants reinstated, but they cannot force the money to actually flow, or restore trust in a system that betrayed it.
“This is like the Titanic hitting the iceberg,” said Steve Shoptaw, who runs the Center for Behavioral and Addiction Medicine at UCLA, which shrank by 40% due to funding cuts. “People are still eating at the table, music’s still playing, and yet the ship is sinking.”
Federal science agencies lost about 20% of their staff in 2025. More than 10,000 doctorate-trained experts left the federal government. About half of the NIH’s 27 institute and center director positions remain unfilled. You can restore a budget line. You cannot restore institutional knowledge that walked out the door.
The brain drain is real
A Nature poll found that 75% of American scientists are considering leaving the country. Europe saw the opportunity and seized it. The European Union launched its “Choose Europe for Science” initiative, committing 500 million euros to attract researchers who no longer see a future in American labs. France separately pledged 100 million euros. Aix-Marseille University alone received nearly 300 applications from Americans for its “Safe Place for Science” program.
New international student enrollment at U.S. universities fell 17% from 2024 to 2025, the sharpest drop in a decade outside the pandemic year. Among schools reporting declines, 96% cited visa concerns as a contributing factor.
“It really is a blitzkrieg against science,” said Robert Proctor, a science historian at Stanford. U.S. investments in research, he said, are “really the seed corn of future knowledge.” When you eat the seed corn, nothing grows next year.
What this should tell us
The optimistic reading is that the system held. Congress protected the budgets. Courts blocked the worst executive overreach. American science survived its worst year since the McCarthy era.
The realistic reading is darker. An administration that controls executive agencies can inflict enormous damage without touching a budget line. It can freeze grants, fire staff, gut leadership, cancel programs, and create so much uncertainty that researchers stop applying, students stop enrolling, and the best minds start looking at plane tickets to Amsterdam. By the time Congress passes a corrective budget, the labs are already dark.
Federal research investment returns $5 to $20 for every dollar spent. It produced the internet, GPS, MRI machines, CRISPR, and the semiconductor industry. Nearly 2,000 members of the National Academies of Sciences, Engineering, and Medicine signed a letter warning of the threat to this legacy. They were not being hysterical. They were reading the budget.
Congress deserves credit for holding the line on funding. But that line is not enough. The real question is whether the U.S. can rebuild the human infrastructure it spent a year destroying: the early-career researchers who left for Europe, the lab technicians who were laid off, the graduate students who chose different careers, the international talent that now sees America as hostile territory. Money is the easier problem. Trust is the harder one.
The administration framed its cuts as fighting waste. The NIH said a “major reset was overdue” and blamed the Biden administration for prioritizing “ideological agendas over scientific rigor.” But when 81% of junior tenure-track scientists fear that funding disruptions could cost them tenure, and when 41% of all researchers have shifted their work to align with federal priorities, that is not a reset. That is a chill. And the chill will outlast whatever budget Congress passes next.
The Trump administration’s assault on the federal research enterprise in 2025-2026 represents the most aggressive attempt to restructure U.S. science policy since the creation of the modern grant-making system after World War II. It proceeded on multiple simultaneous fronts: direct grant terminations, proposed budget cuts, indirect cost cap attempts, agency staff reductions, and targeted defunding of disfavored research topics. Congress and the courts checked some of these efforts. Others inflicted damage that no appropriations bill can reverse.
This analysis examines the full scope of what happened, what was blocked, and what the structural consequences may be for American scientific capacity.
The multi-front assault
Grant terminations
Beginning in February 2025, the administration terminated or froze 5,844 NIH grants and 1,996 NSF grants, approximately 7,840 awards in total. By November 2025, the cuts totaled about $3 billion in remaining funds. The administration disproportionately cancelled projects related to diversity, equity, and inclusion; health disparities; vaccine hesitancyReluctance or refusal to vaccinate despite vaccine availability, driven by factors such as distrust, safety concerns, or complacency rather than lack of access.; infectious diseases; and climate science.
Grant recipients received form letters citing vague changes in “agency priorities” as justification. A federal judge called these explanations “void and illegal” and “breathtakingly arbitrary and capricious” under the Administrative Procedure Act. Judge Rita Lin in the Northern District of California found the terminations constituted “illegal viewpoint discriminationA First Amendment doctrine prohibiting the government from restricting or penalizing speech based solely on the viewpoint it expresses. under the First Amendment”, rejecting the government’s argument that grant allocation was a form of protected government speech.
Courts ordered thousands of grants reinstated, but implementation has been incomplete. A STAT survey of nearly 1,000 NIH-funded researchers found that only 35% of those whose grants were cut or delayed had their funding fully restored by the end of 2025. Among those without full restoration, a third reported losses between $100,000 and $500,000.
Proposed budget cuts
The FY 2026 budget request sought a 35% reduction to non-defense R&D, totaling $32 billion. Agency-specific proposals: NIH cut by 40% ($18 billion), NSF by 57%, CDC by 53%, NASA’s science directorate by 47%. Adjusted for inflation, this would have reduced spending to 1991 levels, an unprecedented reduction over the past half-century.
Indirect cost cap
In February 2025, the NIH announced a flat 15% indirect cost reimbursementFederal funding paid to universities on top of direct research costs, covering shared expenses like lab space, utilities, and administration. Rates are individually negotiated and typically exceed 50%. rate on all grants, replacing negotiated rates that averaged above 50%. This represented a fundamental restructuring of the cost-sharing model between the federal government and research institutions. The NIH’s own estimate projected $4 billion in annual savings, which translates directly to $4 billion in annual losses for universities.
Economic modeling estimated the policy would cause $16 billion in economic losses and 68,000 job losses annually. Multiple federal courts issued injunctions, and the First Circuit Court of Appeals unanimously affirmed that the NIH violated statutory law and its own regulatory procedures. Congress subsequently added language to the FY 2026 appropriations bill designed to prevent future indirect cost cap attempts.
Workforce reduction
Federal science agencies lost approximately 20% of their staff in 2025. The EPAEicosapentaenoic acid, a long-chain omega-3 fatty acid found primarily in marine sources. EPA reduces inflammation and is associated with cardiovascular benefits. and NASA were among the hardest hit. The FDA lost more than 20% of its workforce. Across science agencies, more than 25,000 people departed, many through a voluntary resignation incentive program, though others were fired in mass layoffs. More than 10,000 were doctorate-trained scientists.
At NASA’s Goddard Space Flight Center, 13 buildings and more than 100 science and engineering laboratories were slated for closure. The campus workforce was reduced from more than 10,000 to approximately 6,600. The center’s library, NASA’s largest, housing some 100,000 volumes, closed on January 2, 2026.
At the NIH itself, about half of the 27 institute and center director positions remain unfilled. Remaining staff report struggling to keep up with the workload of processing and approving research grants.
The congressional response
Both chambers, controlled by Republicans, rejected the administration’s proposed cuts in the FY 2026 appropriations process. Key outcomes:
- NIH: $48.7 billion, a $415 million increase over 2025. Cancer research boosted by $128 million, Alzheimer’s by $100 million, ALS by $15 million. Monthly reporting requirements added for grant awards, terminations, and cancellations.
- NSF: 3.4% cut, not 57%.
- NASA science: 1.1% cut, not 47%.
- DOE Office of Science: ~2% increase to $8.4 billion, not a 14% reduction.
Congress also included provisions to block future indirect cost cap attempts and to limit the administration’s practice of multi-year lump-sum grant funding, which had been used as an accounting mechanism to obscure reduced grant volumes.
“It illustrates that there’s still strong bipartisan support for the federal government playing a critical role in supporting research,” said Toby Smith, senior vice president at the Association of American Universities. “Yes, we’ve got the money now from Congress. Will they move it out the door? Will they have the staff to do that effectively?”
Downstream effects: the human cost
The STAT national survey provides the most granular picture of the damage at the lab level. Among NIH-funded researchers:
- 27% laid off lab members
- 42% cancelled planned research
- 47% paused experiments or studies
- 61% adjusted project timelines or milestones
- 66% counseled students to consider careers outside academia
- 81% of junior tenure-track scientists feared that disruptions could threaten their tenure
Researchers studying health disparities were disproportionately affected: 26% had grants terminated, 11 percentage points above the overall rate, and 68% shifted their work to topics aligned with federal priorities.
A clinical trial in Puerto Rico studying diabetes prevention lost participants who feared that their association with a Harvard-run project would draw government scrutiny. At the University of Washington, doctoral enrollment in medicine fell by a third as principal investigators could not guarantee continued grant funding.
The brain drain
A March 2025 Nature poll found 75% of American scientists considering leaving the country. Nearly 2,000 members of the National Academies of Sciences, Engineering, and Medicine signed a letter warning of existential threat.
The European Union responded with the “Choose Europe for Science” initiative, committing 500 million euros over 2025-2027 to attract researchers. France added 100 million euros separately. The European Research Council doubled its relocation grants from 1 million euros to 2 million euros. Aix-Marseille University received nearly 300 applications from Americans for its “Safe Place for Science” program.
International student enrollment dropped 17% in the 2025-2026 academic year, the largest decline in a decade outside the pandemic. Among institutions reporting decreases, 96% cited visa-related concerns. This pipeline constriction compounds the immediate workforce losses: fewer international PhD students means fewer researchers training in U.S. institutions, feeding back into reduced domestic capacity over a 5-10 year horizon.
Structural assessment
The federal government’s research funding model, established after WWII, operates on a simple principle: the government provides money, universities provide the talent and infrastructure, and the resulting innovations benefit the public. Economic analysis estimates this system returns $5 to $20 for every dollar invested.
The Trump administration’s 2025 actions attacked this model at every level simultaneously. Even with congressional and judicial correction, four structural vulnerabilities have been exposed:
1. Executive unilateralism. The administration demonstrated that it can inflict significant damage to the grant system through executive action alone, without congressional authorization. Grant freezes, terminations, staffing cuts, and agency reorganizations do not require legislation. Courts can intervene, but litigation takes months or years, during which researchers lose staff, pause experiments, and close labs.
2. The trust deficit. The multi-year grant model depends on researchers trusting that committed funding will actually arrive. That trust has been severely damaged. The NIH’s shift to lump-sum multi-year funding, while it moved money out the door, funded roughly 5-10% fewer projects. Researchers are now rationally hedging, applying for non-NIH funding, shrinking lab operations, and advising students against academic science careers.
3. Workforce irreversibility. It takes 8-12 years to train a PhD scientist. The 25,000+ personnel who left federal agencies, the early-career researchers pushed out, and the international students who did not enroll represent capacity that cannot be rebuilt on the timeline of a single budget cycle. Europe’s active recruitment efforts accelerate this loss.
4. The chilling effectThe deterrence of lawful behavior, such as research or free speech, caused by fear of punishment or uncertainty rather than direct prohibition. People self-censor to avoid potential consequences. on research direction. When 41% of researchers shift their work to align with perceived federal priorities, that is self-censorship operating at systemic scale. Research topics that the current administration disfavors, including health disparities, climate change, and infectious disease epidemiology, will see reduced attention regardless of their scientific merit or public health importance.
Congress held the budget line. That matters. But the budget is the floor, not the ceiling, of what a functional research enterprise requires. The ceiling is the confidence that the system will honor its commitments, the talent that chooses to stay, and the freedom to pursue knowledge without ideological vetting. On all three counts, American science ended 2025 diminished.
