What Happened to NSF
Two things happened at NSF that point in opposite directions, and both matter for anyone applying in 2026. Congress appropriated $8.75 billion for NSF in the FY2026 Commerce, Justice, Science spending bill, signed into law on January 23, 2026 as part of a minibus package (covered by the Congressional Research Service in report R48783). That figure is 3.4% below FY2024 levels, according to the American Astronomical Society's analysis of the enacted bill. But it explicitly rejects the White House budget request, which proposed cutting NSF to $3.9 billion, a reduction of roughly 55% that NSF's own projections said would drop funding rates to approximately 7%. The Senate Commerce Committee described the final appropriation as preventing an existential threat to U.S. science funding. What the budget did not protect is what happened inside the agency. Beginning in mid-April 2025, DOGE personnel arrived at NSF headquarters and initiated a review of the agency's grant portfolio. On April 18, 2025, NSF published a statement announcing terminations of grants deemed misaligned with administration priorities. By late May 2025, the agency had terminated 1,752 grants with a combined value of approximately $1.4 billion, according to the termination list documented by the Consortium of Social Science Associations. Director Sethuraman Panchanathan resigned on April 24, 2025, in the fifth year of his six-year appointment. Science magazine reported that White House orders to accept the 55% budget cut and eliminate half of NSF's approximately 1,700-person workforce may have precipitated his departure. Chief of Staff Brian Stone has been performing the duties of director since, with no permanent replacement announced as of February 2026. The agency also reduced new grant commitments even before the DOGE interventions. NSF funded approximately 8,800 new research project grants in FY2025, roughly 20% fewer than the approximately 11,000 funded in FY2024, a deliberate decision to manage future-year obligations given budget uncertainty. The practical result for applicants: appropriated money exists for new grants, but the agency that reviews proposals, manages awards, and runs programs has fewer people and less institutional continuity than it had 12 months ago. NSF acknowledged this by fundamentally changing its merit review process. For the parallel situation at NIH, where DOGE terminated more than 600 grants worth up to $8.2 billion, see our NIH grants guide. Search current NSF opportunities on Funding Landscape
What DOGE Terminated and Why It Matters
The terminations were not random. NSF's April 18, 2025 statement specifically cited diversity, equity, and inclusion research and studies related to misinformation and disinformation as categories being cut. The breakdown by directorate, based on NSF's published termination list, reveals how concentrated the damage was. The STEM Education directorate (EDU) lost 839 grants totaling $888 million, accounting for nearly two-thirds of the total terminated dollar value. The Social, Behavioral, and Economic Sciences directorate (SBE) lost 320 grants totaling $91 million. Together, these two directorates accounted for roughly 66% of all terminated grants by count. Nature reported that hundreds of additional grants were terminated after Director Panchanathan's departure, suggesting the final scope exceeds the 1,752 count from the May 2025 list. What this means for current applicants depends on your field. Physical sciences, engineering, computer science, and geosciences experienced fewer terminations and continue to fund research in their traditional areas. Researchers in these directorates face the indirect effects of reduced staffing and review capacity, but lower direct termination risk. STEM education and social sciences took the heaviest losses. Researchers in these areas should check NSF's Updates on NSF Priorities page for the agency's current position on what it funds, and should consult program officers before investing time in proposals. Interdisciplinary researchers whose work could be framed in multiple ways should understand what the agency currently prioritizes. This is not about disguising research. It is about connecting genuine scientific work to the problems the agency is funded and directed to solve. The terminated grants were active awards. Researchers whose mid-stream funding was cut lost both remaining budgets and the ability to complete ongoing work. Some have filed legal challenges, but no broad restoration had occurred as of February 2026. Search for open NSF research grants
🔍 Search related opportunities now
The Merit Review Overhaul
NSF made the most significant changes to its peer review process in years, formalized through PAPPG 24-1 Supplement 1 (designated NSF 26-200), effective December 15, 2025. Before December 2025, proposals required a minimum of three external reviews. Panel discussions were standard practice. Panel summaries provided detailed synthesis of reviewer assessments and often ran several paragraphs. Program officers recommended awards based on panel consensus. After December 2025, the minimum dropped to two reviews, and one of those can be conducted internally by NSF staff rather than an external expert. Panel discussions are optional rather than required. Panel summaries are limited to three to five sentences. Program officers have substantially more individual discretion in award decisions. NSF cited a proposal review backlog following the November 2024 government shutdown and significant workforce reductions as reasons for the changes. Some directorates lost up to one-third of their program officers through buyouts, early retirements, and departures during 2025. The previous process, which required coordinating three external reviewers for every proposal and running formal panels, was unsustainable with a smaller workforce. What this means for applicants: Less external feedback on declined proposals. The three-to-five-sentence summary limit means substantially less guidance for resubmission than previous cycles provided. Keep detailed notes about your proposal decisions during writing so you can diagnose problems yourself if the proposal is not funded. More variation between programs. With panel discussions optional and program officer discretion increased, the review experience will differ across directorates and programs. Two proposals of similar quality may receive different levels of scrutiny depending on how individual programs implement the changes. Program officer relationships matter more. When one of two reviews can be internal, the program officer's understanding of your work and field becomes a larger factor in outcomes. Contacting the program officer before submitting was always good advice. In the current environment, it materially affects your odds. Faster decisions are possible but not guaranteed. Fewer required reviews and optional panels could shorten turnaround times, though this depends on staffing levels in individual programs.
Policy and Compliance Requirements for 2026
PAPPG 24-1 Supplement 1 (NSF 26-200), effective December 8, 2025, introduced requirements that apply to all submissions. These are compliance items independent of the merit review changes, and missing any of them can result in administrative rejection. Research Security Training: All senior and key personnel must complete NSF's research security training modules before submission. This requirement originates from the CHIPS and Science Act of 2022 (Public Law 117-167) and took effect October 10, 2025. Training is available through NSF systems. Failure to complete it before submission results in rejection without review. Equipment Threshold Change: Federal property identification now applies to equipment costing $10,000 or more, up from $5,000. This affects how you track and report equipment purchased with NSF funds. Data Sharing at Publication: All data supporting NSF-funded publications must be shared at the time of publication. Exceptions require explicit documentation in your data management plan. Update your data management plan before submitting. AI and Research Misconduct: The guidance now explicitly addresses fabrication, falsification, or plagiarism through AI-based tools in research misconduct definitions. Document AI tool use carefully and verify all AI-generated content. Foreign Financial Support Reporting: Higher education institutions must report annually on foreign financial support of $50,000 or more from countries of concern. This is an institutional-level requirement, but your institution's compliance status affects eligibility. Funding Threshold Increases: Planning proposals up to $200,000 (was $100,000). RAPID proposals up to $300,000 (was $200,000). EAGER proposals up to $400,000 (was $300,000). Conference grants up to $200,000 without external review. These increased ceilings took effect with NSF 26-200. Confucius Institute Restrictions: Institutions maintaining Confucius Institute agreements cannot receive NSF funding without a director waiver. Foreign Drone Restrictions: Effective December 22, 2025, NSF funds cannot support procurement or operation of unmanned aircraft from covered foreign entities.
Key Deadlines: Spring and Summer 2026
NSF operates differently from NIH in that many research programs accept proposals year-round rather than on fixed annual deadlines. This means opportunities exist continuously, not just at specific dates. Upcoming Fixed Deadlines: S-STEM (Scholarships in STEM) - March 3, 2026. Up to $5 million per award for institutions providing scholarships and support to low-income STEM students. One of NSF's largest education programs that survived the DOGE terminations. Faculty Early Career Development (CAREER) - Late July 2026. The most prestigious NSF award for early-career tenure-track faculty, providing five years of funding for integrated research and education. Specific directorate deadlines cluster around the third to fourth week of July. Check the current CAREER solicitation on Research.gov for your directorate's exact date. No-Deadline Programs: Many NSF programs now operate without fixed deadlines. Standard research proposals to programs in condensed matter physics, chemical catalysis, environmental engineering, ecology, and dozens of other areas can be submitted at any time throughout the year. Research.gov lists whether each program has a deadline or accepts proposals continuously. Rolling Opportunities: RAPID (up to $300,000 for urgent research needs) and EAGER (up to $400,000 for high-risk exploratory research) can be submitted at any time with program officer approval. These mechanisms bypass the standard timeline entirely and are reviewed on an expedited basis. A Note on Disrupted Programs: Some programs that previously ran regular solicitation cycles may experience delays in posting new opportunities or reviewing submissions because of staffing changes. If a program you have relied on in the past has not posted its usual solicitation, contact the program officer to ask about timing and status. Some programs may have been consolidated, modified, or paused as part of NSF's organizational restructuring. Search NSF grants with upcoming deadlines
How NSF Compares to NIH in 2026
Both NSF and NIH experienced DOGE interventions in 2025, but the nature and scale differ in ways that matter for researchers who could apply to either agency. Budget: NSF received $8.75 billion for FY2026. NIH received $47.22 billion. Both amounts represent Congressional rejection of proposed cuts. DOGE impact by the numbers: NSF saw 1,752 grants terminated worth $1.4 billion. NIH saw more than 600 grants terminated worth $6.9 to $8.2 billion. As a fraction of portfolio, NSF's terminations were proportionally larger. NSF's terminations concentrated in two directorates (education and social sciences), while NIH's spread across multiple research categories including DEI, vaccine hesitancy, transgender health, and infectious disease. Application systems: NIH moved all new funding opportunity announcements from the NIH Guide to Grants.gov (per notice NOT-OD-25-143). NSF continues to use Research.gov as its primary submission and discovery system and has not made a comparable change. Biosketch requirements: NIH now mandates SciENcv-generated biosketches and ORCID IDs for all senior and key personnel (effective January 25, 2026, per NOT-OD-26-018). NSF uses its own biographical sketch format within Research.gov and has not adopted the SciENcv requirement. Review process: NSF reduced minimum reviews from three to two and made panels optional (NSF 26-200). NIH's study section process remains largely intact, though the volume of funded proposals decreased. Preliminary data: NSF continues to welcome exploratory research without extensive preliminary data, especially through EAGER and RAPID mechanisms. NIH, in practice, expects strong preliminary data for competitive R01 proposals. Funding rates: NSF's overall rate was approximately 27% in FY2024, with directorate variation (Engineering 23%, Biological Sciences 19%). NIH R01 rates were approximately 20%. NSF's 20% reduction in new awards during FY2025 means effective competition for FY2026 may be higher than historical rates suggest. The strategic question for researchers whose work could fit either agency is less about funding rates and more about where your specific research connects to current priorities. Both agencies are funding science, but both are also working within political constraints that make topic fit and agency alignment more important than they were two years ago. For the full NIH picture, see our NIH grants guide.
How to Compete When the Agency Is Disrupted
The standard advice for NSF proposals still applies: strong intellectual merit, specific broader impacts, appropriate scope, clean compliance. But several things are different in 2026 and worth addressing directly. Contact program officers before you write. In 2026 this is close to mandatory. With reduced staffing and organizational changes, programs may have shifted scope, changed review timelines, or adopted priorities not yet reflected in posted solicitations. Program officers are listed in every funding announcement and on Research.gov. A five-minute email exchange can prevent weeks of misdirected effort. Do not assume last year's solicitation is current. Some programs have been modified, consolidated, or discontinued as part of NSF's reorganization. Check Research.gov for the active version of any solicitation you are targeting. If a program you previously applied to has not posted its usual solicitation, ask the program officer whether it is still active. Scope proposals tightly. Proposing five years of work on a two-year budget is a consistent disqualifier. With fewer reviewers, optional panels, and more program officer discretion, there is less opportunity for a panel discussion to surface the strong parts of an overambitious proposal. Each proposal needs to stand clearly on its own. Avoid tool-first framing. Proposals that read as a technique looking for a problem consistently score poorly. Lead with the scientific question. Explain why your approach is the right one for that question. Reviewers with deep domain expertise can tell the difference between problem-driven and method-driven proposals. Treat broader impacts as seriously as intellectual merit. Generic statements about publishing results or training students are the most common fixable weakness in unfunded NSF proposals. Specific plans for outreach, education, or broadening participation score better. In the current environment, demonstrating concrete public benefit from your research is more important, not less. Use the right mechanism. Beyond standard research grants, NSF offers RAPID (up to $300,000) for urgent research, EAGER (up to $400,000) for high-risk exploratory ideas, planning grants (up to $200,000) for developing larger proposals, and conference grants (up to $200,000) without external review. These mechanisms can be faster and face different review processes. Prepare for less feedback. If your proposal is not funded, expect three to five sentences of summary rather than the detailed feedback of previous cycles. Keep notes during writing about key decisions and alternatives considered so you can diagnose problems yourself for resubmission. For researchers also considering federal R&D contracts, BAAs and OTAs continued operating through the DOGE period with less disruption than NSF's grants programs.
What to Do Before the Next Deadline
A practical checklist for researchers targeting NSF in spring or summer 2026. Registration and compliance: - Complete NSF's research security training modules (required under the CHIPS and Science Act). This must be done before submission for all senior and key personnel. - Verify your institution's SAM.gov registration is active and not expiring within 60 days. See the SAM.gov registration guide. - Confirm your Research.gov account is active and institutional permissions are current. - Update your data management plan to address the data-sharing-at-publication requirement from NSF 26-200. Finding the right program: - Search Research.gov for programs matching your discipline. Note whether each program has a fixed deadline or accepts proposals continuously. - Check NSF's Updates on NSF Priorities page for the agency's current direction. - Read past awards in your target program to understand what has been funded recently. Research.gov lists awarded proposals with abstracts. Pre-submission: - Email the program officer 6-8 weeks before your intended submission date. Ask whether your research fits the program's current priorities. Ask about expected review timelines given staffing changes. - Get institutional sign-off on your budget well before the deadline. Many institutions have internal deadlines 5-10 business days before the NSF deadline. - If collaborators are at other institutions, confirm their SAM.gov registration and research security training are current. Broader funding options: - If your research has defense applications, BAAs and OTAs continued funding through the DOGE disruptions with fewer portfolio-wide terminations. - If SBIR was part of your funding plan, see our SBIR guide for the current reauthorization status. The program remains frozen. - For tech startups that relied on NSF as a primary funder, diversifying applications across multiple agencies reduces exposure to disruption at any single agency. Search all open NSF opportunities