NSF’s Broader Impacts merit-review criterion has not been rewritten for 2026. There is no new PAPPG chapter, no revised criterion wording, and no discrete “2026 policy” on Broader Impacts specifically. What is true is that a real, substantive reframing of the criterion — NSF’s April 2025 Statement of Priorities — remains fully in force and continues to shape how every 2026 proposal is reviewed, alongside two narrower policy notices that took effect in the same window but do not touch Broader Impacts wording at all. This post separates what actually changed from what has not, for applicants and research-administration staff preparing proposals under current guidance. Last verified against nsf.gov 2026-08-16.
The two-criterion structure is unchanged
NSF merit review still applies two co-equal, National Science Board-approved criteria to every proposal: Intellectual Merit (the potential to advance knowledge) and Broader Impacts (the potential to benefit society and contribute to specific, desired societal outcomes). Both criteria are evaluated against the same five review elements — the proposed activity’s potential to advance knowledge or benefit society, how creative/original/transformative the concepts are, whether the plan is well-reasoned and well-organized with a mechanism to assess success, the qualifications of the individual/team/organization, and whether adequate resources are available. This structure dates to 1997 and PAPPG Chapter III.A did not change it for 2026.
NSF’s own Broader Impacts guidance still lists example societal-outcome categories — STEM education and educator development, public engagement and scientific literacy, societal well-being, a diverse and globally competitive STEM workforce, partnerships between academia and industry or other sectors, national security, U.S. economic competitiveness, and improved research/education infrastructure — and is explicit that the list is illustrative, not exhaustive or prescriptive. NSF also still recognizes three legitimate ways to deliver a broader impact: through the research activity itself, through activities directly related to the research, or through complementary activities supported by the research. A bolted-on outreach event is only one of three valid paths, not the default expectation.
What actually changed: the April 2025 Statement of Priorities
The real shift governing 2026 proposals dates to April 18, 2025, when NSF Director Sethuraman Panchanathan issued a Statement of Priorities that substantively reframed how broadening-participation and outreach activities within Broader Impacts should be scoped. Applicants are directed to prioritize the America COMPETES Reauthorization Act of 2010’s broader-impacts goals, and broadening-participation or outreach activities “must aim to create opportunities for all Americans everywhere” and “should not preference some groups at the expense of others” — NSF’s own wording. Practically: framing an activity around open, non-exclusionary eligibility (by geography, socioeconomic status, institution type, first-generation status, and similar) remains acceptable; a plan that preferences or excludes participants by protected class or characteristic does not. NSF states directly that it will not fund proposals whose broader impact is narrowly limited to subgroups defined by protected class or characteristic, and that DEI frameworks are not an acceptable basis for project design.
This restriction targets how broader-impacts activities are scoped, not research topics generally. Research where a protected characteristic is intrinsic to the research question itself — NSF’s own example is disability-focused assistive-technology research — remains fundable on its own terms. Legally mandated broadening-participation programs and cross-institution-type partnerships (research-intensive universities working with community colleges, minority-serving institutions, or EPSCoR jurisdictions, for example) continue to be supported. Because this reframing is genuinely contestable and has drawn scrutiny from the research community, applicants and proposal-development offices should treat NSF’s own nsf.gov/updates-on-priorities page as the authoritative source, not a secondary summary, when scoping a specific plan.
What has not changed for 2026 proposals
Two things that got real 2026-effective policy notices are easy to mistake for a Broader Impacts change — they are not one. NSF Policy Notice 26-200 (effective for awards made on or after December 8, 2025) revised SBIR/STTR Project Pitch requirements, review and return-without-review criteria, several budget thresholds, the equipment-tracking threshold, research-security provisions, and the research-misconduct definition. NSF Policy Notice 26-202 (effective for awards made on or after January 22, 2026) tightened the public-access requirement, requiring zero-embargo Author’s Accepted Manuscript deposit for newly funded awards. Neither notice revises the Broader Impacts criterion, its review elements, or its reviewer guidance. The current PAPPG remains NSF 24-1, in effect for proposals due on or after May 20, 2024, with these two notices layered on top as targeted supplements — not a renumbered PAPPG 25-1 or 26-1.
One thing worth watching: the draft Guidance on Financial Assistance
NSF has deferred the PAPPG 26-1 revision it had previously signaled, citing Executive Order 14332 and an OMB Uniform Guidance streamlining effort, and instead published a draft full replacement document called the Guidance on Financial Assistance (GFA), docket NSF-2026-OTR-0001, via a Federal Register notice dated June 24, 2026, with a revised draft text released in July 2026. The GFA is out for public comment through August 24, 2026 and is not yet adopted policy. Whether the final GFA touches Broader Impacts review language specifically is not yet knowable from the draft stage alone. Applicants should keep working from current PAPPG 24-1 guidance and NSF’s Broader Impacts page until an adoption decision is announced, and research-administration offices tracking NSF policy should flag the comment-period close date for a status check rather than assume any outcome in advance.
Writing a Broader Impacts section under current guidance
None of the above changes the basic discipline of a strong Broader Impacts section, and the practical guidance for PIs is unchanged in substance:
- Connect it to the science. A plan is stronger when the broader impact grows out of the research activity itself or activities directly tied to it, rather than reading as a generic, swappable outreach add-on.
- Name a mechanism to assess success. Review element three explicitly asks for this on both criteria — specify what will be measured, when, and by whom, not just an aspiration.
- Use NSF’s societal-outcome categories as a checklist, not a script. STEM education, public engagement and scientific literacy, workforce diversity and competitiveness, infrastructure, cross-sector partnerships, national security, and economic competitiveness are all legitimate framings — pick the one the actual project supports, don’t force-fit one for its own sake.
- Frame broadening-participation activities around open eligibility, not protected-class targeting. Under the April 2025 Statement of Priorities, describe who can participate in inclusive, non-exclusionary terms (open application, geography, institution type, first-generation status) rather than restricting eligibility by protected characteristic — unless the research question itself is intrinsically about a specific population, in which case that framing is different and should be described as such.
- Match ambition to award size. NSF’s proportionality principle expects broader-impacts plans to be assessed using metrics appropriate to the resources actually provided — a small standard grant does not need an evaluation apparatus sized for a center-scale award.
What research-administration and proposal-development offices should do
Three concrete actions for 2026: first, make sure any internal Broader Impacts guidance, templates, or boilerplate language your office circulates reflects the April 2025 Statement of Priorities framing, not the pre-2025 baseline — older internal guidance describing broadening-participation activities in protected-class terms needs updating. Second, if your office tracks PAPPG version numbers as a compliance trigger, note that NSF has not issued PAPPG 25-1 or 26-1 as of this writing; the operative document is still 24-1 plus Policy Notices 26-200 and 26-202. Third, calendar August 24, 2026 as a check-in point for the draft GFA comment period close, and revisit NSF’s public-access and PAPPG pages directly once any adoption decision is announced, since a full PAPPG replacement would affect far more than Broader Impacts alone.
Related CASRAI resources
- NSF Broader Impacts Statements: What the Review Criteria Reward — the full evergreen breakdown of the five review elements and NSF’s societal-outcome categories.
- NSF CAREER Award: Eligibility, Proposal Requirements, and How to Win One — how Broader Impacts integrates with CAREER’s research/education synergy requirement.
- How to Apply for an NSF Grant — the full proposal-preparation process, including the Project Summary’s Intellectual Merit and Broader Impacts sections.
- NSF Public Access Repository (NSF-PAR) — the mechanics behind Policy Notice 26-202’s zero-embargo deposit requirement.
- NSF’s Draft PAPPG Replacement: The Guidance on Financial Assistance (GFA) — full detail on the GFA docket, comment period, and what a full PAPPG replacement could mean.
