On Dec. 15, 2025, the U.S. National Science Foundation (NSF) is relaxing some of its requirements for the review of grant proposals in an effort to cope with a backlog of applications and reduced staff numbers. Staff members at the agency, one of the world’s leading funders of basic science, say that the new strategy could have benefits, but raises the risk of grant applications being inadequately reviewed.

Previously, proposals for research grants were generally required to undergo at least three external reviews by scientists from outside the agency. Now, two reviews are enough and, under some circumstances, one can be an internal review by a scientist at the NSF.

The change in policy was described in an update of the agency’s Proposal and Award Policies and Procedures Guide that was published on 8 December. Nature also obtained the more detailed internal guidance supplied to NSF programme officers, who are in charge of the review process. To get a better understanding of the changes, Nature spoke to five NSF staff members. All requested anonymity because of concerns about retaliation.

Substituting an internal review for an external review could save time, programme officers say. “I’m glad to have a little more flexibility,” one says, especially when they are handling proposals that do not meet basic criteria for funding. “But I would hate to see two reviews becoming the norm for the proposals that are legitimate candidates for funding.” Programme officers say that they value a range of external expert perspectives.

Like other science agencies, the NSF relies heavily on external expertise to assess grant proposals. Proposals are sent to peer-review panels of between three and ten independent specialists, each of whom produces reviews of a handful of proposals. To supplement panellists’ evaluations, programme officers also solicit other external experts for ad hoc reviews.

During day-long discussion sessions, panels consider panellists’ reviews and any ad hoc reviews, then produce assessments of proposals. Programme officers use these assessments to make recommendations to division directors, who decide which proposals are funded.

Programme officers also write summaries of these assessments, which can provide feedback for researchers. Until now, those summaries were highly detailed and multiple paragraphs long. But new internal guidance limits summaries to three to five sentences. It is a shame to limit the summaries, which are “a more helpful synthesis” than other documents from the review process, says Laurel Yohe, a bioinformatics researcher at the University of North Carolina, Charlotte.