Nonprofit Refuses $1.5M Science Grant Due To New Federal DEI Rules

Forbes

a day ago

The Carpentries, a nonprofit that has trained over 100,000 researchers in coding and data skills, turned down a $1.5 million NSF grant after being asked to strip diversity-related content from its programming. For an organization with just three months of cash on hand, it was a decision that threatens the organization's very existence.
The Carpentries' experience reveals the real-world consequences of the Trump administration's policy shift prohibiting diversity, equity, and inclusion (DEI) efforts in federally funded science programs. These restrictions, added to NSF's Grant General Conditions on May 19, 2025, are reshaping what kinds of research the US government will fund and which organizations can participate. For a community-led group like The Carpentries, whose mission is fundamentally inclusive, compliance would have meant abandoning core values. So they said no.
Why organizations like The Carpentries are crucial for national competitiveness
Since its founding in 1998, The Carpentries has grown from a single lesson program into a global nonprofit teaching foundational data and coding skills to novice researchers across 71 countries. In its 2024 annual report, the organization cites more than 4,600 workshops, over 5,100 trained instructors, and more than 100,000 learners served worldwide over the past decade. Their entire curriculum is open source, volunteer-led, and centered around inclusivity and reproducible science.
The organization's mission—empowering a diverse, global scientific community capable of interpreting and leveraging data—aligns with broader economic imperatives. According to the National Skills Coalition and the Federal Reserve Bank of Atlanta, over 92% of jobs in the U.S. economy require digital skills, yet one-third of workers lack this foundational capability. Similarly, according to a June 2025 report from the Organisation for Economic Co-operation and Development (OECD) many countries are now shifting labor markets toward skills over credentials, and digital fluency has become essential to economic competitiveness. The share of U.S. employers having difficulties filling jobs due to a lack of available talent increased from just greater than 30% in 2013 to more than 70% in 2023. All of this is unfolding just as the rise of artificial intelligence across industries is accelerating demand for data-literate workers—especially those with the skills to manage, interpret, and audit complex computational systems.
I have written previously about the tight relationship between long-term productivity growth and basic scientific research. Many say the administration of Donald Trump has resulted in nothing short of a war on science. In this context, The Carpentries is more than a niche teaching organization, but rather a contributor to economic resilience and, from that vantage point, an important public good. By providing accessible pathways to digital literacy, it equips future data workers with the technical skills increasingly required across all sectors of the economy. Its volunteer-led model, multilingual curriculum, and global reach also help close persistent gaps in access to technical education. In this sense, it is also crucial for workforce development. As digital tools become foundational to modern research, business, and public service, organizations like The Carpentries are building the knowledge necessary for data-driven innovation and national competitiveness.
The grant and a line in the sand
In September 2024, The Carpentries submitted a proposal to the National Science Foundation through a program called Pathways to Enable Open-Source Ecosystems, or POSE, a federal initiative designed to stimulate the growth and long-term sustainability of open-source projects that serve the general research enterprise. Phase II of POSE provides substantial funding for well-established organizations, like The Carpentries, to formalize governance structures, improve contributor engagement, and expand community participation.
The Carpentries' proposal requested $1.5 million over two years to strengthen multilingual programming, build mechanisms for recognizing and retaining community contributors, and test a more flexible membership model to develop greater financial stability. In February 2025, the proposal was recommended for funding, meaning it had been rated highly meritorious through peer review by leading scientists, was prioritized by NSF program staff, and only required final administrative approval to be awarded.
But in May, the NSF informed The Carpentries that their project had been flagged for diversity-related content. In communications shared with me by Directors Erin Becker and Kari Jordan, the agency wrote: 'Your project contains activities for the retention of underrepresented students which has a limitation or preference in outreach, recruitment, participation that is not aligned to NSF priorities.' Later, NSF announced new general conditions that included that grantees must certify that they did not—and would not—'operate any programs that advance or promote DEI' where DEI is a federally recognized abbreviation for the phrase 'diversity, equity, and inclusion.' While long embraced by universities and nonprofits as a framework for expanding access and representation, DEI has also become a point of contention in U.S. conservative politics, where it is often viewed as an instrument of soft political power within traditionally left-leaning institutions.
For an organization whose mission is grounded in inclusive access to data skills, this was untenable. The Carpentries formally withdrew. 'We are unable to certify that our organisation does not and will not operate any programs that advance or promote DEI' they wrote to the NSF.
Inclusion Isn't Peripheral—It's the Point
For The Carpentries, diversity, equity, and inclusion are not ancillary values—they are embedded in the organization's core design. As Becker and Jordan emphasized to me, their mission is not about preference but about access: providing subsidies for participants from low-income backgrounds, translating materials into local languages, and using teaching methods that welcome learners of all levels and identities. This ethos has allowed The Carpentries to train over 100,000 people in foundational coding and data skills across more than 70 countries.
In response to the NSF's new restrictions, the team briefly discussed whether they could reframe their mission in more politically neutral terms—perhaps using language like 'democratizing data science.' But they rejected the idea. 'Where do we draw the line?' Becker asked. 'If we scrub the word 'diversity,' do we rewrite our code of conduct next?' Jordan was blunt: 'We want to be explicit—this is for everyone.'
That commitment to inclusivity has required grappling with real tensions. Wondering how much viewpoint diversity the organization would really be open to, I asked how far their openness reached. Would they be open to specifically Republican points of view? Becker recalled a recent example from their instructor training curriculum, which had listed belief in a young Earth as a common scientific misconception. A community member flagged the example as unnecessarily divisive, pointing out that some people who hold that belief could still be effective coding instructors. After a public and respectful debate, the community chose to remove the example—not because they endorsed creationism, but because they wanted to maintain a space where a truly broad range of perspectives could contribute to shared learning. 'If we all care about science,' Jordan said, 'we should be able to work through our challenges together.'
The Broader Implication
The implications of this case extend beyond one nonprofit. Universities, companies, and research organizations across the country now face a stark choice: adhere to restrictive federal conditions that may conflict with longstanding institutional values or forgo the funding that supports their operations. The NSF's policy shift risks disqualifying organizations not because their work lacks scientific merit, but because of differences in how they define fairness, access, or institutional responsibility. That sets a troubling precedent, not just for science, but for American institutions more broadly. While federal funders have a legitimate interest in setting ethical and legal boundaries, the principle at stake is that participation in public research should be judged by quality and contribution, not political conformity.