The Bill & Melinda Gates Foundation, one of the world’s top biomedical research funders, will from next year require grant holders to make their research publicly available as preprints, articles that haven’t yet been accepted by a journal or gone through peer review. The foundation also said it would stop paying for article-processing charges (APCs) — fees imposed by some journal publishers to make scientific articles freely available online for all readers, a system known as open access (OA).

The Gates Foundation is the first major science funder to take such an approach with preprints, says Lisa Hinchliffe, a librarian and academic at the University of Illinois Urbana–Champaign. The policies — which take effect on 1 January 2025 — elevate the role of preprints and are aimed at reducing the money the Gates Foundation spends on APCs, while ensuring that the research is free to read.

But the policy’s ramifications are unclear. “Whether this will help the open-access movement or not, it’s hard to know,” Hinchliffe says. On the one hand, more research will become freely available in preprint form, she notes. On the other, the final published versions of articles, known as the version of record, might become harder to access. Under the revised rules, after sharing their manuscript as a preprint, authors will be allowed to submit it to the journal of their choice and will no longer be required to select the OA option.

“Our decision is driven by our goals of immediate access to research, global reuse and equitable action,” says Ashley Farley, programme officer of knowledge and research services at the Gates Foundation in Seattle, Washington. Grant recipients will still be required to post their preprints under a licence that allows their contents to be reused, she says. The foundation plans to publish the full policy within the next couple of weeks.

OA efforts

The Gates Foundation announced in 2015 that it would require its grant recipients to make their research articles freely available at the time of publication by placing them in open repositories. It later joined cOAlition S — a group of mainly European research funders and organizations supporting OA academic publishing — and endorsed the group’s Plan S, by which funders mandate that grant holders publish their work through an OA route.

But the Gates Foundation’s latest policy puts it on course to diverge from the group. It is not “entirely in line with cOAlition S”, says Johan Rooryck, executive director of the coalition, who is based in Leiden, the Netherlands. Whereas cOAlition S requires either an accepted manuscript or the version of record to be available OA, he says, “the Gates Foundation is clearly of the opinion that the preprint is sufficient”. He notes that the group allows for “a lot of leeway in policies” between its members, adding that the Gates policy continues to uphold key aspects of Plan S, such as promoting authors’ retention of rights to their accepted manuscripts.

The coalition has been examining the role of preprints in OA, but it’s a long way from adopting any related policy changes, Rooryck says. A document released by the group last year discussed the issue, and the coalition is gathering feedback from the research community through a survey open until 22 April. No decisions will be made on adopting any proposal before the end of the year.

Another difference between Plan S and the Gates policy is their stances on APCs. “Ending support for APC payments is not the cOAlition S policy, I can be very clear about that,” Rooryck says. “That’s a decision that Gates has taken. It’s not a decision that we, as cOAlition S, are ready to make by 1 January 2025.”

Ending support for APCs is a “very sensible plan” given the unsustainable increase in such charges in recent years, says Lynn Kamerlin, a computational biophysicist at the Georgia Institute of Technology in Atlanta. “The Gates Foundation plan is the open-access plan I would have liked to see when Plan S was announced.”

Juan Pablo Alperin, a scholarly-communications researcher at Simon Fraser University in Vancouver, Canada, notes that APCs are “inherently an unjust way” of supporting OA. “Stopping support for APCs sends a signal to the larger community, including the community of funders, that this mechanism is not a way forward,” he says.

Effects on publishing

It’s hard to predict the effects of the Gates policy on scientific publishing, says Hinchliffe. Some grant holders might find it harder to publish in OA journals, and rely more on preprints to disseminate their work. But others might continue to publish through OA journal routes, especially if they have other funding sources to cover the APCs, or if their institutions’ libraries have agreements with publishers to reduce the costs of OA publishing.

Although the Gates Foundation is a big funder — with a budget of US$8.6 billion in 2024 — it still funds only a modest percentage of the world’s research, Hinchliffe notes, and it’s not clear whether other funders will follow suit. Some, even among those that require OA publishing, already refuse to cover APCs.

Another potential consequence of the policy is that there might be a difference in the quality of a manuscript freely available as a preprint and its final version behind a paywall. In certain cases, people with access to the final version are going to be in a better position to avoid particular kinds of mistake than are those who rely solely on the preprint, Hinchliffe says. Kamerlin notes that an increasing number of preprint publishers allow authors to update their preprints as many times as necessary, which could ease that concern.

Farley says that there is growing evidence that errors in early versions of preprints are addressed quickly, “as there is a much broader pool of researchers to read and evaluate the preprint”. The foundation will provide grant recipients with a list of recommended preprint servers “that have demonstrated a level of checks that ensure the scientific validity of research”, she adds. It has also invested in a new preprint service called VeriXiv, “which will set new standards for preprint checking”.

Some authors might well choose not to publish formally in journals, deciding that the preprint is enough, says Alperin. “I don’t see that as being a problem in itself,” he says. “Sometimes, the goal of a journal publication has been a negative force in science, encouraging people to focus on publishing in a particular journal when the goal should really be to do high-quality research and to ensure that it is communicated and that it reaches the right audience.”

Publishers contacted by Nature’s news team said they are still assessing the Gates policy. (Nature’s news team is editorially independent of its publisher, Springer Nature.) “We are reviewing the implications of the Bill & Melinda Gates Foundation’s new open-access policy and what it means for how we support their researchers,” said a spokesperson for the publisher Elsevier in a statement.

Roheena Anand, executive director of global publishing development and sales at the publisher PLOS, which is based in San Francisco, California, said in a statement that PLOS has already recognized that the APC model of OA publishing creates inequities. “We are committed to finding sustainable and equitable alternatives. That’s why we have launched several non-APC models and are also working with a multi-stakeholder working group,” she says, “to identify more equitable routes to knowledge-sharing beyond article-based charges.” She added that there is a risk that, without established alternatives, researchers funded by the Gates Foundation will revert to publishing their work behind paywalls. “PLOS’s newer business models offer one possible alternative.”

In an article announcing the changes, Estee Torok, a senior programme officer at the Gates Foundation, wrote that the organization has paid around $6 million in APCs per year since 2015. “We’ve become convinced that this money could be better spent elsewhere to accelerate progress for people,” she wrote. Farley says that the foundation plans to invest in more equitable OA models, such as ‘diamond OA’, a system in which publishers don’t charge fees to authors or readers, as well as preprint servers and other platforms and technologies for research dissemination.