The NIH Public Access Policy: Q&A for Authors

We’ve received many questions from authors, librarians, and research administrators about the NIH decision to accelerate implementation of its public access plan, going into effect on July 1, 2025. The policy will now require all NIH grantees to deposit research articles upon acceptance, for immediate public availability on the date of publication through the NIH public access repository, PubMed Central. Below are some responses to questions we’ve received: 

1. Researchers are currently signing agreements with publishers for publications that may be released after July 1. These agreements might still contain year-long embargo provisions that conflict with the new NIH policy. Is this a valid concern based on what you are seeing?

The NIH revision of policy NOT-OD-25-101 states that “Author Accepted Manuscripts […] with acceptance dates on or after July 1, 2025, are subject to the Policy.” If an author signs a publisher agreement that includes a one-year embargo before July 1st, and the article is accepted on or after July 1st, then the publisher agreement and NIH requirements are indeed in direct conflict. This is a realistic problem given some publishers’ workflows where, for example, they require authors to agree and select up-front, upon submission, how they would like their article published  (open access or not), which can occur months before actual acceptance of the article.

But to be clear, neither the funding date nor the submission date impacts the applicability of the policy—the policy only applies to articles accepted on or after July 1. 

2. What will happen for authors who want to publish with journals that do not allow for zero-embargo deposit of articles they publish? 

Several major commercial publishers (including mega-publishers Elsevier, Springer Nature) per their publicly stated policies, at present require embargos for deposit of articles in public access repositories, and there are no indications that these publishers will change their policies to accommodate the NIH on July 1. 

This creates a contradictory policy environment: the NIH policy gives authors the obligation (policy mandate¹) and the NIH obtains the right (via the federal purpose license²) to make an immediate repository deposit. At the same time, publishers may of course set their own policies. For instance, a publisher can decide to refuse publication of an article that carries a pre-existing contractual burden such as an immediate deposit requirement (either from a sponsor or employing institution), or they impose conditions for doing so such as  (1) an Article Processing Charge being paid by the author. It is important to stress that how these publishers might enforce their policy positions remains an open question, and of course their positions could evolve over time.

It seems likely that many publishers will try to route authors to paid OA whenever they realize the author has a zero-embargo deposit obligation during (or even after) the publication process. This mirrors the actions some publishers are taking with respect to authors subject to Plan S (open access requirements from a coalition of largely European funders).  

Accordingly, authors should pay close attention to publisher policies before submitting; if grant funds or other funding is not available for these fees, it could mean a long period of wasted review and effort if the author or their institution is unwilling or unable to pay the publisher’s fees. 

3. Some publishers are rolling out new fees for publication of articles that are subject to funder or institutional deposit requirements. Must authors pay these fees to comply with their grants? 

Many open access journals, including nonprofit journals such as PLOS One and for-profit journals such as those in Elsevier’s portfolio for which it offers authors an OA option, have long required payment for publication through article processing fees (APCs). For these OA journals, their terms normally align with public access requirements from funders such as the NIH. This is not a new development. 

What is new is a new class of fees imposed by a handful of publishers for publication in journals that do not by default allow for immediate open access at the site of publication. These have generated some confusion.

For example, the American Chemical Society (ACS) has created an entirely new pathway requiring authors to pay for compliance with funder and institutional policies. Under ACS’s “Article Development Charge,” ACS requires authors to pay $2,500 for the right to deposit their articles with no embargo. More specifically, ACS states that this fee will be assessed when: 

• The article has been submitted to an ACS hybrid journal;
• The manuscript includes funder- or institution-recommended language regarding zero-embargo green open access; and
• The corresponding author’s institution does not participate in an institutional open access agreement, and they have not indicated that they intend to publish the version of record as immediate ‘gold’ open access.

The Institute of Electrical and Electronics Engineers (IEEE) is another publisher to implement a new fee, which it calls a “repository license fee.” Announced in March 2025, the repository license fee purports to grant back to authors the right to post articles in a repository. IEEE charges $1,275 for the fee, but it only applies in situations where authors are mandated to deposit the Author Accepted Manuscript version of their article under Creative Commons CC-BY license. 

For the most part IEEE’s Repository License Fee does not apply to U.S. federal funder grant requirements. Neither the NIH nor any other federal funder has required that grantees share their article under a CC-BY license. In fact, several agencies have considered this requirement (and, for the record, we think it’s a good idea), but  have rejected it for now. See, for example, the Department of Energy public access plan, p. 8 or the National Science Foundation public access plan, p. 12.

So, must NIH-funded (or other federally-funded) grantees pay these fees? While these policies seem to have been created in response to federal public access requirements, at least IEEE’s policy does not seem to apply. As for ACS, it has quite a lot of practical power to accept or reject publication, so whatever lack of clarity there may be, we’ll have to wait and see how ACS approaches enforcement in practice.³ That same thing is true for other large publishers who may be thinking about implementing new fees—unfortunately, if publishers want to dictate these fees and make them mandatory, authors are in the short term left with few good options aside from publishing their work elsewhere. 

4. Do any publishers allow for zero-embargo deposit without paying an additional open access or publication fee? 

Many publishers allow for zero-embargo deposit of articles without any fee. A good resource to check if a publisher allows this is the JISC open policy finder (formerly Sherpa-Romeo). Sage and the American Association for the Advancement of Science (AAAS) are two of the biggest publishers to indicate that they have no objections to authors complying with their grant obligations to make the Author Accepted Manuscript available immediately upon publication in a public access repository.

A handful of other major publishers are unclear on their stance. For instance, Wiley states on its self-archiving page that authors only have the right to share the peer-reviewed accepted manuscript after a 12-month embargo. Yet, the standard Wiley copyright transfer agreement for non-OA articles also states that “[i]n the case of a Contribution prepared under U.S. Government contract or grant, the U.S. Government may reproduce, without charge, all or portions of the Contribution and may authorize others to do so, for official U.S. Government purposes only, if the U.S. Government contract or grant so requires.” Because federal public access policies are built into U.S. grant obligations and it is the U.S. government exercising those rights to share the article (in the case of NIH, through its PubMed Central), it would seem that Wiley is recognizing that authors can comply with their federal grant obligations even under their standard publishing agreement. 

5. What if an author mistakenly agrees to a publication agreement that purports to restrict their ability to comply with federal public access requirements? 

Our experience is that many authors have found publisher explanations of their public access and sharing requirements to be confusing, incomplete, and, in some cases, misleading. Authors are also increasingly presented with publishing agreements upon submission via click-through forms that discourage close reading. These conflicting policies and the manner in which they are presented to authors raise the risk that some authors may be induced to sign publishing agreements that conflict with their grant obligations. If this happens, what are the likely risks for that author or their institutions? A couple of observations: 

Publisher agreements cannot supersede the federal agencies’ pre-existing rights.  As we’ve explained before, we think it very unlikely that these conflicting agreements will limit federal agencies’ rights to share articles publicly, and therefore are unlikely to put  authors (or their institutions) in a position of being in breach of their grant obligations. This is because the NIH and other federal agencies obtain a first-priority license covering the article as a required term of the grant agreement. In effect, this means that the federal government already has its rights secured before the publisher ever enters the picture, and therefore, nothing that a publisher asks an author to sign can give those rights away. Of course, authors must still take the mechanical step of ensuring that their articles are actually deposited with the agency. 

The outcome for authors in these cases will ultimately be dictated by how aggressively the publisher decides to enforce its contractual rights. While some publishers may not continue their prior practice of automatically depositing articles on authors’ behalf under the new policy, there is little they can do to prevent an author from making the deposit. Authors normally would have their manuscript versions, including the Author Accepted Manuscript, as well as access to funder repositories, and funders do have the right to host the article.

6. What happens if an author (or their institution) finds themselves in a conflict with a publisher insistent on enforcing contractual rights to control distribution of the article? 

One scenario, as noted above, is that the publisher could demand payment from the author to make the article compliant with the publisher’s open access or repository deposit policies. We expect that this will be the predominant form of enforcement. What happens if an author refuses to pay? We really don’t know, though you can imagine some possible ways a publisher could exert leverage,  including retracting the article or even refusing to publish that author in the future (though to be clear, no publisher that we are aware of have suggested they will do this). 

There are a few other possible scenarios, such as the publisher issuing a takedown notice to a funder repository. We think this is unlikely, in part because it would require the publisher to directly test the strength of the government’s asserted license, a proposition that could significantly undermine the position many publishers have publicly taken. 

It is also theoretically possible that publishers could pursue legal action against individual authors for breach of contract. Again, we view this as unlikely: this scenario will put publishers in direct conflict with the very authors they want to attract and cultivate, creating a significant disincentive to pursuing this remedy. It could be that a publisher would pursue a case to set a precedent and make an example of non-compliant authors, but this comes with significant risks to the publisher. It’s also worth noting that publishers are unlikely to gain much financially from such a suit. In this scenario, a publisher would be making a breach of contract claim, and to recover anything it would have to prove actual damages which in most cases would be negligible. 

 7. How can institutions proactively protect themselves and their authors from risk, either from grant agencies or publishers? 

We think the most effective way to manage risk is robust communication and guidance from universities and federal granting agencies to grantees who have not yet submitted their articles. For authors, we might suggest:

• Submitting articles to a publisher that allows zero-embargo deposit. For subscription articles, this requires careful review of the author agreement and publisher policy. Applicable policies on publisher websites can often be found by looking for policy on “self-archiving” or “green OA.”
• Publishing in fully open access journals, which have no restrictions for repository deposit. Publishing in a diamond OA journal and publishing under an institutional open access transformative agreement (where your institution has negotiated for OA publishing on behalf of their authors) are pathways without additional cost barriers to the author/grantee to be in compliance.
• Budgeting for article processing charges to allow for gold open access publishing. Per NIH policy (“[r]easonable costs associated with publication that are allowable costs of the project budget may be requested as direct or indirect costs”), authors have the option to budget for an open access publication in their grant, and ensure compliance with federal and publisher policy in that way.

8. How can Institutions support authors who find themselves in conflict with publisher policies? 

Authors who have submitted their article to a publisher without awareness of the policy implications for compliance may find themselves in a difficult position. A publisher may ask an author, who did not budget for a payment in their grant and who does not have institutional open access publishing support, to make a payment (an article processing charge, or an alternative payment) in order to comply with their policy for zero-embargo deposits, and may refuse publication unless such a payment is made.

Institutions should also be aware that an author’s leverage to change terms in a publisher agreement are extremely limited at best. Often, an author’s agreement is presented as a click-through, with no opportunity to modify, and the journal editorial staff an author may contact do not have the authority to change the terms of those agreements. Moreover, authors usually do not have a background that would equip them to negotiate nuanced contract terms, even if such a negotiation were possible. This type of intervention also misunderstands the problem:authors and their institutions have already agreed to terms with granting agencies that allow for deposit of their article under NIH’s public access plan, and do not need to negotiate for that right; but publishers still have the right to refuse publication.

We recommend that institutions: 

• Reach out to publishers to clarify their policy stance regarding NIH’s public access plans; 
• Communicate with authors about the importance of choosing a publication venue that supports a zero-embargo deposit, as well as about budgeting for open access publication in their grant proposals to further broaden their options; and 
• Ensure authors are aware of the institutional support available (open access funding, transformative agreements) as well as alternatives to publishing under APC based models (e.g., diamond open access journals). 

In addition, institutions may also want to remind their authors that deposit in a local university repository may be beneficial, in an environment where the continuity of federal repositories and the ability of the federal government to steward the scholarly record for the long term may be evolving.

1. “The NIH Public Access Policy requires: Submission of an electronic version of the Author Accepted Manuscript to PubMed Central upon its acceptance for publication for public availability without embargo upon the Official Date of Publication.” https://grants.nih.gov/grants/guide/notice-files/NOT-OD-25-047.html ↩︎
2. “By accepting NIH funding, the recipient grants to NIH, as the funding agency, a royalty-free, nonexclusive, and irrevocable right to reproduce, publish, or otherwise use the work for federal purposes and to authorize others to do so, which includes making Author Accepted Manuscripts publicly available in PubMed Central upon the Official Date of Publication. A statement that conveys this point is incorporated into Notices of Award, the terms of Other Transaction agreements, and applicable contracts.” Id.; see also 2 C.F.R. § 200.315(b). ↩︎
3.  For readers who are interested in what these unclear provisions are: Read in combination with the standard ACS publication agreement, it is clear that authors must transfer whatever rights they have at the time they sign the agreement to ACS. It is also clear that the author cannot, subsequent to signing this agreement, authorize redistribution of their article under zero-embargo terms without violating ACS’s posting policies, which are incorporated by reference into its publication agreement. 
   
   That said, it is not clear this ADC fee should apply to NIH grantees under the stated terms. First, the ACS agreement merely asks authors to “transfer their copyright interests in the Article,” but does not require authors to make any representation or warranty that the authors actually hold 100% of the copyright (and authors likely do not, given the pre-existing license granted under the Federal Purpose License). This means authors are not making any promises about other uses of their articles by others, like the federal government, who may hold pre-existing rights.  Similarly, while the ACS agreement prohibits authors from attempting to post articles in public access repositories under their own authority with zero embargo, the agreement does not and cannot restrict federal agencies—which independently hold a non-exclusive right to the article—from doing so themselves. Finally, neither the NIH nor most other federal funders require (though they do recommend) authors to include zero-embargo manuscript language, and so it may be that the ADC is inapplicable in any case because authors could just choose not to include that language. Whether these are oversights by ACS or gaps that it intended to later fill to force authors to pay this fee remains to be seen, but for now it is unclear. ↩︎