CASRAI Dictionary

Author: MCP Service

  • NIH’s 2026 Data Management and Sharing Plan Format: What Changed and Why It Matters

    The NIH data management and sharing policy, in force since January 2023, is getting its first substantial format revision. Under guide notice NOT-OD-26-046, the National Institutes of Health has updated the required structure of the Data Management and Sharing (DMS) Plan that accompanies most extramural applications, with the revised elements taking effect for applications submitted on or after May 2026. For grant administrators, the change is not cosmetic: reviewers and program staff will now expect more granular metadata detail at the point of submission, not left for post-award justification.

    The revision arrives as NIH data sharing policy compliance has moved from a novelty to an audited routine. Institutions have spent three years building DMS Plan templates, and many of those templates are now out of step with what NOT-OD-26-046 asks for. This article walks through what changed, what stayed the same, and how research administration offices should adjust their internal checklists ahead of the effective date.

    What NOT-OD-26-046 Changes in the NIH Data Management and Sharing Policy

    The original 2023 policy set out six required elements for a DMS Plan: data type, related tools and software, standards, preservation and access timelines, access and distribution considerations, and oversight. NOT-OD-26-046 keeps that six-element skeleton but sharpens what “sufficient detail” means within each one. The revision responds to two years of program-staff feedback that plans were frequently too vague to be enforceable — naming a repository without specifying a deposit timeline, or citing “standard formats” without naming which standard.

    The most consequential change is a new expectation that plans identify specific, named metadata standards and persistent identifier schemes rather than describing them generically. Plans that previously said data would be shared “using appropriate metadata” will now need to name the applicable schema — for example, a Dublin Core or DataCite metadata profile, or a domain-specific standard where one exists. This mirrors a broader trend across funders and publishers: DataCite, CrossRef, and ORCID have each pushed toward named, machine-actionable identifiers rather than free-text descriptions, and NIH’s revision brings its DMS Plan expectations closer into line with that norm.

    Section-by-Section Checklist for the Revised DMS Plan

    Administrators preparing applications for the May 2026 cycle should work through each element against the following practical checklist. This is not a substitute for the official NIH guidance, but it reflects where NOT-OD-26-046 tightens expectations relative to the 2023 baseline.

    1. Data type

    • Distinguish clearly between data that will be preserved and shared versus data generated but not intended for sharing, with a stated rationale for any exclusions.
    • Estimate approximate data volume and modality (imaging, sequencing, survey, sensor, etc.) rather than describing data in purely narrative terms.

    2. Related tools, software, and code

    • Name any custom code, pipelines, or proprietary software required to access or interpret the data.
    • State whether code will be deposited alongside the data (for example in a version-controlled repository) and under what licence.

    3. Standards

    • Name the specific metadata schema and, where applicable, community data standard (such as those maintained by domain repositories or standards bodies).
    • Confirm which persistent identifier scheme will be used for datasets — a DOI issued via DataCite is the default expectation in most disciplines.

    4. Data preservation, access, and timelines

    • Name the repository and confirm it meets NIH’s criteria for an established, sustainable repository, rather than describing repository characteristics generically.
    • State a concrete deposit timeline tied to publication or award milestones, not an open-ended commitment.

    5. Access, distribution, and reuse considerations

    • Address any human subjects, privacy, or controlled-access considerations explicitly, including how consent language supports secondary use.
    • State the reuse licence or terms under which shared data will be made available.

    6. Oversight

    • Name the institutional office or role responsible for monitoring compliance with the plan across the life of the award.
    • Confirm how deviations from the plan will be reported to NIH, consistent with the enforcement posture program staff have adopted since 2023.

    How the NIH Genomic Data Sharing Policy Interacts With the Revised Format

    Applications involving genomic data must continue to satisfy the NIH genomic data sharing policy (GDS Policy) in addition to the general DMS Plan requirements. NOT-OD-26-046 does not replace the GDS Policy’s separate requirements around Institutional Certifications, data use limitations, or submission to NIH-designated data repositories such as dbGaP. What changes is that the DMS Plan’s “standards” and “access” elements now need to explicitly reference the GDS-specific data use limitation categories and repository submission timelines, rather than treating GDS compliance as a parallel, loosely linked process. For research administration offices handling both genomic and non-genomic awards, this means the DMS Plan template can no longer be a single generic document — it needs a genomic-data branch that pulls in GDS-specific language by reference.

    This tightening reflects a pattern seen across major funders: as more institutions ask “what is a data management plan supposed to demonstrate”, the honest answer has shifted from “an articulated intention” to “a verifiable operational commitment.” Reviewers are increasingly trained to check whether named repositories, standards, and identifiers actually correspond to what the applicant’s field uses, rather than accepting boilerplate language.

    What This Means for Research Administrators

    The practical burden of NOT-OD-26-046 falls squarely on research administration offices, sponsored programs staff, and data librarians who support investigators through the DMS Plan drafting process. Three actions are worth prioritising before the May 2026 effective date:

    • Audit existing DMS Plan templates. Any institutional template built around the 2023 language should be checked against the sharpened standards and access-timeline expectations, particularly where templates currently rely on generic placeholder text.
    • Build a named-standards reference list. Rather than asking each investigator to independently identify the correct metadata schema, administration offices should maintain a discipline-by-discipline reference list of accepted standards and identifier schemes, updated as fields evolve.
    • Clarify oversight roles in writing. Because NOT-OD-26-046 expects a named office or role for monitoring compliance, institutions without a designated data-sharing oversight function should assign one now, rather than leaving it implicit.

    None of this is exotic. It is largely the same discipline that publishers, funders such as UKRI, and initiatives like cOAlition S have already been pushing through open-access and open-data mandates elsewhere in the research lifecycle. What is new is that NIH — as the single largest funder of extramural biomedical research — is applying the same rigour to its own plan format.

    A Standards-Originator Perspective

    Consistent, well-documented metadata practice is the connective tissue that makes data management plans auditable rather than aspirational — the same principle that underpins contributor attribution standards. CASRAI originated the CRediT contributor role taxonomy in 2014. The standard is now stewarded by NISO as ANSI/NISO Z39.104-2022. That history is a useful reference point here: a taxonomy or metadata convention only becomes durable once a neutral standards body maintains it beyond its originating organisation, and once funders and publishers build compliance checks around it rather than treating it as optional guidance. NOT-OD-26-046 is, in effect, NIH doing the same thing for dataset metadata that CRediT did for authorship — moving from generic description toward named, checkable categories.

    For grant administrators, the takeaway is not that the DMS Plan has become harder to write — it is that the plan must now say precisely what the institution intends to do, in terms that a reviewer, a repository, and an auditor can all verify independently. Offices that treat the May 2026 changes as an opportunity to formalise their metadata and oversight practices, rather than a compliance inconvenience, will be better positioned as NIH data sharing policy enforcement continues to mature through the remainder of the decade.

  • No-Cost Extensions, Carryover, and Prior Approval: A Glossary for NIH Grant Administrators

    Search traffic around the NIH Grants Policy Statement no-cost extension provisions spikes every year around the same time: the final months of a project period, when a principal investigator realises the data collection is behind schedule and someone in the sponsored programs office has to explain, again, what counts as an allowable extension and what does not. The terminology in the NIH Grants Policy Statement (GPS) is precise, but it is scattered across dozens of pages, cross-referenced against 45 CFR Part 75 and the Uniform Guidance, and rarely explained in one place for the administrators who have to apply it daily.

    This glossary collects the recurring terms — no-cost extension, carryover, consortium agreements, and program income — that generate the highest volume of long-tail searches from research administrators, and sets out what each one actually means in practice. It follows the same reference format CASRAI uses across its dictionary of research-administration terms: definition first, then the operational detail that determines whether an institution stays compliant.

    The audience is deliberately broad. US-based sponsored programs offices are the primary users of NIH terminology, but institutions that hold both NIH awards and funding from UK bodies increasingly need to translate between systems — a researcher who has learned to apply for grants under UKRI’s rules and then moves onto an NIH-funded consortium project discovers quickly that “extension” and “carryover” do not mean the same thing, or follow the same approval path, on both sides of the Atlantic.

    What the NIH Grants Policy Statement Says About No-Cost Extensions

    A no-cost extension (NCE) lengthens the period of performance on an NIH award without adding funds. The NIH Grants Policy Statement recognises two routes:

    • First, unilateral NCE: the recipient institution may extend the final budget period of a grant once, for up to 12 months, without prior NIH approval, provided the award is not in its final year of a competitive segment for reasons that require agency sign-off (for example, if additional time is needed beyond the single automatic extension, or if the award carries specific terms restricting this authority). This authority sits with the institution’s authorised organisational representative, not the principal investigator alone.
    • Second, NIH prior-approval NCE: a second extension, or any extension where the unilateral authority does not apply, requires a formal request to the assigned NIH awarding component, submitted well before the current project period ends, with a justification tied to the scientific rationale — not simply “more time is needed.”

    The distinction matters because a unilateral NCE that should have gone through prior approval is a compliance finding waiting to happen. Institutions that log every NCE — automatic and approved — in a central tracking system tend to catch these errors before an NIH grants management specialist does.

    Carryover: The Companion Concept to No-Cost Extensions

    Carryover is the mechanism that lets unobligated funds from one budget period move into the next. It is distinct from a no-cost extension, though the two are frequently requested together: a project that needs more time to spend down its budget usually also needs the unspent funds to travel with it.

    Under the NIH Grants Policy Statement, carryover authority depends on the type of award:

    • Awards issued under the Streamlined Non-Competing Award Process (SNAP) generally carry automatic carryover authority, meaning unobligated balances may move forward without a separate request.
    • Non-SNAP awards, and awards with specific restrictive terms, require prior NIH approval before carryover funds can be obligated in the next budget period.

    Administrators should treat carryover and NCE requests as related but separate compliance events. A common error is assuming that approval of one automatically covers the other — it does not. Each has its own authority basis in the GPS, and each needs its own documentation trail in the institution’s grants management system.

    Consortium Agreements and Subrecipient Monitoring

    Multi-institutional NIH awards are structured through consortium agreements — legally binding documents between the prime recipient and each participating (subrecipient) institution. The NIH Grants Policy Statement requires that these agreements flow down all applicable federal terms, including the prime award’s specific conditions, and that the prime recipient carry out subrecipient monitoring consistent with the Uniform Guidance (2 CFR 200.332).

    Where searches for “nih grants policy statement” cluster most heavily around consortium terms is usually at renewal or no-cost extension time, because an NCE on the prime award does not automatically extend the subaward — each consortium agreement needs its own amendment, on its own timeline, coordinated by the prime institution’s sponsored programs office. Institutions that manage large multi-site NIH awards (common in clinical trials networks and multi-PI R01s) typically maintain a master tracking sheet cross-referencing prime award dates against every subaward’s period of performance, precisely to avoid a subrecipient continuing work after its own agreement has technically lapsed.

    Program Income: An Often-Overlooked GPS Category

    Program income is gross income earned by the recipient that is directly generated by a supported activity, or earned as a result of the award — for example, fees from clinical services delivered through a research protocol, registration fees at a conference funded by the grant, or proceeds from the sale of research-generated data or materials. The NIH Grants Policy Statement sets out three methods for handling it: deduction (reducing NIH’s contribution), addition (adding it to the project’s committed funds), or cost-sharing/matching. Which method applies is usually specified in the notice of award, and applying the wrong method is a recurring finding in post-award reviews.

    Because program income rules interact with the same budget period concept that governs carryover and no-cost extensions, administrators processing an NCE should verify whether unspent program income also needs to be reported and carried forward under the applicable method.

    What This Means for Research Administrators

    The terms in this glossary are not abstract definitions — each one triggers a distinct approval workflow, a distinct system-of-record entry, and in some cases a distinct federal reporting obligation. The practical implications for institutional research offices are:

    • Track authority levels separately. Unilateral NCE authority, prior-approval NCE, automatic carryover, and prior-approval carryover each rest on different sections of the GPS. Conflating them in institutional policy documents is a common source of audit findings.
    • Coordinate consortium timelines actively. A no-cost extension on the prime award is not self-executing across subawards; each consortium agreement needs its own amendment.
    • Build cross-training between NIH and non-NIH portfolios. Institutions managing both NIH grants and awards from bodies such as UKRI, Horizon Europe, or other funders that researchers may apply for grants through, benefit from a single internal glossary that maps equivalent-but-different terms (for example, NIH’s “no-cost extension” against a UK funder’s “extension request”) so that PIs moving between funding streams are not caught out by assuming identical rules.
    • Document the scientific justification, not just the administrative request. NIH awarding components scrutinise the rationale behind prior-approval NCE and carryover requests; a request that reads as purely administrative is more likely to be questioned or delayed.

    These distinctions also matter for research integrity and reproducibility more broadly. Standards bodies such as NISO and initiatives supported by SPARC and cOAlition S increasingly expect that data management, sharing, and reporting obligations tied to federal awards are tracked with the same rigour as the funding mechanics themselves — an NCE that extends a project’s timeline, for instance, typically extends the data-sharing and reporting deadlines tied to NIH’s data management and sharing policy as well.

    A Reference Point, Not a Substitute for the Statement Itself

    No glossary replaces a careful reading of the current NIH Grants Policy Statement, which is updated periodically and should always be consulted in its original form for the specific fiscal year and award terms in question. What a reference like this can do is give research administrators — particularly those new to federal awards, or those managing a mixed portfolio spanning NIH grants and non-US funders — a shared vocabulary for the conversations that recur every budget cycle: whether an extension is unilateral or requires approval, whether carryover is automatic or restricted, whether a consortium agreement needs its own amendment, and how program income should be handled once it appears on the books.

    As federal reporting requirements continue to tighten and data-sharing obligations extend further into the post-award period, the administrative precision this glossary describes is likely to matter more, not less. Institutions that build internal reference materials mapping these terms — and keep them current against each GPS revision — will spend less time reconciling compliance findings after the fact.

  • UKRI’s New Funding Model Explained: What Research Administrators Need to Know for 2026

    UK Research and Innovation (UKRI) is in the middle of the most significant overhaul of its application and grants-management infrastructure in a generation. The UKRI new funding model consolidates the disparate systems that research offices have used for years — most notably the ageing Je-S (Joint Electronic Submission) system — into a single, browser-based UKRI Funding Service, alongside a redesigned Funding Finder for discovering live opportunities. For institutions managing a share of UKRI’s more than £8 billion in annual research and innovation funding, this is not a cosmetic IT upgrade. It changes how proposals are drafted, how organisational data is captured, and how compliance is monitored across the grant lifecycle.

    For research administrators, the practical stakes are high. Application routes, data fields, and reporting obligations that have been stable — if clunky — for over a decade are being replaced council by council, scheme by scheme. Getting institutional processes aligned to the new platform, and doing so with consistent metadata practices, is now a live operational priority heading into the 2026 cycle.

    What the UKRI New Funding Model Actually Changes

    At its core, the transition moves application and post-award activity away from Je-S and onto the UKRI Funding Service, a modern, cloud-based platform intended to serve all seven research councils, Research England, and Innovate UK from a common front end. UKRI has been rolling this out incrementally: rather than a single cutover date, individual funding opportunities and schemes have migrated in phases, meaning many research offices are currently operating a hybrid environment — some calls still routed through Je-S, others exclusively through the new service.

    Alongside the application platform itself, UKRI has overhauled UKRI Funding Finder, the public search tool researchers and administrators use to identify open and forthcoming calls. The redesigned finder is intended to reduce the friction of tracking opportunities across nine constituent councils with historically inconsistent naming conventions, deadlines, and eligibility criteria. Institutions that previously relied on manually curated spreadsheets or council-specific alerts to track UKRI funding opportunities are being encouraged to migrate to the centralised finder and its associated notification tools.

    The stated ambition behind the wider UKRI funding portal consolidation is simplification: fewer bespoke application forms, a common set of reusable organisational and project data, and — eventually — a single sign-on and case-tracking experience regardless of which council is funding the work. UKRI has also continued to update its standard UKRI terms and conditions for research grants, clarifying obligations around data management, open access compliance, and financial reporting as part of the same modernisation effort.

    Why This Is a Governance-Standards Issue, Not Just an IT Migration

    Research offices tend to experience system migrations as a training and helpdesk problem. That framing understates what is actually at stake. Every time a funder changes its application architecture, it also changes — implicitly or explicitly — the data model institutions must feed into it: how principal investigators are identified, how organisational affiliations are recorded, how costings and terms are structured, and how outputs and outcomes are reported back.

    This is precisely the territory that research information management standards exist to govern. Persistent identifiers such as ORCID for individual researchers and the Research Organization Registry (ROR) for institutions reduce the ambiguity that free-text name fields introduce into any funder’s database — ambiguity that becomes costly when a research office has to reconcile its internal systems (CRIS, HR, finance) against a funder’s records at renewal, reporting, or audit time. Where a new platform like the UKRI Funding Service standardises identifier capture at the point of application, it materially reduces downstream reconciliation work for research offices — provided institutions actually populate those fields consistently rather than treating them as optional.

    The same logic applies to reporting metadata more broadly. Bodies such as DataCite and CrossRef have spent over a decade demonstrating that consistent, machine-readable metadata at the point of creation is far cheaper than retrospective clean-up. A funder-level platform consolidation is an opportunity to embed that discipline institutionally — or, if handled poorly, an opportunity to multiply inconsistency across an even larger data set.

    Practical Steps for Institutional Research Offices

    Ahead of further rollout through 2026, research offices should treat the transition as a data-governance exercise as much as a systems one:

    • Audit current Je-S versus Funding Service coverage. Maintain an internal register of which schemes and councils have migrated, since guidance and support arrangements differ between the two systems during the transition period.
    • Standardise identifier capture now. Ensure ORCID iDs are mandatory in internal proposal-approval workflows for all named investigators, and confirm the institution’s ROR identifier is used consistently wherever organisational affiliation is recorded, rather than relying on free-text institution names.
    • Re-brief costing and contracts teams on updated terms and conditions. Changes to UKRI’s standard grant terms and conditions can affect audit obligations, data-sharing requirements, and financial reporting deadlines; contracts staff should not assume prior guidance still applies unchanged.
    • Update internal training materials and templates. Application forms, checklists, and costing templates built around Je-S field structures may no longer map cleanly onto the Funding Service’s data model.
    • Assign ownership for Funding Finder monitoring. As legacy alert mechanisms are retired, someone in the research office should own migration to the new finder’s notification tools to avoid missed deadlines during the changeover.

    What This Means for Research Administrators

    The immediate operational burden of the UKRI new funding model falls on research offices, not on principal investigators. Administrators are the ones reconciling which schemes sit on which platform, retraining academic staff on new interfaces, and absorbing the ambiguity of a multi-year, phased rollout rather than a single clean cutover. That burden is real, but it is also time-limited and manageable with the right internal governance discipline.

    The larger, more durable implication is about data quality. Institutions that use this transition as a forcing function to standardise identifier use, tidy internal CRIS records, and align local metadata practices with the fields the Funding Service expects will find themselves with cleaner, more defensible data for REF-related reporting, funder audits, and institutional benchmarking well beyond this migration. Those that treat it purely as a login-page change will likely recreate today’s reconciliation headaches inside tomorrow’s system.

    For sector bodies such as ARMA, NCURA, and EARMA, the consolidation also strengthens the case for shared guidance on identifier practice and metadata hygiene across UK and international funders — reducing the risk that every major funder’s platform change becomes a bespoke crisis for research offices to absorb alone.

    Looking Ahead

    UKRI’s platform consolidation will continue in phases through 2026 and beyond, and further schemes are expected to move from Je-S onto the Funding Service as confidence in the new platform grows. Research offices that get ahead of the transition — auditing coverage, enforcing identifier discipline, and refreshing internal guidance on the current terms and conditions — will be better placed to absorb whatever comes next, whether that is further UKRI consolidation or similar modernisation efforts from other national funders following the same trajectory.

  • Detecting Paper Mills: How Contribution Taxonomy Can Flag Implausible Authorship Patterns

    Paper mill fraud academic publishing schemes have moved from a peripheral integrity concern to a systemic threat that publishers, funders and institutions can no longer treat as isolated incidents. Retraction Watch and COPE have both documented a sharp rise in bulk retractions tied to fabricated manuscripts, often submitted in coordinated batches across unrelated journals. What is changing in 2026 is not just the scale of the problem but the toolkit available to detect it — and structured contributor role data is emerging as one of the more promising, and underused, signals.

    The CRediT taxonomy — CASRAI originated the CRediT contributor role taxonomy in 2014, and the standard is now stewarded by NISO as ANSI/NISO Z39.104-2022 — was designed to make authorship transparent by breaking a byline down into fourteen discrete roles, from Conceptualization and Data Curation through to Writing – Review & Editing. That transparency has an unintended but valuable side effect: it produces machine-readable metadata that can be pattern-matched at scale. Where a narrative acknowledgements paragraph hides inconsistency, a structured taxonomy exposes it.

    Why Paper Mill Fraud Academic Publishing Schemes Rely on Authorship Opacity

    Paper mills manufacture manuscripts, sell authorship slots, and route submissions through compromised peer review to generate publication credit for buyers who had no genuine involvement in the work. The business model depends on authorship remaining a black box: a byline lists names, not verifiable contributions. Editors and integrity teams investigating suspected paper mill fraud academic publishing cases have historically had to rely on tell-tale linguistic artefacts, template phrasing, image duplication, or reviewer-ring detection — all after-the-fact forensic work.

    Structured contribution data changes the economics of that opacity. When a submission system requires every author to declare a CRediT role, a paper mill operator selling a middle-authorship slot must also assign that buyer a plausible role. In practice, mills tend to default to generic, low-specificity roles — commonly Writing – Review & Editing or Supervision — applied uniformly across large numbers of authors who otherwise share no institutional, disciplinary or geographic connection. That uniformity is itself a signal: genuine multi-author teams typically show role differentiation that tracks the actual division of labour.

    Statistical Signatures of Implausible Authorship

    Several patterns recur across research misconduct case studies involving suspected paper mills, and each becomes more detectable once contribution roles are captured as structured fields rather than free text:

    • Role clustering without task correlation. A disproportionate share of authors assigned identical roles (for example, every author credited with Formal Analysis) on a manuscript whose subject matter would not plausibly require that many analysts.
    • Absent core roles. No author credited with Conceptualization, Methodology or Data Curation — the roles most directly tied to originating and executing a study — while several are credited with Writing or Supervision, roles more easily claimed without hands-on involvement.
    • High author-to-role ratio with low role diversity. A large author list mapping onto only two or three of the fourteen CRediT categories, rather than the fuller spread expected of a genuinely collaborative project.
    • Cross-manuscript author recombination. The same small pool of names recurring across dozens of ostensibly unrelated manuscripts, each time in a similar and narrow role, submitted to a similar cluster of journals within a short window.
    • Institutional and disciplinary mismatch. Authors credited with roles requiring domain expertise (Methodology, Investigation) whose institutional affiliation or publication history shows no prior connection to the field.

    None of these signatures is individually conclusive — legitimate collaborations can produce unusual role distributions, particularly in large consortium studies. But taken together, and cross-referenced against ORCID identifiers, CrossRef metadata and DataCite records, they give integrity teams a quantitative starting point rather than a purely qualitative hunch. This is precisely the shift that distinguishes structured taxonomy-based detection from earlier approaches focused only on fabrication, falsification and plagiarism as textual or image-level artefacts.

    From Fabrication Detection to Contribution Forensics

    Fabrication, falsification and plagiarism have long been the three canonical types of research misconduct recognised by funders and integrity offices, including in frameworks referenced by ICMJE and COPE guidance. Data fabrication in research — inventing results that were never generated — remains the most damaging category because it corrupts the evidence base itself, not just the credit attached to it. Paper mills frequently combine fabricated datasets with fabricated authorship, which is why contribution metadata analysis complements rather than replaces existing image-forensics and statistical-anomaly tools (such as those used to detect duplicated Western blots or implausible p-value distributions).

    What contribution taxonomy adds is a layer that operates before peer review even begins. A submission platform that enforces CRediT declaration at manuscript intake can flag anomalous role patterns automatically, routing suspect submissions for enhanced editorial scrutiny before reviewer time is spent at all. Some publishers already screen for reviewer-author citation rings and template language; extending that screening to role-distribution analysis is a logical next step, and one that requires no new standard — only consistent enforcement of the existing ANSI/NISO Z39.104-2022 taxonomy at the point of submission.

    What This Means for Research Administrators

    For research administrators, integrity officers and institutional leaders, the implications are practical rather than theoretical:

    • Mandate CRediT at institutional repositories, not only at journals. Institutions that require CRediT statements for internal reporting — REF 2029 preparation being one UK driver — build a parallel dataset that can be cross-checked against journal-level declarations for inconsistency.
    • Treat role data as an integrity input, not a formatting requirement. Research integrity offices should incorporate contribution-role review into misconduct triage workflows alongside existing checks for duplicate publication and undisclosed conflicts of interest.
    • Pair CRediT with persistent identifiers. The detection value of contribution data depends on being able to link it reliably to a real, verifiable researcher. ORCID iDs and ROR-identified institutional affiliations are what make cross-manuscript pattern analysis possible at all.
    • Expect funder scrutiny to increase. As UKRI, Horizon Europe and NIH tighten data-sharing and integrity expectations, evidence of contribution-level authorship verification is likely to become part of what funders expect institutions to demonstrate, not merely journals.
    • Build internal case libraries. Documented research misconduct case studies — including near-misses caught by role-pattern anomalies — help integrity committees calibrate thresholds and avoid both false accusations and missed detections.

    None of this requires new technology from research offices themselves. It requires consistent adoption of an existing, freely available taxonomy, and a willingness to treat authorship metadata as something that can — and should — be audited.

    A Structural Response to a Structural Problem

    Paper mill fraud academic publishing is fundamentally a structural exploit: it takes advantage of the gap between what a byline claims and what actually happened during a study. Structured contribution data narrows that gap without requiring investigators to prove intent or reconstruct a fabricated dataset from scratch — it simply makes the shape of a suspicious authorship list visible. As adoption of ANSI/NISO Z39.104-2022 continues to widen across journals, preprint servers and institutional repositories, the marginal cost of running this kind of pattern analysis keeps falling, while the cost of ignoring it — in retractions, reputational damage and wasted funder investment — keeps rising.

    The taxonomy will not stop paper mills on its own. Coordinated action from publishers, funders, and integrity bodies such as COPE remains essential, alongside continued scrutiny from watchdogs like Retraction Watch. But for an integrity ecosystem that has historically had to detect fraud after the fact, contribution-role metadata offers something rarer: a signal available before a flawed paper ever reaches print.

  • Retraction Transparency: Why Author Contribution Statements Are Now a Governance Tool

    Retraction in academic publishing has moved from a rare editorial embarrassment to a routine, high-volume governance problem. Springer Nature disclosed that it retracted 1,462 papers across its portfolio in 2025 — a figure that, alongside a January 2026 Nature editorial on the subject, has reframed how publishers, institutions and funders think about the paper trail behind a byline. The editorial’s central argument was not that retraction volumes are alarming in isolation, but that the publishing ecosystem still lacks a reliable, structured way to establish who did what on a paper once something goes wrong.

    That gap is precisely where structured author contribution disclosure now sits. What began in 2014 as a mechanism for fairer academic credit is increasingly being read by editors, integrity officers and institutional research offices as something closer to an audit trail: a documented record of who contributed which specific tasks to a study, retrievable long after publication when a correction, expression of concern or retraction notice becomes necessary.

    This shift matters for research administrators specifically, because it changes what “good practice” looks like at the point of submission. Contribution statements are no longer a courtesy line for the acknowledgements section — they are becoming evidence that institutions and journals may need to produce during a formal investigation.

    Retraction in Academic Publishing: From Rare Event to Routine Governance Signal

    The scale of the Springer Nature figure is instructive. A number in the thousands, drawn from one large multi-journal publisher in a single year, signals that retraction of research papers has become a standing feature of scholarly communication rather than an exceptional event confined to high-profile fraud cases. Retraction Watch has tracked this trend for years through its public database, documenting causes ranging from image manipulation and data fabrication to authorship disputes, undisclosed conflicts of interest and — increasingly — undisclosed or improper use of generative AI in manuscript preparation.

    What the January 2026 Nature editorial added to this picture was a governance argument: publishers cannot investigate at this volume using ad hoc correspondence and memory. Journals need structured, machine-readable metadata about contribution and responsibility captured at submission, not reconstructed after the fact from email threads and co-author recollection. Research integrity issues surface months or years after publication, often when the researchers involved have moved institutions, changed collaborators, or in some cases become uncontactable. A contribution statement recorded at submission time, tied to a persistent identifier, survives all of that.

    Why Contributor Role Taxonomies Function as Audit Trails

    This is where the Contributor Roles Taxonomy becomes relevant to integrity investigations rather than just credit allocation. CASRAI originated the CRediT contributor role taxonomy in 2014. The standard is now stewarded by NISO as ANSI/NISO Z39.104-2022. Its fourteen defined roles — including conceptualisation, data curation, formal analysis, investigation, methodology, and writing (original draft, review and editing) — were designed to solve a credit-allocation problem: junior researchers, data specialists and methodologists were frequently under-recognised by the traditional single-line “authorship” convention.

    The governance value is a by-product of that original design. When a journal applies structured roles at submission, and links each contributor to an ORCID identifier, it creates a queryable record: who claimed responsibility for the statistical analysis, who curated the dataset, who wrote the manuscript. If a subsequent research integrity investigation identifies fabricated data in a specific figure, that record narrows the inquiry to the individuals who claimed the relevant roles — data curation, formal analysis, investigation — rather than treating the full author list as equally implicated or equally responsible. This is precisely the function COPE (Committee on Publication Ethics) guidance has long urged: that authorship and contribution be documented in a way that supports fair, evidence-based adjudication of misconduct allegations, rather than blanket sanctions against every listed author.

    COPE Guidelines and the Documentation Gap

    Retraction guidelines COPE has published over the past decade consistently emphasise two things: that retraction decisions should follow a documented, defensible process, and that all listed authors should be given the opportunity to respond before a notice is issued. Both requirements depend on knowing, precisely, who is responsible for what. In practice, many journals still rely on a single free-text contribution paragraph — “X and Y designed the study; Z performed the experiments” — that is neither standardised nor easily machine-searchable across a portfolio of thousands of papers.

    Structured CRediT statements close that gap. Because the taxonomy uses a fixed, finite set of roles, publishers can query metadata at scale: which papers list a given researcher under “data curation,” across how many journals, in how many retracted studies. Crossref and DataCite metadata schemas already support structured contributor role fields, meaning this information can, in principle, travel with the persistent identifier record rather than remaining locked inside a PDF. That is the technical foundation an audit-trail function requires — and it is largely already in place; the remaining barrier is consistent adoption and consistent metadata deposit by journals and platforms.

    What This Means for Research Administrators

    For institutional research offices, this shift has practical consequences that go beyond publisher policy:

    • Institutional research integrity offices should expect to be asked for contribution records during misconduct investigations initiated by journals or funders, not just the reverse. Retaining structured contribution metadata alongside grant and output records strengthens an institution’s ability to respond quickly and specifically.
    • ORCID linkage is no longer optional infrastructure. With ORCID adoption now effectively mandated across most major funders and publishers, institutions should ensure researcher profiles are current and that contribution claims on outputs are verified, not simply self-reported and forgotten.
    • Authorship disputes should be resolved before submission, using structured role assignment as the basis for discussion, rather than settled informally and revisited only when a correction becomes necessary.
    • Research integrity training should reference contribution statements explicitly, framing them as a professional and accountability record, not an administrative afterthought completed in the final minutes before submission.
    • Institutions preparing for REF 2029 and equivalent national assessment exercises should treat consistent, verifiable contribution metadata as an asset that supports both credit allocation and defensibility should any submitted output later face scrutiny.

    The direction of travel is consistent with wider open science governance trends — UKRI’s evolving open access policy, NIH data sharing enforcement, and Horizon Europe’s research integrity expectations all point toward increased structured disclosure as a condition of funding and publication, not a voluntary enhancement.

    Conclusion: Structured Disclosure as Standard Practice

    Retraction in research has always carried reputational weight for the individuals and institutions involved; what has changed is the expectation that the process leading to a retraction decision be documented, structured and defensible from the outset. The Springer Nature 2025 figure and the Nature editorial that followed it are unlikely to be the last signals in this direction. As integrity investigations grow in frequency and complexity — compounded by emerging challenges around AI-assisted manuscript preparation and image generation — publishers, funders and institutions will continue to look for metadata standards that were originally built for credit and recognition, and increasingly ask them to do double duty as accountability infrastructure. Contribution taxonomies stewarded through formal standards bodies, deposited consistently in persistent-identifier metadata, and linked to verified researcher identifiers are best positioned to meet that demand.

  • What the cOAlition S 2026-2030 Strategy Means for Your Institution’s Open Access Policy

    In November 2025, cOAlition S — the funder alliance behind Plan S — published its strategic direction for 2026 to 2030, and the cOAlition S strategy 2026 document marks the most significant recalibration of the coalition’s open access policy since Plan S launched in 2018. For research offices, the headline is not that open access requirements are loosening. It is that the coalition is walking away from the transformative agreements it once tolerated as a bridge to full open access, while redirecting funder investment toward diamond open access and reaffirming, unchanged, its core mandate for immediate open access under a CC BY licence.

    For institutions that built compliance workflows around transformative “read and publish” deals, this is not a minor administrative footnote. It changes which routes to compliance your researchers can rely on, which publisher agreements your library or research office should keep renewing, and where new funding lines for open infrastructure are likely to appear over the next four years. This explainer sets out what has actually changed, what has stayed the same, and what research administrators should be updating in internal open access policy documents now.

    cOAlition S 2026 to 2030 strategy: the end of transformative agreement support

    Transformative agreements — the large “read and publish” or “publish and read” contracts negotiated between library consortia and major publishers such as Elsevier, Springer Nature and Wiley — were always framed by cOAlition S as a temporary compliance route, not a destination. They allowed institutions to satisfy Plan S requirements without every researcher negotiating individual article processing charge (APC) waivers or rights-retention clauses, on the understanding that publishers would use the revenue to convert subscription journals to fully open access titles within a defined period.

    The cOAlition S 2026 to 2030 strategy confirms that this transitional support is over. Funders in the coalition will no longer treat transformative agreements as an acceptable long-term compliance mechanism, reflecting a long-standing concern, raised repeatedly since the original Plan S implementation guidance, that many transformative deals converted little content to genuine open access while entrenching APC-based publishing economics at large commercial publishers. Institutions that have relied on a transformative agreement as their primary Plan S compliance route should treat this as a signal to diversify, not a cliff-edge deadline to panic over, but the direction of travel is unambiguous: the coalition wants funded researchers publishing through routes that do not depend on subscription-linked bundled contracts.

    Expanding diamond open access investment

    The counterpart to withdrawing support from transformative agreements is an expansion of funder investment in diamond open access — journals and platforms that charge no fee to authors and no subscription fee to readers, typically run by universities, learned societies, library consortia or national research infrastructures rather than commercial publishers. This builds on work already under way through the Action Plan for Diamond Open Access and the DIAMAS project, which have mapped the diamond OA landscape across Europe and identified capacity, quality-assurance and sustainability gaps in community-owned publishing.

    For research offices, the practical implication is that the menu of “approved” publication venues under funder open access policy is likely to broaden to include more diamond OA journals and platforms, some of which may be less familiar to academic staff than established commercial titles. Research administrators should expect funders to ask institutions to help identify, vet and promote diamond OA venues in their disciplines, and to factor diamond OA into researcher guidance rather than treating APC-based gold open access as the default paid route to compliance.

    What hasn’t changed: the immediate CC BY mandate

    It would be a mistake to read the cOAlition S 2026 to 2030 strategy as a loosening of open access requirements. The core Plan S principle — that research arising from coalition-funded grants must be made immediately open access, with no embargo, under a CC BY licence (or, in defined circumstances, CC BY-ND) — remains fully in force. So does the Rights Retention Strategy, which allows authors to apply an open licence to the author accepted manuscript at the point of submission, regardless of the publisher’s own policy, precisely so that researchers are not locked into paying an APC or relying on a transformative agreement to comply.

    In other words, Plan S gold open access — publishing in a fully open journal, whether APC-funded or diamond — remains one valid compliance route, as does green open access via rights retention and immediate deposit in a repository. What the 2026 to 2030 strategy removes is institutional and publisher reliance on transformative agreements as a proxy for compliance, and what it adds is funder appetite to underwrite non-APC alternatives so that gold open access is no longer synonymous with paying a commercial publisher.

    cOAlition S members and the wider funding landscape

    cOAlition S members include UK Research and Innovation (UKRI), the Austrian Science Fund (FWF), France’s Agence Nationale de la Recherche (ANR), the Dutch Research Council (NWO), Science Foundation Ireland, Luxembourg’s Fonds National de la Recherche, and the European Commission acting on behalf of Horizon Europe, among other national and regional funders. Because these funders between them cover a substantial share of publicly funded research across Europe, a strategic shift of this kind has effects well beyond the institutions directly funded by coalition members — publishers and platforms that want to remain viable routes to compliance for European-funded authors have to adapt their offerings accordingly, and non-coalition funders elsewhere often watch cOAlition S’s direction when reviewing their own mandates.

    Research offices supporting international, multi-funder projects should note that not every funder a researcher reports to will move at the same pace. Where a project is co-funded by a cOAlition S member and a funder with a different open access policy, the more stringent requirement typically governs compliance — so the withdrawal of transformative agreement support from coalition funders does not automatically loosen obligations under, for example, UKRI’s own open access policy or NIH data-sharing requirements, which continue to operate on their own terms.

    What This Means for Research Administrators

    Institutional open access policies, researcher guidance pages and library-negotiated agreement lists were largely written on the assumption that transformative agreements were a stable, funder-endorsed compliance route. That assumption no longer holds for cOAlition S-funded research. Research offices should treat the following as near-term priorities:

    • Audit compliance messaging. Update internal guidance that lists transformative agreements as a “safe” route to Plan S compliance; reframe them as one option among several, with declining funder endorsement.
    • Reconfirm rights retention workflows. Ensure researchers and grants staff understand that applying a CC BY licence to the author accepted manuscript at submission, under the Rights Retention Strategy, remains a robust and funder-preferred compliance path that does not depend on any publisher contract.
    • Map diamond OA venues in relevant disciplines. Work with library and repository teams to identify credible diamond open access journals and platforms so researchers have a vetted list rather than discovering options ad hoc.
    • Flag co-funding conflicts early. Where projects have mixed funding sources, confirm which policy is most stringent before advising researchers on their compliance route.
    • Watch publisher responses. Expect publishers with large transformative agreement portfolios to adjust pricing, waiver schemes or their own diamond and community-publishing offerings in response to reduced funder backing — track these changes when renewing agreements.
    • Brief research leadership. Associate deans for research and heads of library services should be aware that funder-endorsed compliance routes are shifting, ahead of the next round of institutional open access policy review.

    Conclusion

    The cOAlition S strategy 2026 announcement does not soften Plan S; it sharpens the coalition’s original argument that open access should not simply reproduce commercial subscription economics under a new name. By withdrawing endorsement from transformative agreements and channelling investment toward diamond open access, cOAlition S is betting that community-owned, non-APC publishing can scale to meet demand that was previously absorbed by bundled publisher contracts. Whether that bet pays off will depend on how quickly diamond OA infrastructure can absorb volume without compromising editorial quality or long-term sustainability — a question the Action Plan for Diamond Open Access and DIAMAS project were designed to help answer.

    For research administrators, the practical task is less about reacting to a single announcement and more about maintaining policy documents that can flex as funder expectations continue to evolve through 2030. Institutions that keep rights retention workflows well understood, diamond OA options mapped, and researcher guidance current will be best placed to support compliance regardless of which route to open access their funders favour next.

  • What the ORI 2024 Final Rule Means for Research Administrators in 2026

    Two decades after its last comprehensive revision, the ORI research misconduct policy that governs how US institutions handle allegations against Public Health Service (PHS)-funded researchers has been rewritten. The Department of Health and Human Services’ Office of Research Integrity (ORI) finalised sweeping changes to 42 CFR Part 93 in September 2024, and the compliance clock is now running out for research administrators. For any allegation an institution receives on or after 1 January 2026, the new rule — not the 2005 version still in force today — governs the entire process, from first inquiry to final institutional finding.

    For research integrity officers (RIOs), sponsored programmes staff, and institutional counsel, this is not a distant regulatory footnote. It is an operational deadline that requires revised bylaws, retrained committees, and updated PHS assurances before the new year turns. Institutions that received PHS funding any time in the last twenty years will recognise the stakes: 42 CFR Part 93 is the backbone of every federally compliant misconduct policy at universities, medical centres, and research institutes across the country, and non-compliance can jeopardise an institution’s standing to receive further PHS awards.

    This piece sets out, in plain language, what changed, when it takes effect, and what research administrators need to have in place before the applicability date arrives.

    Why ORI Rewrote Its Research Misconduct Policy

    The federal research misconduct regulations at 42 CFR Part 93 had not been substantively updated since 2005. In the intervening years, research practice changed considerably: the 2008 NIH Public Access Policy, the 2023 NIH Data Management and Sharing Policy, the migration of research records to cloud storage, and the emergence of artificial intelligence tools capable of detecting image manipulation and data falsification all outpaced a regulation written for a largely paper-based research environment. ORI’s final rule, published in the Federal Register in September 2024, is the agency’s attempt to modernise federal research misconduct policy for that reality.

    Crucially, the amended rule does not retroactively apply to open cases. Institutions must continue to apply the 2005 version of 42 CFR Part 93 to any allegation received before 1 January 2026, unless the respondent and the institution jointly elect in writing to proceed under the new rule instead. This creates a transitional period in which RIOs may be running two parallel sets of procedures — a detail that governance documents and training materials need to reflect clearly.

    Revised Inquiry and Investigation Timelines

    The most immediately practical change for administrators is timing. Under the current regulation, institutions have 60 days to complete an inquiry and roughly 120 days to complete an investigation once a formal case is opened. The final rule extends both phases: the inquiry period is lengthened to 90 days, and the investigation period to 180 days, giving RIOs and investigation committees materially more runway to conduct a thorough, defensible review before a finding is reached.

    The final rule also formalises procedural protections around the investigation report. Respondents must be given an opportunity to review the draft investigation report and to submit written comments — generally within 30 days of receipt — before the institution finalises its findings. Notably, ORI declined to adopt a provision from its proposed rule that would have barred institutions from determining “honest error” at the inquiry stage; that flexibility survives into the final text, giving institutions more room to close weak allegations early without triggering a full investigation.

    These extended timelines should be reflected in every institutional misconduct policy, in RIO case-tracking templates, and in any communications sent to complainants and respondents about expected case duration.

    Updated Definitions and New Assurance Requirements

    The revised PHS research misconduct regulations introduce more than 25 defined terms that did not previously exist in the regulatory text, including “institutional record,” “administrative record,” “intentionally,” and “recklessly.” Institutional record is now explicitly defined as the records an institution actually compiled or generated during a misconduct proceeding — excluding materials it never considered or relied upon — which narrows disputes over what must be produced during appeals or ORI oversight review.

    The definition of plagiarism has also been sharpened to explicitly exclude self-plagiarism (text recycling) and authorship disputes from the scope of research misconduct, redirecting those matters to separate institutional processes such as publication ethics review or authorship-contribution reconciliation. This is a useful distinction for institutions that already document individual contributor roles using frameworks such as the CRediT taxonomy — CASRAI originated the CRediT contributor role taxonomy in 2014, and the standard is now stewarded by NISO as ANSI/NISO Z39.104-2022 — since clear contribution records can help RIOs quickly separate a genuine authorship dispute from a fabrication or plagiarism allegation.

    On the assurance side, institutions must update their PHS Research Integrity Assurance and supporting policies and procedures to align with the final rule, and reflect that update in the annual report covering 2025 activity, due to ORI on or before 30 April 2026. ORI has also expanded and clarified its jurisdictional reach — the office of research integrity jurisdiction now extends more explicitly to applicants, subawardees, and PHS-supported research regardless of the specific funding mechanism, closing gaps that previously left some collaborative and subcontracted research outside clear oversight.

    What This Means for Research Administrators

    The practical to-do list for institutions receiving PHS funding is substantial:

    • Revise institutional policy documents to reflect the 90-day inquiry and 180-day investigation windows, the new defined terms, and the narrowed plagiarism definition.
    • Update the PHS Research Integrity Assurance on file with ORI and prepare to certify compliance in the annual report due by 30 April 2026.
    • Train RIOs, deciding officials, and investigation committee members on the revised definitions of intent, recklessness, and institutional record, since these terms carry direct evidentiary consequences.
    • Build a transitional protocol for distinguishing cases governed by the 2005 rule from those governed by the 2026 rule, including the joint-election mechanism for respondents who prefer the new procedures.
    • Coordinate with sponsored programmes and general counsel on subrecipient and subaward language, given the clarified jurisdictional scope over collaborative research.
    • Monitor ORI’s phased guidance releases, which began appearing in December 2025 and are continuing through 2026, for topic-specific implementation detail beyond the regulatory text itself.

    Professional associations that support research administrators — including NCURA and ARMA — have flagged the final rule as a priority training topic for 2026, and institutions should expect sample policy language and compliance checklists to continue circulating through those channels and through compliance-focused organisations such as CITI Program.

    A Modernised Baseline, Not a Finished Job

    The final rule brings 42 CFR Part 93 into closer alignment with how research is actually produced, stored, and scrutinised in 2026 — but it does not settle every open question. Guidance documents are still arriving in phases, institutions are still building parallel procedures for the transitional period, and the practical burden of proof around “intentionally” and “recklessly” will only be tested as new cases move through the revised process. For research administrators, the sensible posture is to treat 1 January 2026 not as a finish line but as the date from which institutional practice will be judged against a materially more detailed, and more demanding, federal standard.

  • Transformative Agreements Are Ending: What Publishers and Institutions Should Do Next

    From January 2025, cOAlition S stopped counting transformative agreements as compliant with Plan S by default, ending five years of tacit funder support for the mechanism that had allowed hybrid subscription journals to offer immediate open access under an interim licence. For publishers and institutions that had built entire open access strategies around read-and-publish deals, the shift is not academic. It forces a hard choice between converting to full open access, negotiating narrower independent agreements, or leaning more heavily on rights retention and repository deposit. Understanding the mechanics of a transformative agreement open access deal — and why funders have lost patience with it — is now essential for anyone managing compliance across a research portfolio.

    The timing matters. Transformative agreements were always framed as a bridge, not a destination: a way to move subscription publishers toward full open access without an abrupt shock to library budgets or editorial pipelines. cOAlition S set an explicit end date for that bridge, and 2025 is when the funding underpinning it formally lapses for many participating publishers. Institutions that assumed renewal on the same terms are now recalculating budgets, contracts, and compliance monitoring in real time.

    Why cOAlition S Withdrew Support for Transformative Agreements

    Plan S, launched by a coalition of research funders including UKRI and several European national agencies, always treated the Plan S hybrid journal as a transitional category. Hybrid titles — subscription journals that also offer an open access option article by article — were permitted under transformative agreements only because those agreements included measurable, time-limited commitments to flip fully to open access. cOAlition S’s own monitoring found that too few agreements were delivering credible transition trajectories at the pace originally envisaged, and continuing to fund the intermediate step indefinitely risked entrenching hybrid publishing rather than ending it.

    The practical effect from 2025 is that cOAlition S funders no longer treat compliance with a transformative agreement as automatically satisfying their open access policy. Authors funded by participating agencies must instead demonstrate compliance through one of the routes Plan S has always prioritised: publishing in a fully open access venue, publishing under an independently negotiated agreement that meets specific transparency and pricing criteria, or exercising rights retention to deposit the accepted manuscript in a repository regardless of the publisher’s own licensing terms.

    The Practical Options: Full-OA, Independent Deals, and Rights Retention

    For publishers, the most direct response is converting hybrid titles to fully open access — removing the subscription dependency entirely and pricing through article processing charges or diamond/no-fee models. This satisfies funders outright but carries real revenue risk for society and mid-sized publishers whose subscription base has not yet been replaced by APC income at scale.

    A second route is the independently negotiated read-and-publish deal: an agreement that continues to bundle reading access with publishing rights, but which is renegotiated outside the cOAlition S monitoring framework, typically at institutional or consortium level with more explicit transition milestones and price transparency. Several national consortia are already restructuring deals along these lines rather than abandoning read-and-publish arrangements altogether.

    The third, and arguably most consequential for institutions, is the cOAlition S Rights Retention Strategy. Under rights retention, authors apply a Creative Commons licence to their accepted manuscript at submission — regardless of where they ultimately publish — and deposit that manuscript in an institutional or subject repository, satisfying an open access repository mandate even when the journal itself remains subscription-based or hybrid. This decouples compliance from publisher business models entirely, which is precisely why funders favour it as transformative agreements wind down. The Wellcome open access policy and the ERC open access policy have both moved in this direction, requiring immediate deposit with an open licence rather than accepting embargoed green open access as sufficient.

    Metadata and Persistent Identifiers: Making Post-TA Workflows Auditable

    Whichever route an institution favours, the compliance burden shifts from the publisher’s contract terms to the research office’s ability to prove, at the level of the individual output, which policy was satisfied and how. That is a metadata problem as much as a policy problem. Funders auditing compliance without a central transformative agreement to point to need reliable, machine-readable evidence: a CrossRef DOI carrying accurate licence and funding metadata, an ORCID iD linking the author to the funder award, a Research Organization Registry (ROR) identifier confirming the affiliated institution, and a repository record with a resolvable identifier showing date of deposit and applied licence.

    This is where the infrastructure built around persistent identifiers earns its keep. A repository deposit under rights retention is only auditable if the manuscript record, the funder grant number, and the author identifier are all linked consistently — otherwise institutions cannot demonstrate compliance at scale across thousands of outputs, and funders cannot verify it either. DataCite-assigned identifiers for datasets and outputs, combined with ORCID and ROR, give institutions a way to generate compliance reports without relying on publisher-supplied dashboards that may disappear along with the agreement itself. Research offices that have invested in clean, well-populated metadata pipelines are considerably better positioned for this transition than those that treated identifier hygiene as a back-office afterthought.

    What This Means for Research Administrators

    Research administrators now need to treat open access compliance as a per-output audit trail rather than a portfolio-level contract. Practical priorities include:

    • Auditing which current agreements are genuinely at risk of losing funder recognition in 2025 and beyond, and flagging affected corresponding authors early.
    • Strengthening repository workflows so that rights-retention deposits happen at submission, not after acceptance, since embargo-based green open access no longer satisfies the strictest funder policies.
    • Ensuring ORCID, ROR, and grant metadata are captured consistently at the point of manuscript submission, not reconstructed retrospectively for a funder audit.
    • Briefing researchers, particularly early-career authors, on how rights retention interacts with journal submission terms, since some publishers still contest the strategy in their own author agreements.
    • Reviewing consortium-level negotiating positions where independent read-and-publish deals are being pursued as a replacement for cOAlition S-recognised transformative agreements.

    Institutions with mature research information management systems — those already aligned with standards from ORCID, DataCite, and CrossRef — will find this transition largely a configuration exercise. Those still relying on manual spreadsheets or publisher-reported compliance figures face a heavier lift, and a compressed timeline in which to build it.

    A Bridge That Has Reached Its Far Bank

    The end of cOAlition S funding for transformative agreements is not a retreat from open access — it is a signal that funders consider the transitional phase complete for a meaningful share of the market, and expect the sector to move to durable models: full open access, transparent independent agreements, or rights retention backed by repository deposit. The institutions and publishers that adapt fastest will be those that have already invested in the unglamorous infrastructure of identifiers and metadata, because that infrastructure is what turns a policy commitment into something a funder, an auditor, or a future researcher can actually verify.

    As with the CRediT contributor role taxonomy — originated by CASRAI in 2014 and now stewarded by NISO as ANSI/NISO Z39.104-2022 — the lesson is consistent: standards and identifiers only deliver value once they are consistently applied across the research lifecycle. Post-transformative-agreement open access will be judged less on rhetoric and more on whether every output can be traced, licensed, and verified.

  • AI and Academic Integrity: How Universities Are Drawing the Line Between Tool and Author

    A consensus has crystallised across the scholarly publishing and research-funding landscape over the past two years: generative artificial intelligence tools cannot be listed as authors on a research output, but their use must be disclosed. What remains unsettled — and what is now the focus of active AI authorship disclosure research among journals, universities, and funders — is exactly how that disclosure should be structured, who is responsible for verifying it, and how institutions distinguish genuine human intellectual contribution from AI-assisted production of text, code, data analysis, or images.

    The policy convergence is real. ICMJE guidance, COPE position statements, and publisher-level policies from major scholarly houses all now hold that large language models and generative AI systems fail the basic test of authorship: they cannot take responsibility for a work’s accuracy, cannot agree to be accountable for its integrity, and cannot hold the legal or ethical liability that authorship implies. What has proliferated instead is a patchwork of disclosure mechanisms — acknowledgement sections, dedicated “AI use” statements, methods-section declarations, and in some cases structured metadata — with no single format yet dominant. That divergence is precisely where contributor role frameworks are proving useful.

    Why AI Authorship Disclosure Research Is Accelerating in 2026

    Three pressures are driving institutions to formalise their approach faster than in previous years. First, the EU AI Act’s phased compliance timeline is pushing research-performing organisations to document AI use in a way that satisfies both scholarly integrity norms and emerging regulatory transparency obligations. Second, UK institutions preparing for the REF 2029 cycle are under pressure to demonstrate that submitted outputs meet originality and integrity standards that predate generative AI, which means research offices need defensible, auditable disclosure practices now rather than in 2028. Third, funders are beginning to ask more precise questions in grant reporting about how AI tools were used in proposal writing, data analysis, and manuscript preparation — a shift that reflects broader AI regulation research funding bodies are grappling with as they update terms and conditions.

    The practical effect is that “disclose AI use” is no longer sufficient as a policy statement. Research offices, journals, and funders are being asked to specify: disclose what, at what level of granularity, verified by whom, and recorded where. This is the gap that structured contributor taxonomies were originally built to close for human contributions — and it is why they are now being extended, cautiously, into AI governance conversations.

    Contributor Role Frameworks as a Principled Dividing Line

    The core conceptual tool available to institutions is not new. CASRAI originated the CRediT contributor role taxonomy in 2014. The standard is now stewarded by NISO as ANSI/NISO Z39.104-2022, and it defines fourteen discrete contribution types — from Conceptualization and Methodology to Writing – original draft and Writing – review and editing — each of which can be attributed to a named, accountable human contributor.

    The taxonomy’s original purpose was to disaggregate authorship credit among multiple humans on a single paper. Its underlying logic, however, generalises usefully to the AI question: authorship requires accountability for a specific, nameable contribution, and accountability requires an agent capable of bearing responsibility. A generative AI tool can plausibly be described as having assisted with tasks that map onto CRediT categories such as drafting text, generating code, or performing data curation — but it cannot occupy the CRediT role itself, because a role assignment implies the assignee can answer for the work’s validity under scrutiny, including retraction proceedings, correction requests, or research-integrity investigations. This is the principled basis journals and institutions are increasingly citing: the same contributor-role logic that separates a data-generating instrument from the humans who interpreted its output can separate an AI writing or coding assistant from the humans who directed, checked, and take responsibility for its use.

    Several publishers now ask authors to describe AI involvement in language that echoes CRediT categories — for example, specifying that a tool was used to support Writing – original draft but that Conceptualization, Formal analysis, and Writing – review and editing remained entirely human. This is a productive middle path: it does not require a new taxonomy, but it borrows the existing one’s granularity to make disclosure statements auditable rather than vague.

    Where Implementation Diverges

    Despite the shared principle, practical implementation varies considerably across the research ecosystem:

    • Placement of disclosure. Some journals require AI use statements in the methods section (treating it as a methodological detail); others require a separate acknowledgements-adjacent declaration; a smaller number embed it in cover letters reviewed only by editors, not readers.
    • Granularity required. Some policies accept a blanket statement (“generative AI was used to improve language and readability”); others, more aligned with CRediT-style precision, require task-level specification of which sections or functions involved AI assistance.
    • Tooling identification. A minority of policies require naming the specific tool and version used, which matters for reproducibility and for tracking model-specific error patterns, but raises practical questions when authors use multiple tools across a long project.
    • Verification mechanisms. Almost no institution has a reliable technical means of verifying that disclosed AI use is complete and accurate; disclosure remains largely an honour system underpinned by researcher attestation, similar to conflict-of-interest declarations.
    • Funder versus publisher scope. Funders such as UKRI and participants in cOAlition S are beginning to address AI use in grant terms, but their focus tends to sit upstream — on proposal preparation and data management plans — whereas publisher policies focus downstream, on the submitted manuscript. Institutions sitting between the two face a compliance gap where neither policy layer fully covers the research lifecycle.

    This divergence is not simply inconsistency for its own sake; it reflects genuinely different institutional risk profiles. A journal’s primary concern is the integrity of the published record. A funder’s primary concern is the integrity of the proposal and reporting process. A university’s research integrity office must satisfy both, plus internal disciplinary and REF-adjacent obligations, which is why many research offices are now building disclosure requirements that exceed the minimum asked by any single external stakeholder.

    What This Means for Research Administrators

    For research administrators, the practical task is less about resolving the philosophical authorship question — that consensus is largely settled — and more about operationalising disclosure consistently across a diverse portfolio of journals, funders, and disciplines. Several concrete steps follow from the analysis above:

    • Adopt CRediT-aligned language in institutional AI-use disclosure templates, so that researchers describe AI assistance using the same task-level vocabulary already familiar from authorship contribution statements, rather than inventing a parallel, less precise disclosure format.
    • Build AI disclosure into existing research integrity and authorship training rather than treating it as a standalone policy, since the underlying skill — accurately attributing who or what did what — is the same competency CRediT training already builds.
    • Track the funder-versus-publisher compliance gap explicitly in grant management workflows, particularly where UKRI or Horizon Europe-funded projects will also be submitted to journals with independent AI disclosure requirements.
    • Maintain records of AI-use disclosures in a form that could support a future research-integrity enquiry, given that verification remains attestation-based and institutions, not authors alone, may be asked to demonstrate due diligence.
    • Monitor evolving guidance from ORCID, DataCite, and CrossRef on whether AI-tool disclosure will eventually be captured as structured, machine-readable metadata rather than free-text statements — a shift that would materially change how research offices audit compliance at scale.

    This agenda sits squarely within the broader landscape of generative AI policy research institutions are now expected to maintain, alongside data management, open access, and research integrity policy suites. It also intersects with wider questions of AI ethics academic institutions face around equitable access to AI tools, disclosure burden on early-career researchers, and the risk that inconsistent policy enforcement disadvantages authors publishing across journals with conflicting requirements.

    A Settled Principle, an Unsettled Practice

    The authorship question itself is close to resolved: AI systems are tools, not authors, across every major scholarly integrity body’s current guidance. What remains genuinely in motion is the practice layer — disclosure format, granularity, verification, and cross-institutional consistency — and this is precisely where AI in research compliance functions will need to mature over the next several REF and grant-reporting cycles. Contributor role frameworks such as CRediT did not anticipate generative AI when devised in 2014, but their core discipline — mapping specific contributions to accountable agents — has turned out to be exactly the conceptual scaffolding institutions now reach for when drawing the line between tool and author. Research administrators who build that scaffolding into existing authorship and integrity workflows now will be far better placed than those who wait for a single global standard to arrive.

  • Plan S Compliance in 2026: A Practical Checklist for Research Administrators

    Seven years after cOAlition S launched Plan S, Plan S compliance has moved from a policy novelty to a routine administrative discipline — but “routine” does not mean simple. As institutions head into the 2026 reporting cycle, research offices are still fielding basic questions from principal investigators about which of the three compliance routes applies to a given grant, and grant management systems still struggle to capture licensing metadata in a form that survives an audit. This piece sets out a practical checklist for research administrators responsible for demonstrating Plan S compliance across UKRI, Wellcome, ERC and NWO-funded outputs.

    The underlying policy has not changed dramatically since its 2018 launch, but its operational surface has. Publishers have consolidated transformative agreements, rights-retention strategies have become the default route for many UK institutions, and funders have tightened monitoring of self-archived manuscripts in approved repositories. The compliance burden has therefore shifted from “which journal did they publish in” to “can we produce the metadata trail to prove the route taken.”

    The Three Plan S Compliance Routes

    cOAlition S — the international consortium of research funders, including UKRI, Wellcome, the European Commission (via Horizon Europe), and several national funders — defines three routes by which a funded output can satisfy Plan S requirements. Every research administrator building a compliance workflow needs a clear internal definition of each.

    • Gold open access route: publication in a fully open access journal or platform, or an open access article within a hybrid journal covered by a transformative agreement, with a compliant licence (typically CC BY) applied immediately on publication.
    • Rights retention route: the author applies a prior licence — usually CC BY — to the accepted manuscript at submission, informing the publisher that a funder mandate requires this. This route has become the default mechanism recommended by UKRI and several other cOAlition S members precisely because it does not depend on publisher cooperation or subscription negotiations.
    • Approved repository route: the author deposits the accepted manuscript (or, where permitted, the version of record) in an institutional or subject repository that meets cOAlition S technical requirements, with no embargo beyond what the funder allows — in most current cOAlition S policies, this means immediate deposit with a compliant licence, not a delayed embargo.

    The critical administrative point is that these routes are not mutually exclusive fallbacks chosen in sequence; they are three parallel doors, and the appropriate one often depends on the venue the researcher has already selected, the transformative agreement coverage at the institution, and the specific funder’s guidance on acceptable embargo periods.

    Funder-Specific Variations Research Offices Must Track

    Although cOAlition S coordinates a shared framework, individual funders retain discretion over enforcement mechanisms, grace periods and sanctions — and these variations are where most non-compliance actually occurs.

    UKRI

    UKRI’s open access policy, which applies to research articles submitted for publication from awards made after its 2022 implementation date, requires immediate open access with a CC BY licence for journal articles, and has extended equivalent requirements to monographs, book chapters and edited collections from 2024. UKRI has been explicit that rights retention is an acceptable and encouraged mechanism, and it does not permit publisher embargoes on journal articles funded under its policy.

    Wellcome

    Wellcome operates one of the more prescriptive policies within cOAlition S: it requires immediate open access with a CC BY licence (or CC BY-ND in limited circumstances) for all research articles, and — notably — it will not pay article processing charges to journals that do not offer a compliant open access route, pushing many Wellcome-funded authors towards rights retention or fully open venues.

    ERC (Horizon Europe)

    The European Research Council, funding under Horizon Europe, requires immediate open access to peer-reviewed publications with no embargo, and mandates deposit of the underlying data in line with FAIR principles where feasible. ERC compliance monitoring is integrated into Horizon Europe’s broader open science requirements, meaning research offices supporting ERC grant holders must track both publication and data-deposit obligations together, not as separate workstreams.

    NWO

    The Dutch Research Council (NWO) aligns closely with the original Plan S text and has been an active proponent of rights retention across Dutch universities, working alongside national consortia to secure publisher acknowledgement of the strategy. Administrators supporting NWO-funded researchers should note that Dutch institutional agreements often pre-negotiate compliance at the consortium level, reducing the burden on individual grant holders but increasing the importance of confirming which agreements are actually in force for a given journal at the time of submission.

    Recording Compliance in Grant Management Systems

    The most persistent operational gap is not policy interpretation but data capture. Many grant management systems were built around financial and reporting milestones, not licensing metadata, and research offices frequently discover compliance gaps only at final report stage — too late to correct.

    • Capture the compliance route at submission, not at publication. Ask researchers to record their intended route (gold, rights retention, or repository) when they submit a manuscript, not retrospectively when compiling a funder report.
    • Store the rights-retention statement text used. If a researcher applies a rights-retention statement to a submitted manuscript, the exact wording and date should be logged against the grant record, since this is the evidence a funder will request.
    • Link repository deposit records to grant identifiers. Persistent identifiers — ORCID for the researcher, DOI for the output, and a funder or grant reference — should be cross-referenced in both the institutional repository and the grant management system so a single query can reconstruct the compliance trail.
    • Flag embargo periods against funder-specific rules. A generic “12-month embargo” field is not sufficient when UKRI and ERC require zero embargo on journal articles while other, non-cOAlition S funders may still permit longer delays; systems should record the applicable policy alongside the embargo value.
    • Audit transformative agreement coverage separately from author self-report. Publisher-reported compliance data under transformative agreements does not always match what the researcher believes was agreed; reconciling the two periodically catches errors before final reporting.

    What This Means for Research Administrators

    The practical effect of seven years of Plan S implementation guidance is that compliance has become a metadata management problem as much as a policy one. Research offices that treat rights retention, repository deposit and licensing decisions as data to be captured at the point of submission — rather than reconstructed at reporting time — spend far less effort on remediation. Given that cOAlition S funders increasingly cross-reference ORCID records, repository metadata and publisher reporting, discrepancies are becoming easier to detect, which raises the cost of poor record-keeping.

    Institutions should also expect continued divergence rather than convergence among individual funder policies. UKRI’s extension of open access requirements to long-form outputs, Wellcome’s refusal to fund non-compliant APCs, and the ERC’s integration of publication and data mandates all point towards funder-specific detail becoming more important, not less, even as the shared cOAlition S framework remains stable.

    Looking Ahead

    As the Plan S open access framework matures alongside broader European and UK open science policy, research administrators should treat compliance recording as core grants infrastructure rather than a compliance afterthought. Institutions that build ORCID-linked, funder-policy-aware metadata capture into their grant management systems now will be better placed to respond as funders sharpen enforcement and as monitoring increasingly relies on machine-readable evidence rather than self-declaration.