Manu Caddie is a member of the inaugural Steering Committee for the Multilateral Mechanism for the Fair and Equitable Sharing of Benefits from the Use of Digital Sequence Information on Genetic Resources, including the Cali Fund, representing Indigenous Peoples and local communities in the Pacific region.
This post is a revised version of an earlier article published on 13 August 2026
Third World Network (TWN) has raised some important questions about a recent study commissioned by the Secretariat of the Convention on Biological Diversity (CBD) on tools and models for making digital sequence information on genetic resources publicly available and accessible in a transparent and accountable manner.
The study deserves scrutiny because its findings could influence consequential decisions about DSI infrastructure, governance and benefit sharing. But that scrutiny also needs to distinguish limitations in the published analysis from shortcomings in the research itself, and questions the study can answer from those requiring further work.
TWN identifies concerns worth pursuing, particularly about participation, accountability and perhaps the evidence supporting comparative feasibility judgments. Its critique is less persuasive where it presents inadequate analysis as complete omission, discounts legitimate operational concerns, or treats an affiliation as evidence of a conflict of interest.
The study provides useful groundwork. Further review should build on that contribution while testing how far its evidence supports the policy conclusions readers might draw from it.
For transparency, I contributed to consultation associated with the study and provided comments during review of its section on Indigenous Peoples and local communities. I also briefly discussed these issues with the author while attending the Genomic Standards Consortium conference in Rotorua, Aotearoa New Zealand in early August 2026.
The research approach deserves fair treatment
The study drew on submissions from Parties and observers, around 40 interviews, an academic survey with 152 respondents, and a private-sector survey with nine respondents. Interview participants included governments, Indigenous Peoples and local communities, academics, database managers, and industry representatives.
It combined qualitative interviews, assessed through thematic analysis, with survey evidence. These are established social science methods. Interviews can be particularly useful for understanding institutional arrangements, implementation challenges and competing expectations.
The most appropriate question is probably whether how the evidence was collected, interpreted and used to support particular conclusions was appropriate and/or legitimate. Given the available time and resources, I think it was both. Was it perfect? Nope, no research ever is.
There are acknowledged limitations: Europe, North America and Oceania were overrepresented in the interviews, while Africa and Asia were underrepresented. Consultation was conducted in English. The surveys focused on academia and the private sector, with no equivalent surveys of provider-country authorities, Indigenous Peoples or local communities, although these groups participated through in-depth interviews and submissions.
These features limit what can be concluded about the distribution of views across constituencies, but they do not invalidate the evidence collected. In particular, the small private-sector survey should be interpreted cautiously.
The interviews were not recorded, and questions were tailored to participants’ expertise. Neither choice is inherently inappropriate. Readers would nevertheless benefit from a concise explanation of how responses were documented, how themes were developed, and how differences between participants informed the assessment. That does not require identifying interviewees. Confidentiality can be maintained while providing anonymised information about recruitment, participation and analysis.
Methodological transparency needs proportionate expectations
A policy report must provide enough detail to support scrutiny while remaining usable for delegations with limited reading and analytical capacity – more pages do not automatically produce a better report.
The distinction between an inadequately explained method and an inadequate method is important. Where the published account leaves questions unanswered, the first step should be to seek clarification rather than assume that the underlying work was deficient.
A short methodological supplement might help readers further understand how evidence informed the study’s feasibility judgments without making the main report substantially longer. More consistent expectations across CBD studies for reporting methods, limitations, disclosures and review would also help consultants and readers alike.
I believe further analysis would require separately commissioned pieces of work. Detailed economic appraisal, for example, needs specialist expertise, sufficiently developed proposals and access to financial information. It is reasonable to identify the need for that work and entirely possible without treating its absence as a failure of an exploratory study.
Feasibility judgments need to be interpreted within their limits
One of the study’s strongest conclusions is that the advantages of new database infrastructure are “consistently outweighed” by costs, complexity, duplication, fragmentation, interoperability problems and risks of reduced access and use.
These seem quite legitimate considerations. Database managers and researchers have practical knowledge of the systems they operate and depend upon. Their evidence shouldn’t be dismissed as institutional self-interest.
However, comparative judgments depend on what an assessment counts as a benefit, a cost or a successful outcome. Technical difficulty, financial burden, institutional willingness and policy desirability are related but distinct questions.
A proposal can be technically possible but difficult to fund. It can face institutional resistance while serving an important public purpose. An inexpensive intervention can be useful without delivering the level of accountability Parties or rights holders expect.
Further assessment should make these distinctions explicit. It should also consider the consequences of retaining existing arrangements, including gaps in provenance information, limited opportunities to question submissions, and uneven participation in governance.
Not all these consequences can readily be assigned a monetary value. They nevertheless still matter when comparing reform options, particularly for Indigenous Peoples and local communities whose interests may receive less attention in assessments centred on existing infrastructure and its users.
The study in this regard provides useful evidence about implementation challenges. Before that evidence is used to rule options in or out, readers need a clear explanation of how those challenges were weighed against the intended benefits.
Accountability needs clearer responsibilities
TWN raises a useful question about what accountability to Parties means in practice.
The study discusses geographical inequalities in database ownership and governance and identifies mechanisms through which Parties could participate. It also addresses transparency, sovereignty and legal certainty. It would be inaccurate to suggest that accountability is absent from the report.
The remaining question is how these elements fit together. Who should answer to whom, for which decisions, and through what procedures? What should happen when a Party, an Indigenous community or another rights holder raises a concern about information held in a database?
These questions extend beyond the technical design of repositories. Some responsibilities may belong to database operators; others may sit with national authorities, research institutions, funders or bodies within the multilateral mechanism.
A useful next step would be to map those responsibilities and use that map when comparing options. Otherwise, there is a risk of expecting databases to resolve every governance problem while leaving other important functions without an institution responsible for delivering them. So accountability should probably be assessed, and in many ways determined, through the operation of the wider system, with clear and realistic obligations for each participant.
Lawful submission is recognised, but implementation remains unresolved
TWN’s criticism concerning lawful submission identifies an important issue, but it seems to overstate the case where it suggests the study ignored it.
The report discusses national compliance requirements, the absence of permit information in major databases, and limitations in existing arrangements for establishing compliance with access and benefit-sharing obligations. It considers permit information, registration, metadata, licensing tools and legal contracts.
A more important consideration is probably how these tools would work together and what assurance each could provide.
Informing a submitter of legal obligations differs from checking evidence of authority to submit. A declaration may support accountability without independently establishing compliance. And analytics may reveal patterns of use while not saying anything about whether the original submission was authorised.
Further helpful work would identify what information databases can reasonably collect and retain, which institutions can assess that information, and how concerns can be referred for investigation or correction.
Databases can support lawful submission without becoming solely responsible for interpreting and enforcing every relevant legal obligation – an effective arrangement could make those limits clear while ensuring that unresolved questions have somewhere to go.
Cost estimates should remain attached to their scenarios
The study considers a full DSI database with functionality comparable to existing global infrastructure and uses an annual operating estimate of approximately US$50 million as a benchmark. It also considers smaller alternatives, including a permit database and a database focused on “valuable DSI”.
For a system intended to replicate extensive existing functionality, current infrastructure offers a relevant, although imperfect, cost comparison. Examining such a scenario is reasonable. The important limitation is that its estimated cost cannot be generalised to every possible accountability reform.
A permit registry, a service agreement with existing repositories, or an arrangement for preserving compliance and provenance information would have different requirements. Their costs would need to be assessed against their own specifications.
The available figures are best treated as indicative and specific to the scenarios assessed. Limited financial detail does not, by itself, establish in any way that an estimate is inflated. Equally, a broad benchmark is insufficient for choosing between substantially different institutional models.
More detailed comparison would require specialist economic work, clearer specifications and better access to operating costs. It should examine both new infrastructure and arrangements that build on existing repositories, including who would bear implementation and ongoing costs.
Registration deserves continued consideration
Public accessibility can coexist with registration, standard terms, provenance information and proportionate safeguards. The practical question is how to implement those measures effectively across a complex database network.
The study rates registration as a medium-difficulty option and discusses implementation challenges as information moves between primary and secondary databases – and smaller repositories with limited resources will likely face particular difficulties. These are substantive concerns. Requiring registration at one point in a network does not automatically create an effective record of subsequent access or use elsewhere.
However, medium difficulty warrants further investigation, especially where an option could make a meaningful contribution to accountability. The emphasis on easier measures in the study’s conclusions should not lead readers to disregard more demanding proposals.
A useful follow-up would examine what registration is intended to accomplish, where and at what thresholds it would be required, how it would work across connected systems, and what resources and safeguards implementation would need.
The same care applies to interoperability and researcher uptake. Poorly integrated systems can fragment information, increase curation burdens and discourage participation. Those risks should be assessed alongside the accountability benefits a proposal is intended to deliver.
TWN’s argument would be stronger if it gave these operational questions more attention and explained how its preferred arrangements would address them.
Provider interests require action across the mechanism
Provider countries, Indigenous Peoples, local communities and originating researchers have legitimate interests in how information is submitted, described, accessed and used.
These may include opportunities to correct inaccurate records, raise concerns about unauthorised submissions, preserve information about provenance and associated Traditional Knowledge, identify collaborators, and receive non-monetary benefits.
From a provider’s perspective, such functions may be central to equitable participation. An assessment focused mainly on searching, downloading and analysing DSI can miss that broader experience of the system.
Nevertheless, these interests do not all translate into responsibilities for database operators.
Some functions may fit directly within database services, such as retaining appropriate metadata or providing a process for raising concerns. Others may require connections to permit systems, research institutions or community governance arrangements. Benefit-sharing programmes and collaboration support may belong primarily elsewhere.
The task is to identify which functions databases should perform, which they should support, and which require other institutions. This would give provider interests a clearer place in the multilateral mechanism while keeping expectations of database operators realistic.
The study’s catalogue of tools offers a useful starting point. Developing the wider institutional arrangements will require work beyond an assessment of databases alone.
Affiliations do not establish a conflict of interest
TWN is entitled to ask for transparent authorship and disclosure of relevant affiliations and interests. Readers should be able to understand who prepared an influential study and what expertise and perspectives informed it.
However, the evidence presented does not establish a conflict of interest.
Membership of a scientific network does not demonstrate that an individual shares every position associated with it, or that their conclusions were improperly influenced. Researchers, advocates, officials, Indigenous representatives and database operators may all have established views and institutional relationships.
Those relationships can warrant disclosure but they don’t automatically invalidate a person’s work.
The appropriate safeguards are transparent methods, relevant disclosures, diverse participation and independent scrutiny. Unsupported assumptions about an author’s loyalties do not strengthen an assessment of the evidence.
It is also unhelpful to divide participants into those who support open science and those who support accountability. People can support both while disagreeing about particular institutional or technical arrangements. The quality of those arrangements should be the focus of debate.
Further review should build on work already done
The study received expert input on individual sections, including IIFB review of the section on Indigenous Peoples and local communities, to which I contributed comments. It would therefore be inaccurate to describe it as having received no external review.
Section-level expert input differs from independent review of the entire study, but it is a meaningful form of scrutiny.
A further review could assess the study as a whole, clarify the strength of its comparative conclusions, and identify questions requiring additional research. It could include scientific, technical, legal, economic and Indigenous governance expertise, with opportunities for participation from constituencies underrepresented in the original consultation.
If it happens, that process should be proportionate and time-bound. The author and Secretariat should have an opportunity to respond, and the process should distinguish corrections or clarifications to the existing report from additional work beyond its original scope and resourcing constraints.
The study offers a substantial account of the DSI database landscape and a useful catalogue of possible tools. TWN has identified questions that deserve further attention, particularly about accountability, participation and the evidence needed for policy choices. Its critique becomes less convincing when it treats limited analysis as omission, discounts operational concerns, or infers conflicts from affiliations.
The next stage should build on the research, recognise its acknowledged limits, and establish how databases and other institutions can together support accessible science, Indigenous rights and fair and equitable benefit sharing.
Aotearoa Indigenous Rights Trust 