For most of the open-data era, the prevailing question about research data has been ‘how do we make it more open?’ The CARE Principles for Indigenous Data Governance ask a different and prior question: open for whom, controlled by whom, and to whose benefit? Developed by the Global Indigenous Data Alliance (GIDA), CARE does not reject openness so much as insist that openness alone is an insufficient ethic when the data in question concerns Indigenous Peoples, lands, knowledge and lives. This article looks at what the principles actually say, what they ask of researchers and repositories in practice, and how they relate to the now-familiar FAIR principles. The governing definitions live in the Indigenous data care domain of the CASRAI Dictionary.
What CARE stands for
CARE is an acronym for four principles, each addressing a dimension that purely technical data principles leave unspoken:
- Collective benefit — data ecosystems should be designed so that Indigenous Peoples derive benefit from data, not merely supply it. Benefit includes inclusive development, improved governance and equitable outcomes, rather than value flowing only to external researchers or institutions.
- Authority to control — Indigenous Peoples’ rights and interests in their data must be recognised, including the authority to determine how data about them, their territories and their knowledge are collected, accessed and used. This is the principle that gives ‘data sovereignty’ its teeth.
- Responsibility — those who work with Indigenous data have a duty to do so in ways that are accountable, that support Indigenous self-determination and capability, and that are transparent about how and why data are used.
- Ethics — the rights and wellbeing of Indigenous Peoples should be the primary concern across the data life cycle, with potential harms anticipated and minimised, and benefits assessed against Indigenous values.
Sovereignty before the principles
CARE rests on the broader concept of Indigenous data sovereignty: the right of Indigenous Peoples to govern the collection, ownership and application of data about their communities, peoples, lands and resources. This is grounded in the right to self-determination and is widely connected to the UN Declaration on the Rights of Indigenous Peoples. Sovereignty is the claim; CARE is the practical articulation of what honouring that claim looks like in data systems. The distinction matters because it locates authority with peoples and nations, not merely with individual research participants — collective interests, not only individual consent, are at stake.
How CARE relates to FAIR
The FAIR principles — Findable, Accessible, Interoperable, Reusable — are deliberately about the technical and structural qualities of data. They are agnostic about power, purpose and benefit: data can be impeccably FAIR and still be extracted and used in ways a community never sanctioned. GIDA’s framing is that CARE is complementary to FAIR rather than opposed to it. FAIR governs how data move and connect; CARE governs the relationships and rights that should shape whether, and on what terms, they move at all. The two are meant to operate together — ‘be FAIR and CARE’ — a pairing examined more fully in our comparison of FAIR and CARE. In practice this means a repository can pursue interoperability and reuse while still applying access conditions, attribution requirements and governance rules that reflect a community’s authority.
Traditional Knowledge Labels in practice
One of the most concrete tools to emerge in this space is the system of Traditional Knowledge (TK) Labels and Biocultural (BC) Labels, developed through the Local Contexts initiative. Where a formal copyright or licence often does not fit Indigenous knowledge — because the knowledge is collectively held, governed by protocols, or culturally restricted — TK Labels add a layer of community-authored provenance and protocol information to a record. A label can communicate, for instance, that material is associated with a particular community, that it carries seasonal or gender-based access protocols, or that it should be attributed in a specific way. The labels do not by themselves enforce restrictions in a legal sense; their power is in carrying the community’s own guidance alongside the data so that downstream users know there are protocols to respect and whom to engage.
What this asks of researchers and repositories
Operationalising CARE is less about a checklist than about shifting when and with whom decisions are made. In practice it means engaging Indigenous communities as governance partners early — at the design stage, not after collection; negotiating terms of access, use and benefit-sharing rather than assuming open-by-default; supporting community capability so that authority to control is real rather than nominal; recording provenance and protocols, for example through TK Labels, so that context travels with the data; and building access and consent models that can express collective, conditional and revocable permissions instead of a single irreversible release.
For repositories and research information systems, the implication is that metadata and access frameworks need to accommodate governance that is relational and ongoing, not a one-time licence choice. None of this is in tension with giving contributors proper recognition — the same instincts that drive structured attribution through the CRediT taxonomy apply to acknowledging Indigenous knowledge holders — but it does require treating communities as authorities rather than sources. The vocabulary for doing this consistently is maintained in the CASRAI Dictionary, and the steady adoption of CARE alongside FAIR is gradually making ‘open’ a more honest word.