There is a reassuring story that technology often tells about itself. We are told that if data is well organized, safely stored, properly anonymized and made available through the right platform, then its benefits will naturally follow. Better information will lead to better decisions. Better decisions will lead to better outcomes. In many contexts, that can be true.
But it isn’t a given.
When the data concerns Indigenous Peoples, their territories, their cultures, their languages, their resources or their knowledge, the central issue is not only how well the data is managed. It is who has the right to govern it.
This is where Indigenous Data Sovereignty becomes so important. It is sometimes described as a technical question about data ownership, hosting or access permissions. Those things matter, but they are only part of the picture. Indigenous Data Sovereignty is about the right of Indigenous Peoples to exercise authority over data that relates to them, consistent with their rights to self-determination, governance, culture and development.
That is a much bigger idea than a setting in a database.
The data is already there
I would say that one reason this conversation is urgent is that Indigenous data is everywhere. It appears in government censuses and administrative systems, public-health records, land registries, environmental impact assessments, school reports, satellite imagery, academic research, biodiversity databases, social-media platforms and commercial supply chains. Increasingly, it is also generated through mobile phones, sensors, drones, cameras, DNA sampling and artificial intelligence systems.
The people and institutions holding that data are often not the people and communities to whom it relates. In many cases, Indigenous Peoples have been studied, classified, mapped and measured for generations without having meaningful control over how the resulting information is used. Sometimes the data has been collected in the name of protection or development. Sometimes it has been collected to make a territory easier to manage, tax, acquire, conserve or exploit.
The effects of this history are still present. A community may know that an outside research team holds detailed information about its medicinal plants, sacred sites or customary governance. It may not know where the files are, who has downloaded them, whether they are now feeding a commercial product, or how they can be corrected or withdrawn. A government may possess data that a community needs to understand a proposed mine or road, while the community itself has no meaningful access to it.
We should not be surprised that trust is often fragile.
Control does not mean isolation
There’s a common misunderstanding that I have encountered. Indigenous Data Sovereignty is not an argument that all data must be closed, or that Indigenous Peoples should be excluded from science, policy or technology. It is an argument that Indigenous Peoples must have a real say in the terms of participation. They should be able to decide when sharing makes sense, with whom, for what purpose, and under what protections.
Sometimes the appropriate decision will be to share data publicly because it strengthens a land claim, supports climate adaptation, contributes to a community-designed health intervention or helps hold a government accountable. Sometimes it will be to share information with a limited group under agreed conditions. And sometimes the correct decision will be not to share it at all.
The ability to make that decision is the point.
There is a useful contrast here with the language of open data. Openness can be valuable, especially when public institutions are withholding information that affects people’s lives. But a blanket commitment to make data findable, accessible and reusable can overlook history, power and risk. Data about a community is not simply a public good because it is valuable to someone else. Information that is open to a university, a company or a government may become inaccessible or harmful to the community that provided it.
The Global Indigenous Data Alliance’s CARE Principles offer a helpful way to think about this. CARE stands for Collective Benefit, Authority to Control, Responsibility and Ethics. The principles do not replace technical approaches to good data management; they add the questions those approaches often leave out. Who benefits? Who decides? What responsibilities remain after collection? Are rights and wellbeing treated as primary rather than as an afterthought?
Consent is not a one-time transaction
In conversations about data, people often reach for the word consent as if it resolves everything. A form was signed. A meeting was held. A box was ticked. But consent that is rushed, poorly explained, or disconnected from local decision-making structures is not meaningful consent. Nor is consent that cannot be revisited.
Free, Prior and Informed Consent (FPIC) is particularly relevant here. It is not a single signature obtained at the start of a research project or mapping exercise. It is a process. It requires freedom from pressure, information that is understandable and relevant, time before a decision is made, and a genuine possibility of saying yes, no or not yet. It also requires respect for the decision-making institutions that a people themselves recognize.
Data projects change. A dataset that begins as local environmental monitoring may later be combined with other records, used in a new study, uploaded to a repository or incorporated into a commercial tool. A photo taken for a community archive may later be used to train an image-recognition system. Consent for the first activity cannot automatically be assumed to cover every future use.
That is why governance has to be designed for the whole data lifecycle: collection, storage, analysis, sharing, reuse, archiving and deletion. It should include clear answers to basic questions.
Where is the data stored? Who can access it? Can a community withdraw material? Who decides whether a request from a researcher, a donor or a company is appropriate? What happens if the project ends or the partner organization changes?
Building institutions, not just projects
One of the most promising directions in this work is the focus on long-term Indigenous-led data capacity. Too many projects treat data governance as a service provided by an outside organization. Experts arrive, conduct training, establish a platform, publish results and leave. The community is left with an account on a system it may not control, a report it may not have requested, and expectations that cannot be sustained.
An alternative is to invest in the institutions and people who will make decisions over time. This can mean supporting Indigenous organizations to employ data specialists, develop their own protocols, establish community review processes, choose appropriate infrastructure, and negotiate with governments and research institutions from a stronger position. It can mean resourcing translation, youth participation, elders’ guidance, local language materials and the patient work of reaching agreement across a community.
This approach also improves data quality. Communities understand the places, relationships and changes that an outside team may not see. They can identify when an indicator does not make sense, when a boundary on a map is inaccurate, when a category erases an important difference, and when a result needs context. They can help define measures of wellbeing that are not limited to income or service delivery but include language, culture, land, food systems, safety and collective life.
A responsibility for everyone else
Indigenous Data Sovereignty is sometimes presented as something Indigenous organizations must solve on their own. That framing is unfair. Governments, donors, researchers, technology companies and NGOs all have responsibilities because they are often the institutions with the greatest resources and control.
For governments, this includes making public data accessible to Indigenous Peoples in forms they can use, collecting disaggregated data with appropriate safeguards, and recognizing Indigenous governance in national data systems. For researchers, it means creating agreements before collection, budgeting for community leadership and data governance, and refusing to treat a publication deadline as more important than a community’s process. For donors, it means funding the less glamorous but essential work of governance, infrastructure, staff time and long-term stewardship.
For technology companies, the questions are becoming even more pressing. Digital systems can create useful opportunities for language revitalization, environmental monitoring, communications and community archives. But they can also make extraction faster and less visible. Information placed online may be scraped, copied or used to train systems far beyond the original context. Communities should not have to become experts in every new platform simply to protect themselves from harm.
The responsibility is to design with communities from the beginning, to be transparent about uses and risks, and to respect a decision not to participate.
Whenever I hear an organization say that it wants to work with Indigenous data, I think the first question should be simple: who will still have authority when this project is over?
If the answer is a university, a ministry, or an NGO, then the project may be improving an external system without changing the underlying relationship. If the answer is an Indigenous people, community or organization, supported with the resources and authority to make decisions, then we may be moving toward something more durable.
Data can help make rights visible. It can support planning, advocacy and accountability. It can connect local realities to national and international debates. But it will only do those things justly if the people represented in the data are not reduced to its subjects.
Indigenous Data Sovereignty asks us to recognize something fundamental: data about a people is connected to that people’s right to determine its own future.
Leave a Reply