The AI with Purpose series showcases how faculty and staff are thoughtfully using AI to enhance student learning, improve efficiencies and model critically informed use — while clarifying the university’s values, building AI literacy and strengthening U of A leadership in this rapidly evolving space.
As post-secondary institutions — including the University of Alberta — race to integrate artificial intelligence into the research landscape, a critical conversation is unfolding at the intersection of technological innovation and Indigenous self-determination. Indigenous data sovereignty — the right of Indigenous peoples to govern the collection, ownership and application of data that impacts their communities and lands — stands as an essential framework for ensuring that AI development and use do not replicate colonial patterns of extraction.
In a university setting increasingly focused on open access publishing, this means moving beyond simple data privacy to active governance, ensuring that Indigenous knowledge and data cannot be used without explicit Indigenous consent, says Dr. Chris Andersen, vice-provost and dean, College of Social Sciences and Humanities (CSSH).
“The problem with using an open data model is that when you democratize knowledge, it may seem like it's great for everyone, but from an Indigenous perspective, it may not be,” he says. “Historically, it actively harmed Indigenous communities because people used it to tell all these deficit-based stories, since all they ever collected information on were the things they thought were wrong with us.”
The exponential growth of open access (OA) — which removes paywalls and barriers to publishing — is fundamentally intertwined with the rapid evolution of AI, which uses massive amounts of data scraped from the internet for training and development. When AI tools vacuum up this data, they indiscriminately absorb Indigenous stories, language recordings, sacred knowledge and cultural and biographical information. This creates friction over whether "free to read" automatically implies "free to train." It is a complex question, and in the context of Indigenous scholarship, it is profoundly consequential.
Shape: A Strategic Plan of Impact emphasizes a One University model and Braiding Past, Present and Future: the Indigenous Strategic Plan centres self-determination, relationships and accountability. The ethical integration of AI needs to consider how to include Indigenous rights, governance and responsibility.
While AI presents distinct challenges in the context of Indigenous data sovereignty, its presence does not alter the fundamental principles of ownership and governance, he explains. Rather, AI simply acts as a new tool that must respect these established boundaries, guided by a framework that balances two essential principles: data for governance (securing good data) and the governance of data (maintaining control over how data is produced and preserved).
Because AI models rely on biased datasets that risk cultural misuse and stereotyping, robust safeguards are essential, including Indigenous-led consent reviews, bias testing, limits on secondary use and clear accountability.
Framing open access through an Indigenous lens
“There is tension between concepts like open data and open science and how data is collected in Indigenous communities,” says Dr. Matthew Wildcat, an associate professor in the Faculty of Native Studies. He leads the Relational Governance Project, an Indigenous policy concept that explores the practice of how First Nations co-govern service delivery organizations.
The path forward must include engagement with the First Nations principles of ownership, control, access and possession (OCAP) and collective benefit, authority to control, responsibility and ethics (CARE), Matthew says. While Canada's OCAP principles grant First Nations total control over how their data is managed, the global CARE principles complement this by focusing on the human rights and well-being of the people involved. Together, they protect Indigenous data from exploitation under the guiding motto: "Nothing about us, without us."
“Within a history of having Indigenous jurisdiction erased and extractive research, an emphasis on control is important,” says Matthew.
Dr. Kisha Supernant, the inaugural CSSH associate dean-Indigenous Engagement as of July 1, 2026, director of the Institute of Prairie and Indigenous Archaeology (IPIA) and a professor in the Department of Anthropology, sees this firsthand. Her work supporting Indigenous Nations in locating potential unmarked graves — research that is for and with, rather than on Indigenous communities — has brought the need for data sovereignty and Indigenous ways of knowing into urgent focus, placing it at the centre of archaeological practices.
“Indigenous-engaged research requires that we rethink how data is collected, stored and analyzed across all stages of the research process,” she explains. “These conversations should begin at the inception of a project, including the possibilities and challenges of working within colonial institutions.”
“In archaeology, there are legal and regulatory barriers to upholding data sovereignty and following OCAP principles,” adds Kisha, noting that at the IPIA, they have developed agreement templates that guarantee community sovereignty over all short- and long-term data. “We can also act as data stewards if the community does not have existing infrastructure to store the data securely, but the community always retains sovereignty.”
Embedding Indigenous data sovereignty
To address the persistent data divide that often leaves Indigenous communities at a disadvantage, building capacity within these communities is crucial. Chris thinks the U of A must invest in long-term partnerships through community-led data roles, training and funding guidelines that respect OCAP and CARE.
“It can’t just be a "free-for-all,” he says. “The goal should be to strengthen the ability for Indigenous peoples to manage and use our own data for our own priorities.”
Indigenous collaboration at the conceptual phase of research, embedding it as a mandatory practice, must be normalized, he explains. This could include re-evaluating institutional metrics to assess how researchers demonstrate relationship-building and shared budgets before securing ethics approvals and funding, as well as how this work counts toward tenure and promotion.
“From a co-governance perspective, in order for this to work, it should be seen as a default part of doing community-engaged research, or any kind of research that makes use of Indigenous data.”
He views the role of senior administrative leaders, including his own, as an opportunity to ground these principles into everyday practices, whether in the context of AI or not.
“It’s really important to have conversations — even uncomfortable conversations — in ways that don't ruin relationships,” he says. “In the era of Reconciliation, I want us all to walk away with what a sense of public responsibility actually looks like. It’s a North Star we can move toward.”