Annual Conference 2025: International input on Indigenous research and AI

– The development of the research ethics guidelines has been central to the committees' work since their establishment 35 years ago. The question we ask today is: "Do we need specific guidelines for these areas?", said Helene Ingierd, Director of the National Research Ethics Committees (NREC), when she opened the conference. 

The first speaker, Mandy Downing, discussed the experiences of the Australian Institute of Aboriginal and Torres Strait Islander Studies (AIATSIS), where she co-chairs the Research Ethics Committee. Their role is to assess and approve research projects involving Indigenous peoples. An important principle is Indigenous leadership, and 12 out of the 18 team members are Aboriginal or Torres Strait Islander people.

– AIATSIS' aims include telling the story of Indigenous Australia and creating opportunities for people to encounter, engage and be transformed by that story as well as the facilitation and support of cultural resurgence.

Downing illustrated how Australia is home to over 280 distinct native title locations each with its own Indigenous group as Traditional Owners. At the same time, Indigenous peoples make up only three percent of the population and lack a unified representative body, such as the Sámi Parliament.

– Lack of involvement can cause harm

AIATSIS has developed a reseach Code of Ethics based on the UN Declaration on the Rights of Indigenous Peoples. These guidelines are built around four overarching principles: Indigenous self-determination, Indigenous leadership, impact and value, and sustainability and accountability.

– Overall, one must assess the impact of the research on Aboriginal and Torres Strait Islander peoples. Should there be representation in research? Lack of involvement can cause great damage.

The treatment of Indigenous people in Australia has necessitated the development of their own ethical guidelines for research, Downing argues. Historically, Indigenous communities have often been subjected to unethical research practices, including both physical harm and cultural exploitation and remain the most over researched population who receive the least benefit from research, she said.

Private financing = no prior approval

Robert Henry from the Canada Advisory Board for the Institute of Indigenous Peoples' Health spoke about Canada's history of colonization and the treatment of Indigenous Peoples in research. He highlighted how research conducted in the name of science has led to abuses with traumatic effects spanning multiple generations of Indigenous communities. Henry also shared some of the challenges they continue to face in Canada today.

– One major issue is that privately funded research often does not require ethical approval through a university, and, as a result, does not necessarily adhere to ethical research guidelines and principles that university researchers have to follow. This means that research involving Indigenous Peoples can take place without their knowledge or consent, and they have limited options to address their concerns.

A key question Henry asked was about who controls the data collected for research.

– Many indigenous communities lack the facilities needed to store data, he explained.

He emphasized the importance of ensuring that Indigenous communities have control over the data collected about them, particularly because such data is often used to shape policies that directly impact their lives.

Who should own the data?

Coppelie Cocq from Umeå University gave a presentation on the advantages and challenges of open science in relation to Sámi research. In highly digitalized countries, efforts are underway to transition research data to open science. In Sweden, the national goal is to implement this by 2026.

– This transition presents challenges for Sámi research, as data concerning Indigenous peoples should be owned by the Indigenous communities themselves. At the same time, many Sámi communities lack the necessary expertise to manage questions related to data access.

She emphasized that Sámi research involves a complex network of actors, making it essential to develop solutions collaboratively. Additionally, she highlighted that allowing more time for research would foster stronger relationships between researchers and research participants.

Contributes to increased awareness

Siv Kvernmo at UiT The Arctic University of Norway is the chair of the Expert Ethics Committee for Sami Health Research. At the Annual Conference, she talked about their experiences with the use of ethical guidelines for Sami research since they were launched in 2019. Over the past five years, the committee has processed 60 applications. Among the applications that are rejected or required to be revised, the committee often finds that the ethical guidelines are not sufficiently addressed, or that the project's partnerships with Sami participants are not described well enough.

One of the most important roles of the committee, Kvernmo explains, is as an advisor to researchers in application processes.

– The guidelines and the committee's work have helped to increase awareness of Sami health research and the challenges it presents.

In Norway, too, the ethical guidelines were created to ensure a Sami perspective on the health research that involves them. The need for separate guidelines became clear when health research projects that include Indigenous peoples began to increase over the past decade.

Historically, abuse of the Sami population has also been committed in research in Norway:

– Racism, discrimination and stereotypes have dominated research, says Kvernmo.

The ethical guidelines can help to ensure that this type of abuse is not repeated in the future, and that the perspectives of Indigenous peoples are better safeguarded.

The last speaker on the first topic of the day was Sean Denham, chair of the National Committee for Research Ethics on Human Remains (the Human Remains Committee). The background for the committee's work stretches back to 1915, when researchers dug up skulls from unmarked Sami burial sites in Neiden, Finnmark, to make skull measurements. The skulls were stored at the University of Oslo for many decades before they were returned and reburied in 2011.

The case revealed a gap in the Norwegian system: None of the existing research ethics committees had sufficient expertise on how this type of reversal should be carried out. The human remains committee was thus established to ensure a more ethical and respectful handling of human remains.

Today, access to the Sami remains is regulated by the Sami Parliament, although a large collection is still kept at UiO. One challenge the committee faces is misunderstandings among researchers, who may believe that the Sami Parliament is against research on the remains of Indigenous peoples. That is not the case, according to Denham.

– There is nothing inherently unethical about studying Sami remains – it's about how you do it, he emphasized.

Are separate guidelines needed?

Part 1 of the conference concluded with a panel discussion featuring the speakers, exploring the question: "Does research involving Indigenous peoples require its own guidelines?"

The panelists agreed that separate guidelines are necessary for research conducted for and with Indigenous peoples. Historically, research has often been carried out on Indigenous communities without their involvement or consent. In Australia, AIATSIS’ experience is that guidelines now help ensure that Indigenous perspectives are heard.

Downing pointed out that such guidelines can also reduce the need to include Indigenous representatives as experts in every assessment. At the same time, Denham emphasized that these guidelines should be developed by Indigenous peoples themselves, with support from universities.

New guidelines for AI use in Europe

After lunch, artificial intelligence was on the agenda, once again raising the question of whether separate guidelines are needed for the use of AI in research.

Daniela Melandri, Policy Officer at the European Commission, presented the Living Guidelines on the Responsible Use of Generative AI in Research, developed by the European Research Area Platform. The rapid advancement and widespread availability of large language models like ChatGPT have created uncertainty within research communities, particularly as AI is increasingly used to streamline various aspects of the research process, Melandri explained.

She noted that there is a need to raise awareness about how AI tools, such as large language models, are not specifically designed for research support. These models generate text that may appear accurate but can contain fabricated information.

– Our intention is not to prevent researchers from using generative AI, but to promote responsible use at this early stage while the tools are still new, she emphasized.

In developing the guidelines, the team has sought to position AI as a tool that can be used effectively — provided researchers are aware of its limitations. This approach aims to help build AI literacy within the research community.

According to Melandri, The European Commission's guidelines are aiming to be  inclusive, iterative, dynamic, and general, allowing them to be adapted to different institutions and disciplines. A new version of the guidelines will soon be available. Given the rapid evolution of AI technology and its use, the guidelines must remain adaptable. Their development is a process of co-creation, where stakeholders can contribute via a feedback form.

– The new version includes topics such as the environmental impact of AI and examples of potential risks, such as the possibility of data being stolen when entered into AI systems, Melandri concluded.

Is there a need for separate guidelines on the use of AI?

The presentations for the day were concluded by the committee chairs, who addressed the question of whether separate guidelines are needed for the use of AI in research in Norway.

The general consensus was that the existing research ethics guidelines in Norway can also be applied to AI, with the European Commission's guidelines serving as a useful supplement.

– ERA's guidelines make it clear that the responsibility for research methods, conclusions, and ethical assessments lies with the researcher and the institution, regardless of whether AI was used to reach the results, emphasized Heidi Østbø Haugen, chair of the National Committee for Research Ethics in the Social Sciences and the Humanities (NESH).

While the potential of AI is great across many fields, it was stressed that the research communities in Norway must also work to avoid major pitfalls, such as systematic bias and the fabrication of research results.

Hallvard Fossheim, chair of the National Committee for Research Ethics in Science and Technology (NENT), pointed out that there are also alternative ways in which we can address these challenges without creating separate guidelines.

– Providing clear guidance when we are asked for advice, ensuring that we don't set a high threshold for offering advisory opinions, and holding open meetings and events are other possible measures to address the challenges related to AI, Fossheim concluded.