McMaster University
The use of artificial intelligence in health policy for Indigenous peoples: a co-design approach with Aymara Indigenous people from Bolivia
Abstract
dc:description.abstractBackground: AI tools have the potential to transform health policy. However, governments and international organizations need research evidence to decide whether to allow or prohibit their use in health policy. On the other hand, vulnerable populations, such as Indigenous peoples, have suffered from marginalization, resulting in poor health and lack of access to basic health care. Therefore, more evidence is needed to see if AI further oppress those communities. Objective: To determine the contribution and ethical implications of using ChatGPT in the generation of ethical frameworks for health policy for Indigenous peoples. Methodology: This is an exploratory case study with two different methodologies, i.e., narrative review and the co-design approach. The methodology of narrative review was applied for Chapters 2 to 5, and the co-design approach was applied to Chapters 6 and 7. Chapters 2 and 3, documents some of the ways in which Indigenous peoples have been marginalized throughout history making their situation visible in order to promote respect for their human rights and equity in health policies. Chapter 4 explores how trust and overreliance on new applications of artificial intelligence, such as ChatGPT, can influence critical thinking, considering health policymaking as a contextual example. Chapter 5 presents some insights from Bolivia's approach to incorporating traditional Indigenous medicine into its national health system, from which other countries could learn. In Chapters 6 and 7, a co-design approach was developed with members of the Bolivian Aymara Indigenous community to generate with ChatGPT 4o a set of ethical principles and values with their indicators that could be included in health policy for Indigenous peoples. Chapter 6 evaluates the partnership and Chapter 7 provides evidence of the potential role of ChatGPT 4o in health policymaking so that governments or international organizations can make better informed decisions about encouraging, discouraging, or even prohibiting the use of AI in health policies for Indigenous peoples. Discussion and conclusions. Participatory-based methods in global health research are key to work collaboratively with those who have been silenced or sidelined by power dynamics in different contexts. The co-design approach was useful in research, but also to build human relationships and put human faces to abstract problems. Despite the limitations, drawbacks and disadvantages of some AI tools, in a study working with members of the Aymara Indigenous people in Bolivia, ChatGPT proved to be very useful to generate a set of ethical principles and values and indicators that could be part of a pool of ideas that health policymakers might apply to generate more equitable health policy. A study that helps demonstrate a proof-of-concept that could be helpful in future practice. However, the use of AI tools requires health policymakers to be trained not only in the use of AI, but also in the ethical concerns that the processes or outputs may bring. This dissertation has opened new avenues for further research, such as comparing results with those of other chatbots, obtaining feedback from relevant stakeholders, creating new GPTs that promote equitable health policies for Indigenous peoples, and exploring possible links between Indigenous traditional medicine and environmental health policies.
Degree
thesis:*- Department dc:contributor.department
- Global Health
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Rosas-Jiménez, Carlos Alberto
- Advisor dc:contributor.advisor
-
- Schwartz, Lisa
Subjects
dc:subject × 7Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Identifier URI
- https://doi.org/10.71548/242