The DOI matches its Crossref record; 72 of the 72 references it cites were found in a registry or an index; 31 of the 72 sources its claims cite were checked for support.
Designing Responsible AI with Indigenous Communities: Towards Participatory AI with ovaHimba and San People in Namibia
Open accessAbstract
This paper investigates the co-design of responsible Artificial Intelligence systems within three interconnected AI-driven projects with Namibian Indigenous communities. Three small-scale projects; Smart Hydroponics, Himbaverse, and Resilience App were explored and prototyped through a co-design process. The research aims to promote responsible AI development that enhances trust and aligns with local community aspirations, needs and goals of sustainability and integration of Indigenous Knowledge Systems (IKS) with advanced technology to empower local cultures. The research approach is based on participatory design and technology prototyping with ovaHimba and San communities in Namibia. It shows that transparency and active participation empower communities to contribute to the development of evolving knowledge bases and RAI system capabilities. Preliminary findings reveal that responsible AI design must be deeply contextual and shaped by local cultural values, knowledge systems, and community-defined objectives. The study contributes to the growing discourse on participatory AI development, demonstrating how culturally responsive AI systems can empower Indigenous communities and promote equitable technological integration.
Overview
CoChat Summary
TL;DR
This paper reports participatory co-design of responsible AI with ovaHimba and San communities in Namibia through three small-scale projects: Smart Hydroponics, Himbaverse, and a Resilience App. It argues that trust, cultural validity, and relatability must be grounded in local values, knowledge systems, and long-term community relationships rather than imported RAI principles alone.
Key findings
- Across three cases, participatory AI was found to be “necessary and possible” with rural Indigenous communities, with trust negotiated through transparency, demonstrations, and culturally meaningful metaphors; one elder said, “we will trust the system when we have seen it functioning.”
- In the Smart Hydroponics case, the San community co-designed AI assistants named /Ham-AI (“Sun AI”) and |Xaru-AI (“Food AI”); TinyML was considered for resource-constrained settings, but OpenAI’s GPT-4o mini API was used for the prototype due to time constraints.
- In Himbaverse, an AI-generated VR world produced inaccurate representations such as cement-based huts, lifeless vegetation, Western-style clothing, and women walking in non-traditional patterns; the co-designed framework-based world was praised for dynamic, contextual, culturally valid representations.
- In the Resilience App, a smart mirror using ChatGPT and DeepFace detected emotions and generated motivational content; participants valued Ju/’hoansi wise sayings, but noted text-based outputs as a limitation and requested audio narration and animated characters.
Methods
- Community-based participatory design with long-term partners: ovaHimba in Otjisa (Kunene Region) and Ju/’hoansi San in Donkerbos (Omaheke Region), conducted July–December 2024 with 19 Honours software engineering students and researchers.
- Two field trips of 3–5 days to each remote community: initial trips in August 2024 and follow-up evaluations in November 2024; each trip had teams of 5–10 members and 5–8 co-design, discussion, or feedback sessions, with approximately 10–20 participants per session.
- Data included field notes, observations, and video recordings; student groups conducted thematic analysis, and prototypes/technology probes were iteratively refined with community feedback under ethical protocols and community consent.
Limitations
- The paper states that long-term engagement is required to support AI literacies beyond this study, that the work involves small-scale prototypes, and that lack of digital infrastructure and connectivity—especially in Otjisa—required low-tech, off-grid options and remote researcher support.
- The paper also reports that text-based outputs limited accessibility and cultural engagement in the Resilience App, that female participants were more reserved in group discussions and needed safe spaces, and that some older participants perceived the system as emotional therapy, challenging the limits of the small-scale AI.
- Evident from the text: findings come from only two Namibian Indigenous communities and three small-scale projects, with prototype evaluations rather than full deployments, and no quantitative outcome measures are reported.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
23 Valid · 1 Not valid · 16 Not checked · 73 Own results and interpretations
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“IKS are an essential part of the cultural fabric of many communities and provide valuable insights on sustainability, resource management, and education [12, 21, 39].”
§ 2 Literature Review > 2.3 Indigenous Knowledge and Participatory Design in AI
What the sources say
- Supportedcites [12] [12] Fridolin Vrosansen Borolla, Henita Rahmayanti, Otib Satibi Hidayat, and Khairul Hafezad Abdullah. 2024. Indigenous Knowledge and Sustainable Resource Management-A Bibliometric Review. Fronteira: Journal of Social, Technological and Environmental Science 13, 4 (2024), 88–103. Google Scholar a [...] resource management, and education b [...] their knowledge is used in digital systems source
The Introduction says Indigenous traditions offer valuable insights into sustainable resource management and describes this knowledge as rooted in cultural heritage. It also says the expertise has been recognised as important to science education. (LLM, conf=0.90)
- Not checked — no full text of the sourcecites [21] [21] Rhina O. Ilagan. 2024. Integration of Indigenous Knowledge Systems and Practices (IKSPs) in the K to 12 Curriculum for Sama Bajaus: A Case Study. International education forum 2, 9 (2024), 82–92. Go to Citation Crossref Google Scholar source
- Supportedcites [39] [39] Louis Nyahunda. 2024. Integration of indigenous knowledge systems (IKS) into climate change mitigation and adaptation endeavours: milestones and gaps in South Africa and Zimbabwe’s climate policy frameworks. Climatic Change 177, 11 (2024), 1–16. Go to Citation Google Scholar source
The Conclusion links IKS to sustainable development, while §2.3 gives examples of soil conservation and farming practices. Section 2.2 describes IKS as a source of knowledge about climatic hazards beyond formal education, supporting the education point in a climate-related sense. The paper’s discussion is focused on climate adaptation rather than education generally. (LLM, conf=0.86)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“These findings highlight that some cultural representations are inherently process-driven and cannot be accurately conveyed as isolated moments or objects [52].”
§ 3 Co-Design Methodology and Empirical Cases > 3.2 Cultural Relevance and Validity
What the sources say
- Supportedcites [52] [52] Kaulyaalalwa Peter, Isaac Makosa, Selma Auala, Laurinda Ndjao, Donovan Maasz, Uariaike Mbinge, and Heike Winschiers-Theophilus. 2025. Co-creating a VR Narrative Experience of Constructing a Food Storage Following OvaHimba Traditional Practices. In Proceedings of the 2025 ACM International Conference on Interactive Media Experiences. ACM, USA, New York, 418–423. Digital Library Google Scholar a [...] and heritage in extended realities b [...] distinct heritage over the past decade c [...] Himbaverse d [...] conveyed as isolated moments or objects source
Section 4.2 says the OvaHimba community treated context and process as essential, in contrast to the designers’ focus on singular scenes and objects. Section 5.2 likewise describes the importance of movement and demonstrations for representing embodied practices. Together, these support the claim that isolated objects or moments are insufficient for some cultural representations. (LLM, conf=0.88)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Participatory design is central to the realisation of RAI in these contexts, which prioritises community participation throughout the AI lifecycle, ensuring that community members actively shape AI systems to reflect their needs, values, and knowledge systems [10, 15].”
What the sources say
- Supportedcites [10] [10] Abeba Birhane, William Isaac, Vinodkumar Prabhakaran, Mark Diaz, Madeleine Clare Elish, Iason Gabriel, and Shakir Mohamed. 2022. Power to the People? Opportunities and Challenges for Participatory AI. In Proceedings of the 2nd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization. Association for Computing Machinery, New York, NY, USA, 1–8. Digital Library Google Scholar a [...] to AI in local global south environments b [...] their needs, values, and knowledge systems c [...] and cultural contexts of the Global South d [...] diversity and cultural preservation e source
The Introduction calls participation a key component of the AI development and deployment lifecycle. Section 3.2 illustrates this in an Indigenous context: the Māori community asserts that technology using its data must benefit Māori people and be built by the community itself, drawing on its knowledge of the language. The Conclusion likewise treats community knowledge and expertise as essential to AI aimed at justice and prosperity. (LLM, conf=0.88)
- Partly supportedcites [15] [15] Fernando Delgado, Stephen Yang, Michael Madaio, and Qian Yang. 2023. The Participatory Turn in AI Design: Theoretical Foundations and the Current State of Practice. In Proceedings of the 3rd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization. Association for Computing Machinery, New York, NY, USA, 1–23. Digital Library Google Scholar a [...] their needs, values, and knowledge systems b [...] rather than imposing external values sourceExperimental — accuracy not yet measured
The Introduction (no page given) describes calls for affected communities to shape AI design so systems reflect their values and needs. But the paper does not establish this specifically for Indigenous contexts or their knowledge systems. Section 3.2.1 (no page given) also reports that community stakeholders participated throughout the project lifecycle in only 3 of 80 projects, rather than describing lifecycle-long participation as current practice. (LLM, conf=0.86)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Beyond aligning with high-level ethical principles, such as fairness, trustworthiness, transparency, and accountability, RAI must also adapt to the socioeconomic and cultural contexts it seeks to serve [8, 13, 20].”
What the sources say
- Partly supportedcites [8] [8] Selma Auala, Uariaike Mbinge, Laurinda Ndjao, Keith Craig Van Niekerk, Nestor Nathingo, Ranta Nehoya, Rachel Charlotte Smith, Kasper Rodil, Gereon Koch Kapuire, Donovan Maasz, et al. 2025. Creating a Responsible AI Framework for Cultural Validity: Exploring Heritage Futures with an Ovahimba Indigenous Community of Namibia. In Proceedings of the 5th Biennial African Human Computer Interaction Conference. ACM, USA, New York, 358–368. Digital Library Google Scholar a [...] and heritage in extended realities b [...] and cultural contexts it seeks to serve c [...] practices were reflected in re sourceExperimental — accuracy not yet measured
Section 4.1 says responsible AI should be adapted to different cultural contexts through local co-design and attention to social structures. Section 4.3 also identifies class among the differences that AI development should account for. The paper does not clearly address adaptation to economic conditions more broadly, so the socioeconomic part of the claim is only partly supported. (LLM, conf=0.78)
- Supportedcites [13] [13] Lu Cheng, Kush R Varshney, and Huan Liu. 2021. Socially responsible ai algorithms: Issues, purposes, and challenges. Journal of Artificial Intelligence Research 71 (2021), 1137–1181. Go to Citation Google Scholar source
In §5, “Open Problems and Challenges” (page unavailable), the paper says context is central to socially responsible AI and warns that interventions can fail when introduced into a new social context. Its discussion of context-aware algorithms also calls for evaluation across cultural or demographic groups, while §3.1 describes harms linked to income and access to internet service. Together, these support the paraphrase that responsible AI must account for the socioeconomic and cultural contexts it serves. (LLM, conf=0.86)
- Partly supportedcites [20] [20] Sabrina Goellner, Marina Tropmann-Frick, and Bostjan Brumen. 2024. Responsible Artificial Intelligence: A Structured Literature Review. arxiv:https://arXiv.org/abs/2403.06910 [cs.AI] https://arxiv.org/abs/2403.06910 Go to Citation Google Scholar sourceExperimental — accuracy not yet measured
Section 4.1 (page unavailable) defines responsible AI as going beyond technology to include laws and societal standards, supporting the need to consider social context. It does not say that RAI must adapt to the specific socioeconomic and cultural contexts it serves; that is a more specific requirement than the passage establishes. (LLM, conf=0.82)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: unclear — It is unclear whether it could be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Although participatory design has been slow in its direct engagement with AI, the focus on democratic technology design and computational empowerment at the community and policy levels has great potential to develop sustainable and decolonial approaches to AI in local global south environments [10, 16, 63].”
What the sources say
- Supportedcites [10] [10] Abeba Birhane, William Isaac, Vinodkumar Prabhakaran, Mark Diaz, Madeleine Clare Elish, Iason Gabriel, and Shakir Mohamed. 2022. Power to the People? Opportunities and Challenges for Participatory AI. In Proceedings of the 2nd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization. Association for Computing Machinery, New York, NY, USA, 1–8. Digital Library Google Scholar a [...] to AI in local global south environments b [...] their needs, values, and knowledge systems c [...] and cultural contexts of the Global South d [...] diversity and cultural preservation e source
Section 2.3, “The Emergence of Participatory AI,” says participatory approaches emerged in AI relatively recently and that early AI research missed earlier waves of participatory research. This supports the attributed claim that participatory design was slow to engage directly with AI. (LLM, conf=0.92)
- Not checked — no full text of the sourcecites [16] [16] Chris Frauenberger, Rachel Charlotte Smith, Daria Loi, Laura Forlano, and Ole Sejer Iversen. 2024. Emerging Technologies and Alternative Futures. Routledge, London, UK, 139–164. Crossref Google Scholar a [...] to AI in local global south environments b [...] and cultural contexts of the Global South c [...] the development and design process d [...] entangled in our everyday lives source
- Could not verifycites [63] [63] Rachel Charlotte Smith, Marie-Monique Schaper, Mariana Aki Tamashiro, Maarten Van Mechelen, Marianne Graves Petersen, and Ole Sejer Iversen. 2023. A research agenda for computational empowerment for emerging technology education. International Journal of Child-Computer Interaction 38 (2023), 100616. Google Scholar a [...] to AI in local global south environments b [...] Smith et al. sourceExperimental — accuracy not yet measured
Section 4.1 discusses research gaps in participatory approaches to emerging-technology education, and Section 4.3 describes limited access to tools for exploring AI and ML. Neither passage says participatory design as a field has been slow to engage directly with AI; searches for that assertion and close paraphrases found no such passage. (LLM, conf=0.72)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
36 more claims · 2 with their cited sources checked
Create a free account to see all 113 claims and to verify the 3 references still open.
- It may state two separate claims as one. 50
- Probably not a substantial claim. 32
- The quote only partly states this claim. 1
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 3 Co-Design Methodology and Empirical Cases > 3.1 Algorithmic Automation and Trust in AI
“To automate the chatbots’ functionalities, a lightweight large language model, Tiny Machine Learning (TinyML) [27], was integrated, ensuring responsive and efficient interactions.”
§ 3 Co-Design Methodology and Empirical Cases > 3.1 Algorithmic Automation and Trust in AI
“Although TinyML models were considered ideal for resource-constrained environments and would be used for implementation, the OpenAI’s GPT-40 mini model API was used for the prototype due to time constraints”
Citations
Cites
Algorithms as culture: Some tactics for the ethnography of algorithmic systems
Seaver, N. (2017) · Big Data & Society · doi:10.1177/2053951717738104 · Cited by 1,010 · cited 1×
“…algorithm studies to promote human-computer interaction toward understandings of the co-existence of machine and human over time, these efforts often remain fragmented, addressing algorithmic interactions without sufficient attention to broader cultural and socioecological contexts [18, 58, 59].”
Abdul, A., Vermeulen, J., Wang, D., Lim, B. Y., & Kankanhalli, M. (2018) · Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3173574.3174156 · Cited by 743
Trust in AI: progress, challenges, and future directions
Afroogh, S., Akbari, A., Malone, E., Kargar, M., & Alambeigi, H. (2024) · Humanities and Social Sciences Communications · doi:10.1057/s41599-024-04044-8 · Cited by 507
Akata, Z., Balliet, D., de Rijke, M., Dignum, F., Dignum, V., Eiben, G., Fokkens, A., Grossi, D., Hindriks, K., Hoos, H., et al. (2020) · Computer · doi:10.1109/mc.2020.2996587 · Cited by 415
Participatory Design that Matters—Facing the Big Issues
Bødker, S., & Kyng, M. (2018) · ACM Transactions on Computer-Human Interaction · doi:10.1145/3152421 · Cited by 361
Power to the People? Opportunities and Challenges for Participatory AI
Birhane, A., Isaac, W., Prabhakaran, V., Diaz, M., Elish, M. C., Gabriel, I., & Mohamed, S. (2022) · Equity and Access in Algorithms Mechanisms and Optimization · doi:10.1145/3551624.3555290 · Cited by 330
Socially Responsible AI Algorithms: Issues, Purposes, and Challenges
Cheng, L., Varshney, K. R., & Liu, H. (2021) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.12814 · Cited by 261 · cited 1×
“Beyond aligning with high-level ethical principles, such as fairness, trust-worthiness, transparency, and accountability, RAI must also adapt to the socioeconomic and cultural contexts it seeks to serve [8, 13, 20].”
The Participatory Turn in AI Design: Theoretical Foundations and the Current State of Practice
Delgado, F., Yang, S., Madaio, M., & Yang, Q. (2023) · Equity and Access in Algorithms, Mechanisms, and Optimization · doi:10.1145/3617694.3623261 · Cited by 256
Socially Responsible AI Algorithms: Issues, Purposes, and ChallengesPreprint
Cheng, L., Varshney, K. R., & Liu, H. (2021) · Journal of Artificial Intelligence Research · doi:10.48550/arxiv.2101.02032 · Cited by 249
Can artificial intelligence be decolonized?
Adams, R. (2021) · Interdisciplinary Science Reviews · doi:10.1080/03080188.2020.1840225 · Cited by 194 · cited 1×
“Decolonial approaches to AI design advocate for technologies that are co-created with African communities to address specific challenges, such as digital exclusion and resource constraints [2, 23, 65, 66, 77].”
Research works printed but not linked: 4
- Rachel Adams, Fola Adeleke, Ana Florido, Larissa Galdino de Magalhães Santos, Nicolás Grossman, Leah Junck, and Kelly Stone. 2024. Global Index on Responsible AI 2024 (1st ed.). Global Center on AI Governance, South Africa.
- Jane D Gallamaso. 2024. INDIGENOUS CULTURE IN THE AGE OF ARTIFICIAL INTELLIGENCE: CHALLENGES AND MORAL RESPONSIBILITY. Ignatian International Journal for Multidisciplinary Research 2 7 (2024) 824–837.
- Genner Llanes-Ortiz et al. 2023. Digital Initiatives for Indigenous Languages. UNESCO Publishing, Paris, France.
- Namibia. Office of the President and Namibia. Vision 2030 Project Office. 2004. Namibia Vision 2030: Policy Framework for Long-term National Development (Main Document). Vol. 2. Office of the President, Windhoek, Namibia.
Metadata
- Authors
- Indongo, N. N., Smith, R. C., Winschiers-Theophilus, H., Kandjengo, S., & Mbinge, U.
- Affiliations
- Namibia University of Science and Technology (NA)
- Aarhus University (DK)
- Department of Software Engineering, Namibia University of Science and Technology, Windhoek, Namibia
- Donkerbos Community, Donkerbos Community, Donkerbos, Namibia
- Otjisa Community, Otjisa Community, Otjisa, Namibia
- ORCID
- Venue
- Proceedings of the 19th Participatory Design Conference 2026, Vol. 1: Full Papers
- Publisher
- ACM
- Published
- 2026-06-14
- Pages
- 446-456
- OpenAlex
- W7162321549
- Cited by
- 1
- References
- 52
- Citations by year
- 2026: 1
- FWCI
- 1.36
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Locations
- Participatory Design Conference (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Information Systems › ICT in Developing Communities
- Social Sciences › Business, Management and Accounting › Management of Technology and Innovation › Innovative Approaches in Technology and Social Development
- Keywords
- responsible AI, participatory AI, indigenous knowledge, Ovahimba, San people, co-design, participatory design, indigenous peoples, cultural responsiveness, trust, hydroponics, Namibia
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Designing Responsible AI with Indigenous Communities: Towards Participatory AI with ovaHimba and San People in Namibia. https://app.cochat.ai/papers/10.1145/3796624.3796667
- The paper
- Indongo, N. N., Smith, R. C., Winschiers-Theophilus, H., Kandjengo, S., & Mbinge, U. (2026). Designing Responsible AI with Indigenous Communities: Towards Participatory AI with ovaHimba and San People in Namibia. Proceedings of the 19th Participatory Design Conference 2026, Vol. 1: Full Papers. https://doi.org/10.1145/3796624.3796667
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.