Skip to main content
SearchLoginLogin or Signup

Reimagining AI

The talent and innovation of the global South will contribute to the world's technological development, and in particular to that of artificial intelligence.

Published onSep 13, 2021
Reimagining AI

This chapter argues that the talent and innovation of the global South will contribute to the world's technological development, and in particular to that of artificial intelligence. Historically excluded people should be enabled to participate in the governance, design and production of technologies that respond to their needs.

According to Srinivasan (2019),1 the way people in the global South use and experience digital technologies could help bring a different understanding to tech innovation and its applications in the real world and the ways in which they are built for and by users. Yet, much of the Global South remains on the fringes of the knowledge economy. Often, people in these countries remain on the outside of innovation and design decision making that remains consolidated in limited learning. 

The capacity of the global South to produce knowledge and innovate offers huge potential for the future of how technology is designed and used. There are several reasons for betting on innovation from the global South: the number of people living in these regions of the planet, the cultural, ethnic and epistemic diversity and the complexity of the problems that need to be addressed. Furthermore, in order to advance innovation in low-resource settings, innovation from the South may also offer unique insight into how to design low resource computing and novel innovations to address the global environmental crisis. Despite the lack of infrastructure, institutional neglect and the enormous inequalities that exist in the counties of the global South, this chapter will highlight some examples of how people put in place mechanisms to move forward with technology production and deployment creatively and collectively, such as Gambiara in Brazil and Jugaad in India. Even in the cases of greatest adversity, the capacity for agency and resilience is evident. 

It is important to reverse the global asymmetries of power and change the paradigms of knowledge and technology production and the use of artificial intelligence for what is coming to be known as “the quantification of life” through data collection and observation. Public policies are also key to enable more people to actively participate in processes of social transformation brought on by the development and use of new technologies, and to ensure that those technologies respond to the needs of communities, rather than worsening structural violence and the destruction of the planet.

Diversity and equity are fundamental conditions for achieving autonomy, social and epistemic justice and a dignified life. If we want more technologies developed for the common good, we need to transform the systemic logics embedded in current technological production, many of which advance exclusion and domination .

Cultures of innovation: everyday innovation from the margins

In contexts where scarcity prevails, innovation is an everyday practice; it is a culture and a way of life. The Cuban artist Ernesto Oroza2 explains the architecture of necessity and the possibilities of technological disobedience: “Objects are born or transformed out of necessity. Every object is a potential object, [...] the object as a physical diagram, a structure that makes unity of understanding needs and the decisions of how to overcome them.”3 To understand vernacular technological transformations, Rai demonstrates how in India, jugaad is a form of everyday hacking, an amateur, manual, material, creative form associated with the so-called lower castes.4 Jugaad is not simply a frugal form of innovation but an ecology that can represent a twist on the forms of the organized, the methodical and the professional innovation; jugaad is a practice and a capacity for affect and affectation. Jugaad is a critical method that dissolves the thinking/doing dualism and future/past temporalities.5 

Similar forms of innovation from the margins can be seen in the practice of the gambiarra in Brazil. The gambiarra can be understood as an approach to technological remediation6 and an alternative way of producing artifacts that expresses the inventive capacity of the peripheries.7 For Boufeur gambiarra refers to the practices of adaptation, improvisation or patching of an artifact.8 (Figure 1) It has the purpose of creating immediate and improvised solutions through specific artifacts. It implies an alternative intervention and a material reappropriation,9 and an attitude to problem solving using an "immediate inventive reasoning" determined by the momentary and unconventional circumstance.

Figure 1. "Gambiarra" (improvised solution) made in a bicycle to carry objects. Foto: Rodrigo Boufleur, São Paulo, Brazil, November 2004. Public Domain.

These cultures of innovation are united through their conditions, means and ends: having minimal resources, seeking unforeseen ways of creating, and responding immediately to the specific needs of a community. Thus, based on Boufleur, it is possible to say that both the gambiarra and the jugaad are based on the articulation of three elements: a social or economic condition, the existence of specific needs, an environmental condition, the availability of material resources, and a cultural character that promotes forms of solution not foreseen by the conventional.10 

These practices aimed at reinventing material culture can be considered as valuable strategies for any process associated with the possibility of invention and appropriation of artefacts and technologies. Learning to observe ecologies of innovation in order to reverse the conditions of inequality allows us to broaden the spectrum of participation and the possibilities of making innovation real for all. Talking about innovation in the field of artificial intelligence in unequal, marginalized contexts with low rates of automation implies exploring how cultures and ecologies of innovation manifest in people's lives and considering mechanisms to incorporate embodied, embedded and everyday experiences of the community.11

Gendered innovation

Marginalized women and girls in the global south are agents of everyday innovation, but they are absent from the circles of privilege that can participate in the governance of smart systems, in their design, or the auditability of their use. However, they do contribute as a workforce to the assembly of devices, to the processes that involve content filtering and contribute with their data collected through mining. How can we expand the platforms of participation so that ordinary women can contribute with their experience, their skills and their sensitivity to build technologies that enable their personal and community growth and dignity? 

Intersectionality, decolonization, and ecofeminism provide theoretical and methodological frameworks that allow us to reflect on the development of artificial intelligence initiatives with considerations of power asymmetries, ecology and gender taken into account. Feminist principles for data collection and use,12 feminist manifestos,13 and ecofeminists are seeking to put the values of sustainability on par with the search for technological equity,14 while techno-feminists imagine creativity through play, narratives and speculative fiction (such as the Oracle for transfeminist technologies,15 or the Digital Witchcraft Institute16) as spaces for “advanced and non-conforming digital technologies for posthuman futures.”17

These provocative visions on AI show that if we broaden the sources of innovation and participation in the complete cycle of AI and its governance, not only will women have more possibilities to improve their living conditions, but they will also contribute their unique perspective from their experience as agents of change. 

Alternative and diverse AI

Currently, the development and governance of artificial intelligence is largely dominated by the ethical frameworks, interests and institutions of the industrialized countries of the global North.18 These values, interests and institutions have led to the concentration, homogenization and integration of data, technologies and infrastructures. This concentration of resources and capabilities in the hands of a few players contributes to deepening scientific, technological and social asymmetries at the global - and local - levels. Although there are numerous initiatives that promote the development of technologies based on values-oriented to achieve the common good, they cannot necessarily prevent their use against these values, nor can they guarantee substantive changes in the current course of technological development. If these technologies are deployed in contexts other than those in which they are produced, or for populations that do not have autonomy over them, there will be tensions that reproduce inequalities, exclusion or violence. For this reason, the logic underlying current technological production needs to be made visible, and its relation to worldviews and economic interests can in turn contribute to generating opportunities for alternative models of technological development. It is a pending task for international organizations, governments, companies and the community of developers to become aware of the implications of developing technologies, with consideration for their impact on communities and the environment.

Cultural and linguistic diversity

It is well known that speech recognition and natural language processing are not available or sufficiently developed for languages other than English. Around the world, many communities do not have English or other dominant languages as their mother tongue. To overcome this deficit, the Papa Reo project19 developed an automatic transcription tool for Maori20 that uses speech to text API for the first-ever synthesised Māori voice. According to their website, this tool "provides the foundational data set for the Papa Reo project” and “a multilingual language platform grounded in indigenous knowledge and ways of thinking and powered by cutting edge data science.” This project shows how technologies can be developed to serve the needs of a specific community and provides the community the opportunity to express their identity and sense of belonging. (Figure 3)

Figure 3. Kōrero Māori

Non-Western ethics

Many non-Western cultures share the principle of relationality as a fundamental dimension of community life. In sub-Saharan Africa this is associated with the Ubuntu philosophy,21 while in Andean communities, its ayni, which manifests in community practices of reciprocity such as tequio or minga. Unlike solidarity, which is unidirectional and voluntary, reciprocity is a social commitment in which all parties have equal co-responsibility in mutual aid. These principles of relationality and reciprocity are materialized in some initiatives such as Papa Reo or the principles of indigenous data governance (CARE). They are also embodied in the closely interdependent relationship between humans, non-humans and nature. 

Case study: Papa Reo

Papa Reo is an initiative of a group of people from the Maori community. It seeks to develop a proposal from the community for the community. This anchoring in the community implies that its proposals and their development respond to the needs of that community. It is not a technological proposal adapted for their language or with values different from those of their community. Papa Reo is part of an indigenous innovation movement that has shown how to contribute from different types of cosmovision to the development of artificial intelligence, and other initiatives that promote the development of standards-based on indigenous principles or methodologies that do not involve extractive processes or relationships.

Data stewardship principles

Papa Reo developed its own software license, kaitiakitanga, which responds to the interests of its community. Unlike the principle of ownership, which implies appropriation from a private or commercial perspective, the vision of data protection and stewardship implies that data is a common good of the community and therefore must be protected:

Te Hiku Media have developed a Kaitiakitanga licence, which states that data is not owned but as cared for under the principle of kaitiakitanga and any benefit derived from data flows to the source of the data. Kaitiakitanga is a principle that expresses guardianship rather than ownership of data. Te Hiku Media are merely caretakers of the data and seek to ensure that all decisions made about the use of that data respect its mana and that of the people from whom it descends. 

Papa Reo is part of an indigenous innovation movement that has shown how to contribute from different cosmovision to the development of artificial intelligence. As well as other initiatives that promote the development of standards based on indigenous principles or methodologies that do not involve extractive processes or relationships. 

Feminist infrastructures and values

The feminist critique of technology incorporates a fundamental dimension in the debate, which is to consider the values and characteristics that foster gender equality and dismantle oppression. Feminist tech communities in Latin America for instance consider four fundamental axes for developing, and deploying feminist autonomous infrastructures:

  • Consent and intimacy

  • Situated knowledge and memory

  • Seeded connectedness

  • Autonomous decision making

This approach implies “redoing the politics of building community networks, of being distributed and decentralised nodes of power.”22


Innovation is driven by the unanticipated, by the possibility of doing things differently. Innovation in artificial intelligence requires the global (epistemic) South to be enriched by the multiplicity of forms of existence in the world. Thus, it will also open the possibility for future technological pluriverses.23 For artificial intelligence and its associated response to the needs of communities in the global South, we must analyze who, for what ends, and how intelligent systems are governed, produced and used. And we may even question whether they are necessary for those communities in the first place. One way is to start from the basic questions at the local level: What problems do we have? Where can this tool be used effectively? Is this what you need to improve your life in dignity? We should aim for communities to develop the technologies they need, provide them with autonomy, enable them to achieve social justice and reparation.

 Although there is a huge digital divide in the world, the development of intelligent systems, regardless of where they occur, will and are impacting all of us. Therefore, the more we broaden the debate and the basis for participation in the decision-making process, focusing on the needs and interests of those most affected by inequalities and exclusions, the greater our capacity to respond to our challenges to achieve a dignified life and the right to a future for all.

When reimagining AI is important to remember:

  • In contexts where scarcity is the norm, cultures and ecologies of innovation are associated with the opportunity to solve basic needs. These strategies are a way to resist inequalities and exclusion. 

  • Women and girls are agents of innovation. We need to develop platforms and conditions to center women and non-binary people's lived experiences and needs as the goal of the innovation process.

  • Non-Western ethics and epistemologies can help the transformation of governance, values and methodologies in the development and deployment of AI. Local and community experiences that embody a sense of community and relational values show the way to alternative technological developments.

  • For AI to be a tool for social good it must serve the interests and needs of marginalized populations. 


Boufleur, Rodrigo Naumann. “The Gambiarra Issue: Alternative Ways to Develop Artifacts and Its Relations to the Industrial Design,” 2006.

Cifor, M., Garcia, P., Cowan, T.L., Rault, J., Sutherland, T., Chan, A., Rode, J., Hoffmann, A.L., Salehi, N., Nakamura, L. (2019). Feminist Data Manifest-No. Retrieved from:

Cortés, N. (Coord.) Nos permitimos imaginar. México: ILSB, 2020

Cortés, N. (Coord.) Tecnoafecciones. Por una política de la corresponsabilidad. México: ILSB, 2020.

Deuze, Mark.“Participation, Remediation, Bricolage: Considering Principal Components of a Digital Culture.” The Information Society 22 (2): 63–75, 2006.

D'Ignazio, Catherine, and Lauren F. Klein. Data feminism. Cambridge, MA: MIT Press, 2020.

Escobar, Arturo. "Sentipensar con la tierra." Nuevas lecturas sobre desarrollo, territorio y diferencia. Medellín: Ediciones UNAULA, 2014.

Hine, Christine. Ethnography for the Internet: Embedded, Embodied and Everyday. London: Bloomsbury, 2016.

Ibrahim, Khairunnisa, Samuel Khodursky, and Taha Yasseri. “Gender Imbalance and Spatiotemporal Patterns of Contributions to Citizen Science Projects: The Case of Zooniverse,” 2021.

Jobin, Anna, lenca, Marcelo, & Effy Vayena. (2019). “The global landscape of AI ethics guidelines.” Nature Machine Intelligence, 1(9), 389-399.

‘Kaituhi - Automatic Te Reo Māori Transcriptions’.

Kee, Jac S. Feminist autonomous infrastructure: Technomagical fires to warm your hearts. GenderIT, 2017.

Mhlambi, Sabelo. "From rationality to relationality: ubuntu as an ethical and human rights framework for artificial intelligence governance." Carr Center for Human Rights Policy Discussion Paper Series 9, Harvard University, 2020.

Oroza, Ernesto,

‘Papa Reo’. Accessed 12 May 2021.

Rai, Amit S. Jugaad Time: The Pragmatics of Everyday Hacking in India. Durham, NC: Duke University Press, 2019.

Rosas, Ricardo. “The Gambiarra : Considerations on a Recombinatory Technology.” Essay. In Digital Media and Democracy Tactics in Hard Times, 342–42. Cambridge, MA: MIT Press, 2008.

Srinivasan, Ramesh. “Opinion: The Global South Is Redefining Tech Innovation.” Wired. Conde Nast, July 11, 2019.

The Digital Witchcraft Institute,

Varon, Joana, Costanza-Chock, Sasha and Clara Juliano. The Oracle for Transfeminist Technologies, 2018.

Weiss, Joshua Z. “Belief in the Potential Object: Ernesto Oroza on Technological Disobedience in Cuba.” Walker. March 7, 2018.

Zaragoza, Liliana & Akhmatova, Anna. “Manifiesto por algoritmias hackfeministas.” GenderIT. October 15, 2018.

No comments here
Why not start the discussion?