AI in the Global South: Mapping Sites and Conceptual Vocabularies
(since Dec. 2020)
The possibilities of leveraging Big Data and AI-based interventions are often poised to flow as innovations emerging from the Global North to the rest of the world. Such flows tend to position the Global North as the active center and the Global South as the passive periphery of these innovations. Problematizing these flows, I am working on a new project at Data & Society (D&S) to map the sites and conceptual vocabularies of digital IDs and AI in the Global South. The conversations on data-driven technologies in the Global South operate on a wide spectrum between optimism of leapfrogging and digital transformation of societies on one end and the pessimism of human suffering caused by new forms of data capitalism and colonialism on the other. Set against this backdrop, this project will map this spectrum, based on an ongoing literature survey and interviews with experts, policymakers, practitioners, and activists. It will identify and articulate patterns in these conversations, which cut across various geographical contexts and industry sectors. Furthermore, these patterns will serve as initial resources towards building a common vocabulary of tensions that arise when building, appropriating, and governing data-driven technologies in different parts of the world.
Project Details: Mapping AI in the Global South on Points, D&S Blog.
Mailing List: Join the community of researchers working on AI in the Global South issues
Algorithmic Impact Assessments
(since Oct. 2020)
I am a part of the AIGI team at Data & Society (D&S) working on a research project to map the challenges of constructing algorithmic impact assessments (AIAs) by analyzing impact assessments in other domains—from finance and environment to human rights and privacy. Impact assessment is a promising model of algorithmic governance because it bundles an account of potential and actual harms with a means for identifying who is responsible for their remedy. But doing so successfully with algorithmic systems will require careful attention to how social and political power creates a preference for narrow technical metrics and can fail to adequately measure actual harms experienced by people, communities, and society. We use existing impact assessment processes to showcase how “impacts” are evaluative constructs that enable institutions to act; why it is necessary to always attend to how impacts get constructed; and occasions when the impacts measured do not capture the actual harms experienced by people.
Elizabeth Anne Watkins, Emanuel Moss, Jacob Metcalf, Ranjit Singh, and Madeleine Clare Elish, 'Governing Algorithmic Systems with Impact Assessments: Six Observations', in Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society, AIES’21 (Virtual Event, USA: ACM, 19–21 May 2021). DOI: 10.1145/3461702.3462580
Jacob Metcalf, Emanuel Moss, Elizabeth Anne Watkins, Ranjit Singh, and Madeleine Clare Elish, 'Algorithmic Impact Assessments and Accountability: The Co-Construction of Impacts', in Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, FAccT '21 (Virtual Event, Canada: ACM, 2021), pp. 735-746. DOI: 10.1145/3442188.3445935.
Restoring Credit: How people understand and interact with credit scoring systems
(since Sep. 2017)
This project is a longitudinal qualitative study of the efforts of low-income individuals to improve their creditworthiness within the lending industry in the United States. It traces credit repair journeys of low-income individuals through qualitative interviews and monthly diary entries over a period of one year of select participants in Upstate New York to understand the implications of credit scoring systems for social and economic inequality. How do ordinary consumers make sense of credit scoring systems that appear to be inscrutable? By capturing credit repair journeys, we hope to understand (1) the key concerns that trigger self-awareness of participating in an algorithmic system, (2) the strategies and tactics used to act upon these concerns under conditions of restricted knowledge and access to the system, and (3) the practices of navigating the ever-shifting line between legitimate participation and illegitimate manipulation of the system. This project is a part of a broader collaboration on conceptualizing and organizing for a Digital Due Process Clinic, based on the model of clinical legal education, to study and support data subjects in their everyday struggles in securing fair representation in algorithmic systems.
Grant: Small Grants Award, Institute for the Social Sciences, Cornell University.
Researchers: Malte Ziewitz (Principal Investigator) and Ranjit Singh (Co-Principal Investigator)
Project Site: Restoring Credit, Cornell University
Doctoral Dissertation Research: Following infrastructural changes in the story of India's development through Aadhaar
(Aug. 2012 - Aug. 2020)
How does a citizen become a data subject? This dissertation examines on-the-ground problems and practices in building and appropriation of Aadhaar (translation: Foundation), the biometrics-based national identification infrastructure of India. It advances public understanding of the affordances and limits of biometrics-based data infrastructures in practically achieving inclusive development and reshaping the nature of Indian citizenship. Deploying a mix of interview-based multi-sited ethnographic research and documentary analysis, I examine how various Indian bureaucracies—especially, the Public Distribution System that establishes access to food rights for low-income households—are using Aadhaar to distribute welfare. Aadhaar is imbricated with existing practices of identifying and authenticating each eligible citizen’s claim to government services. I show how making these claims becomes a matter of making as much of the Indian population as possible visible through Aadhaar.
Tracing the sociotechnical, legal, and administrative development of Aadhaar, this dissertation captures the artful blending of the entrepreneurial culture of IT start-ups with the bureaucratic culture of the Indian government. In emerging regimes of data-driven governance, the work of resolving citizens through data is simultaneously a social and a moral problem: social, because making up and interpreting a population as data requires so much work, organization, and discipline; moral, because using data records to represent citizens inevitably involves responding to demands of fairness, accountability, and social justice.
Grant: Doctoral Dissertation Improvement Grant, STS Program, Division of Social and Economic Sciences, National Science Foundation (NSF)
Ranjit Singh, 'Study the Imbrication: A Methodological Maxim to Follow the Multiple Lives of Data', in Sandeep Mertia (ed.), Lives of Data: Essays on Computational Culture in India (Amsterdam: Institute of Network Cultures, 2020), pp. 51-59.
Ranjit Singh, '"The Living Dead": Orphaning in Aadhaar-enabled Distribution of Welfare Pensions in Rajasthan', in PUBLIC Journal: Art Culture Ideas, Vol. 20, no. 60 (2020), pp. 92-104. DOI: 10.1386/public_00008_7
Ranjit Singh, 'Give Me a Database and I Will Raise the Nation-State', in South Asia: Journal of South Asian Studies, Vol. 42, no. 3 (2019), pp. 501-518. DOI: 10.1080/00856401.2019.1602810
Ranjit Singh and Steven J. Jackson, 'From Margins to Seams: Imbrication, Inclusion, and Torque in the Aadhaar Identification Project', in Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (Denver: ACM, 2017), pp. 4776-4824. DOI: 10.1145/3025453.3025910
Life of a Tuple: The Assam NRC as an Infrastructure of Reform in Citizen Identification
(Nov. 2016 - Sep. 2020)
I am part of a research team, which is following bureaucratic trails of documents used in updating the National Register of Citizens (NRC) to differentiate between citizens and illegal immigrants in Assam, a state in the north-east of India. This update has required a series of interrelated technical, legal, bureaucratic, and policy fixes to streamline the collection and verification of document-based evidence in building a genealogical database of citizens in Assam. The people who are left out will eventually become immigrants for the Indian state. Its conceptualization and implementation have been rife with controversy with the involvement of diverse stakeholders such as the state bureaucracy, the Supreme Court of India, civil society groups, information technology companies, and communities of people facing different challenges in enlisting as citizens. The project involves producing a documentary film on who is an Indian citizen and the profound consequences for citizens marginalized in the processes of using documents and data records to map family relations and claim Indian citizenship. Discussions on migration have focused on contests over legal identity of immigrants, making bureaucratic sense of the other. This project contributes to these discussions with ethnographic vignettes of a state-sponsored citizen registration project, which aims to make bureaucratic sense of self (citizenry) in order to differentiate it from the other. It captures how the NRC manifests borders that are not geographical but are experienced as anxieties over new forms of alienation from the Indian state.
Grant: Research Grant Programme, Azim Premji University, Bengaluru, India.
Project Site: Life of a Tuple—Researchers at Work (RAW), Center for Internet and Society (CIS), New Delhi and Bengaluru, India
Back to the Future: Situating 'Technology' in 'Science, and Technology Studies'
(May 2013 - June 2014)
This project explored the historical context within which the discussion paper on Social Construction of Technology (SCoT) emerged and the professional and discursive efforts of STS practitioners to sustain the eventual shift in focus of STS as an academic discipline from socio-cultural explanations of science to those of science and technology. Taking the SCoT paper as an exemplar of this turn to technology, this project traced the drafting, publication, and reception of the SCOT approach between 1982 and 1987 to provide insights into the development of STS as an academic discipline and extend the notion of boundary-work to professions beyond the sciences. Making a claim for expanding disciplinary boundaries is not a singular event in time. Boundary-work is a continuous practical accomplishment of professional practitioners who use accountably rational criteria of intelligibility to determine the boundaries of their professional work. An expansion claim, thus, is only as good as the work done to maintain it.
Supervisor: Michael Lynch
Award: The Sheila Jasanoff Prize for Academic Excellence in Science Technology Studies for the best graduate student paper within the previous three semesters (May 2015)
Invited Talk [Keynote Speaker]: Ranjit Singh, 'Back to the Future: Situating the 'T' in 'STS'', at the Workshop on Social Construction of Technology Coming of Age: New Challenges and Opportunities Ahead, (Trondheim: Norwegian University of Science and Technology, 3-5 June 2014).
Locating Publics: Co-Production of the Bt Brinjal Controversy and Publics in India
(Jan. - July 2011)
This project traced a sequence of historical events between 2005 and 2010 that led up to the National Consultations on Bt Brinjal in January and February, 2010 organized for the Ministry of Environment and Forests (MoEF). It evaluated the persistent notions of science governance in India and placed them within the historicity of the controversy. On 9th February, 2010, the then Indian Minister of State for Environment and Forests (MoSEF), Jairam Ramesh, imposed a moratorium on the agricultural production of Bt Brinjal after organizing a set of public consultations on the issue. The moratorium not only portrays the possibilities inherent within a public debate, but it also marks another significant event in the continuous evaluation of science and its impact on the developing economy of India. Right from the modern Chipko Movement of the early 1970s initiated as a protest against deforestation for industrialization to Narmada Bachao Andolan (Save Narmada Movement) since late 1980s against the construction of Narmada Dam, the public conversations on science in India are marked by distinct peaks of criticism within the generic troughs of belief that development through science is equivalent to progress of the country. In this project, I explored the the Bt Brinjal controversy as yet another critique of this belief system around science-led development in India.
Presentation: Ranjit Singh, 'Testing for the Post-normal age: Investigating Scientific Risk Assessment in Bt Brinjal Controversy', at the Science Studies Reading Group Meeting (Ithaca: Department of Science and Technology Studies, Cornell University, 15 April 2013).
Institution: Faculty of Arts and Social Sciences (FASoS), Maastricht University
Vaacha: A Tribal HealthCare Management System
(Jan. 2007 - July 2008)
This project focussed on building a healthcare information visualization system of patients who came to the health camps organized by Bhasha, an NGO working on the study, documentation, and conservation of marginal languages in the tribal belt of Gujarat. It was an investigation of the circumstances which lead to disease outbreaks in the tribal belt of Gujarat and an intervention in evaluating contexts and devising healthcare policy using data visualization as a tool. In addition to creating a reference index of support systems for patients suffering from various diseases, the project set out to experiment with techniques to represent statistical indicators on quality of life of such patients. My attempt was to visualize health as a way of life.
Supervisor: Binita Desai
Institution: Dhirubhai Ambani Institute of Information and Communication Technology (DA-IICT)
This was my final year research project towards completion of my undergraduate studies in BTech in Information and Communication Technology at DA-IICT.
Ranjit Singh and Ravi Kiran Atluri, 'Democracy and Policy Games: The New Information Panchayats', in Journal of Creative Communications, Vol. 2, no. 3 (2007), pp. 329-344.