AI for All: A Research Agenda for India
Based on the third meeting of the Technology Foresight group, this brief outlines a social science research agenda to support the development of safe and inclusive in AI.
Narratives & Social Frames
Multiple and often competing narratives on complex issues, such as the development of AI, emerge as new knowledge is refracted through social-frames and mindsets of stakeholders. Narratives tend to play an important role in shaping the collective imagination that directs technological trajectories and innovations.
Data and Inequity
AI systems reflect the biases of the data-sets they are trained on, and the subjectivities of individuals developing these systems. As AI systems are deployed across a range of activities, from hiring decisions to policing, there is a risk that AI systems can re-produce existing social bias and inequities. Moreover, these biases can be amplified as they are coded into seemly technical and neutral systems, in ways both visible and invisible, and permeating across a diversity of daily social practices.
Data governance and Privacy
The proliferation of AI applications poses complex and new challenges for data governance and privacy, from the transnational jurisdiction over data flows to the collection and use of personal data. The principle of consent for data-use is further complicated in the context of vast socio-economic inequities in India.
Advances in AI based technologies could improve the quality, accessibility and affordability of healthcare and education for under-served populations. A combination of environmental sensing, machine learning, predictive modelling and robotics could be applied for intelligently controlling river ecology and systems.
Global best practices
Countries across the world have started laying out their national strategy for AI, in order to accelerate AI research and development and identify the ethical and regulatory frameworks required.. A range of transnational challenges around data collection and use, labor and employment conditions, security, and governance also need to be addressed.
Knowledge Co-Creation and Policy Needs
AI is a cultural shift as much as it is a technical one. While technology companies are playing a dominant role in shaping technology trajectories, there is a need for practical and broadly applicable social-systems analysis that thinks through the effects of AI across society, engaging with social impacts at every stage — conception, design, deployment and regulation.