To build AI for all, India must build it with all

India stands at the forefront of AI development, bolstered by Digital Public Infrastructure and a burgeoning innovation ecosystem. Yet, the nation faces a paradox: while India nurtures a substantial number of women in STEM, their presence dwindles as they advance in the AI pipeline. This disparity is evident across various stages: in STEM foundation, women constitute 43% of India's graduates; in the tech workforce, they comprise 26%; and in advanced AI roles, their representation plummets to 12%. Even in senior AI leadership positions, women hold a mere 10%.

This exodus is not merely a matter of employment; it has profound implications on the accuracy and fairness of AI systems. The data feeding AI models is drawn from society, and if society is fraught with inequalities—such as unequal nutrition, education, caregiving duties, workplace discrimination, and language barriers—so too will AI reflect these imbalances.

Consider Asha, a resident of rural Bihar applying for a micro-loan. AI credit models often lean on historical male financial behaviors, potentially undervaluing her creditworthiness. Similarly, Meera, a community health worker in Gujarat, relies on AI tools for maternal health. If these tools are trained on data that does not mirror local nutritional and health conditions, they may yield inaccurate recommendations, impacting maternal care.

Such barriers manifest early. Kavya may forgo robotics due to a lack of adequate school infrastructure, Pooja may struggle with AI education dominated by English, and Nisha, an AI engineer, often finds herself the only woman in the room, lacking influence in product design decisions. AI, when trained on incomplete or biased datasets, perpetuates existing inequalities.

However, it also holds the potential to broaden opportunities. Across rural India, women’s self-help groups have harnessed collective savings and entrepreneurship to foster robust financial ecosystems. When women assume influential roles, AI tends to become more inclusive.

India’s constitutional commitment to equality, established in 1950, lays a robust foundation for inclusive AI. The India AI Mission presents an opportunity to ensure that AI mirrors India’s diversity in languages, cultures, socio-economic conditions, and lived experiences. True AI leadership should extend beyond models, investments, and patents to encompass the participation of diverse voices—researchers, engineers, entrepreneurs, and policymakers from women and marginalized communities.

Ultimately, the success of inclusive AI hinges on whether it genuinely reflects the tapestry of India’s diversity.