Principles over Politics? A Critical Reading of India AI Governance Framework

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

AI governance is shaped by a broad array of considerations, primary among them being the translation of ethical principles into practical policy goals. The principle-based approach focuses on mitigating the risks posed by AI in terms of fairness, transparency, accountability, privacy, explainability and trustworthiness. The role of the state is that of an ethical arbitrator, balancing competing claims of innovation and risk. The recent Indian AI governance guidelines (November, 2025) released by MeITY also talk about a need to balance innovation with societal requirements. This article, by foregrounding the trajectory of the AI governance discourse in India, argues that listing AI ethical principles takes the centre stage as a shorthand to mitigate the adverse impacts of a technology understood as a predetermined artefact and paradigmatic inevitability. The article critically analyses the political theoretical assumptions of AI governance discourse. We begin by asking questions around the implicit technological determinism and inevitability, both in policy documents and in the literature. We then analyse the ways in which AI ethical principles get integrated into the discussions on AI governance, arguing that they speak to the AI governance process more like a shorthand of larger socio-economic and cultural implications of AI and less as a frame of critical inquiry. Finally, in closely reading the guidelines document, and the discourse around it, the article argues for developing a political theory of AI governance as a meta-framework, which would potentially enable a situated and systematic analysis of the domain and a framework for understanding and analysing the co-production of society, politics and technology. While the article argues for a political theory of AI governance, it does not attempt to do so; a comprehensive political theory of AI is beyond the scope of this article.