The inaugural Global Dialogue on AI Governance was held in Geneva on 6-7 July and India was represented by the Minister of State for External Affairs Kirti Vardhan Singh. Mandated by UN General Assembly resolution 79/325 and rooted in the Global Digital Compact agreed at the 2024 Summit of the Future, the Dialogue was deliberately chaired by two small states - Ambassador Egriselda López of El Salvador and Ambassador Rein Tammsaar of Estonia rather than by any AI superpower.
India arrived as a state that had, five months earlier, hosted the largest AI gathering yet held anywhere - the India AI Impact Summit in New Delhi, which drew more than twenty heads of government and participants from over a hundred countries. The UN's new Dialogue, mandated by General Assembly resolution 79/325 and modelled on the non-binding Internet Governance Forum, hands every nation a seat but no gavel. Its currency is influence, not enforcement and influence is precisely the resource India has spent the past two years accumulating.
India's claim to shape global AI rules rests on something few developing economies possess: a working, at-scale demonstration that AI can be built for public purpose rather than private capture. Its Digital Public Infrastructure - Aadhaar, UPI, DigiLocker, the BHASHINI language stack already reaches hundreds of millions, and AI is now layered onto it in concrete ways, from MuleHunter.AI flagging fraudulent "mule" bank accounts to the automated translation of court judgments into regional languages. This is the template India carries to Geneva: governance framed around access and delivery, not only restriction.
Behind the pitch sits real capacity. The IndiaAI Mission, backed by over ₹10,000 crore, is expanding national compute from roughly 38,000 GPUs toward nearly 60,000, and is funding sovereign models - the government-backed BharatGen, Bengaluru's Sarvam AI, Gnani.ai's Vachana speech systems spanning a dozen Indian languages, designed to keep data and cultural representation within national borders. With one of the world's largest AI talent pools and an ecosystem projected to add several hundred billion dollars to GDP by 2030, India can speak as both consumer and producer, a rare vantage in a debate otherwise polarised between Silicon Valley and Shenzhen.
India's substantive contribution is a reframing. Where Bletchley (2023), Seoul (2024) and Paris (2025) fixated on catastrophic and existential risk, New Delhi's summit pivoted the conversation toward present-day harms, equity and deployment, organised around three "sutras" - People, Planet, Progress. In Geneva, Singh pressed the same line: close the capacity gaps of the Global South, or watch AI become a multiplier of inequality rather than an equaliser. "Machines do not quarrel over borders, ideology, or pride; humans do," he told delegates. This development-first framing is India's genuine value-add to a forum where, tellingly, governments were the only stakeholder group to rank capacity-building above safety in more than 1,500 pre-session submissions.
The geopolitics of AI is also shaping fast and here India's timing is fortunate. The United States has largely stepped back from multilateral AI diplomacy, preferring instruments it leads - the G7 Hiroshima Process, the AI Safety Summit series and, under a deregulation-minded administration, resisting binding global rules. China, conversely, champions the UN as the sole legitimately universal venue, courting the Global South with low-cost models and infrastructure. Into that bipolar gap steps India - non-aligned by instinct, credible to both camps, and unwilling to be pressed into either technological sphere as the risk of a "digital iron curtain" grows.
India is already hedging in exactly this register. On the margins of its own summit it joined the US-led Pax Silica semiconductor initiative and signed an India-US AI Opportunity Partnership, while simultaneously launching a trilateral with Italy and Kenya to co-design "sovereign AI" pathways for Africa, linked to a G7-endorsed AI Hub for Sustainable Development. That portfolio displaying one foot in Western supply chains, one in South-South cooperation is the essence of the swing-state role. India's impact in the Dialogue lies less in any single speech than in its capacity to broker between blocs that no longer talk easily to each other.
However for India to make the forum to matter, four moves follow. First, it can become the lead advocate for the Secretary-General's proposed Global Fund for AI, offering its DPI experience as the delivery model and giving capacity-building a concrete institutional home. Second, it can push for interoperable testing and evaluation standards, the practical safety tools it foregrounded in New Delhi, so that governance converges on shared instruments even where national laws diverge, as they will when the EU's AI Act reaches full application in August. Third, it must convert pledges into architecture: the summit's headline investment figures, exceeding $250 billion, are market signals, not commitments, and India's strength depends on translating them into measurable Global South deployment before the Dialogue reconvenes in New York in May 2027. Fourth, it should use the intervening months to consolidate a coalition - Brazil, the African Union, ASEAN partners - so that the next session carries a Global South negotiating bloc, not a scatter of individual voices but at the same time keeping its partnerships with France and USA growing.
The deeper test is philosophical. The world does not need another AI declaration; it needs architecture - standing institutions, funded mechanisms, enforceable red lines. India, uniquely, has the political weight, the technical stack and the market scale to help build it, plus the diplomatic latitude to bridge a fracturing order while doing so. Emmanuel Macron, speaking in New Delhi, argued that the smartest AI is not the most expensive but the one built by people for the right purpose. India's opportunity in Geneva was to make that sentiment operational and prove that a Global South convenor can move a non-binding dialogue toward binding substance.
(Subimal Bhattacharjee is a policy adviser on digital tech issues and the author of 'The Digital Decades: Thirty years of the Internet in India')
Disclaimer: These are the personal opinions of the author
