AI Summit: Hype, Hard Truths, and India’s AI Gap
AI summit showcases India’s global AI ambition, but beneath the diplomatic glow lies a stark gap in compute power, chipmaking, and homegrown AI leadership.
AI Summit Spotlight: Big Declarations, Zero Chipmaking Power
The narrative around the India AI Impact Summit 2026 and its celebrated New Delhi Declaration reads like a triumphal march of the Global South into the commanding heights of artificial intelligence. Framed as the world’s largest AI gathering and wrapped in soaring civilisational language, the AI Summit is projected as a geopolitical turning point where India emerges as a moral and institutional architect of the AI age. Yet a closer, harder look reveals a widening gap between diplomatic theatre and technological reality.
Support Independent Journalism. Public interest stories that affect ordinary citizens — especially those without power or voice — requires time, resources, and independence. Your support — even a modest contribution — allows us to uncover stories that would otherwise remain hidden. Support The Probe by contributing to projects that resonate with you (Click Here), or Become a Member of The Probe to stand with us (Click Here) |
Also Read: DPDP Act and RTI Act: Is the Balance Tilting Toward Secrecy?
While India boasts millions of AI users and one of the world’s largest digital populations, it remains strikingly thin in original AI platforms, foundational models, and globally scaled applications. This critique attempts to peel back the rhetoric to examine how much of the AI Summit was genuine capacity-building and how much could be carefully choreographed hype masking structural weakness in India’s AI ecosystem.
AI Summit: Moral Posturing, Missing Muscle
The narrative portrays the Summit almost as a civilisational awakening, claiming a “recalibration in the global geography of artificial intelligence governance” after 88 nations endorsed the New Delhi Declaration. It celebrates India as a bridge between technological superpowers and developing nations, projecting convening power as though it were technological leadership itself. The language is grand, confident, and glowing — but dangerously light on substance.
Hosting the “largest AI summit in the world” sounds impressive. Davos also hosts the world’s most influential economic conversations — yet Switzerland is not the world’s dominant economic power. Conferences signal aspiration, not capability. The narrative quietly equates diplomatic applause with technological ascendancy, a sleight of hand common in policy spectacle. The Declaration itself, proudly described as “voluntary and non-binding,” is framed as pragmatic wisdom. In reality, it is precisely why it is painless for 88 countries to endorse it. Non-binding declarations — if one may say so — are the international equivalent of LinkedIn likes — generous, costless, and quickly forgotten.
The seven thematic “chakras” of cooperation, access, trust, infrastructure, human capital, and sustainability sound spiritually harmonious, but they are conceptually recycled from every global AI forum of the last five years — the OECD AI Principles, G20 AI frameworks, UNESCO’s AI ethics recommendations, and the UK’s AI Safety Summit declarations. What is missing is anything operational: funding commitments, shared compute facilities, joint research programmes, or enforceable standards. Without money, chips, models, and deployment pipelines, chakras remain slogans. In defence, it can, however, be said that Declarations are only that — pious intentions — and nothing beyond.
Also Read: Are Hospitals Hiring Unregistered Nurses Despite Indian Nursing Council Rules?
The narrative then pivots dramatically into philosophy through the MANAV framing, presented as India’s unique human-centric AI doctrine rooted in “Sarvajan Hitaya, Sarvajan Sukhaya” and articulated by the PM. It is rhetorically elegant and politically attractive — but policy history is littered with beautiful visions that collapsed without institutional muscle. Ethical AI does not emerge from cultural metaphors; it emerges from rigorous datasets, bias audits, accountable algorithms, safety research budgets, and independent regulators. None of these were concretely anchored in the Declaration.
The same pattern follows with the grandly named “Pax Silica,” evoking semiconductors as the bedrock of AI power. There is a temptation to present it as a convergence of technology, finance, and skills, a strategic masterstroke signalling India’s infrastructure awakening.
Yet the uncomfortable reality is this: India currently manufactures almost zero advanced AI chips, relies heavily on imported GPUs, and remains years away from meaningful semiconductor fabrication at scale. Announcing convergence without controlling compute is like declaring naval supremacy without ships.
Where the narrative becomes most misleading is in its quiet avoidance of the central paradox of India’s AI story.
AI Summit Raises Big Questions About India’s AI Core
We are one of the world’s largest consumers of AI. Millions use ChatGPT, Google Gemini, Meta AI, Microsoft Copilot, recommendation algorithms, translation tools, facial recognition systems, fintech risk engines, and health diagnostics every day. India’s startups enthusiastically integrate foreign AI APIs into apps for education, agriculture, HR, lending, and customer service.
- But where are India’s foundational AI platforms?
- Where is India’s large language model competing with OpenAI, Anthropic, Google DeepMind, Meta, or Baidu?
- Where is India’s globally adopted AI cloud stack?
- Where is India’s breakthrough vision model, speech engine, or medical AI system at scale?
Despite years of policy speeches, India has produced almost no globally relevant AI core technologies. Most “AI startups” in India are wrappers around foreign models. They build interfaces, workflows, and local adaptations — valuable, yes — but not technological leadership.
Also Read: Trump’s Tariffs: What the US Supreme Court Just Changed
Even government-backed efforts like AIRAWAT (India’s public AI compute cloud) remain modest compared to US and Chinese infrastructure. India’s total AI research spending is a fraction of what single US tech giants invest annually. The Stanford AI Index consistently shows the US and China dominating model development, patents, compute power, and private investment — with India largely absent from the frontier layer.
In simple terms: India is an AI marketplace, not yet an AI engine.
The laudatory post acknowledges “modest research expenditure” and “limited high-performance compute” almost as passing footnotes — when in truth these are the entire game. AI leadership today is about three brutal realities: massive data, massive compute, and massive capital. Declarations do not train trillion-parameter models. Diplomacy does not manufacture GPUs. Philosophy does not replace cloud clusters.
What the AI Summit brilliantly achieved was narrative dominance. It repositioned India as a moral convenor, gave the Global South symbolic inclusion, and projected technological ambition without exposing technological thinness. This is soft power at its finest — but soft power should not be confused with hard capability.
There is also a deeper risk. By celebrating symbolism too early, India may lull itself into believing it is already an AI leader rather than a late but promising entrant. The danger of hype is complacency. China became an AI powerhouse not through conferences but through relentless state investment in chips, research labs, defence-linked AI programmes, massive data pipelines, and industrial deployment. The US dominates through venture capital, hyperscale compute, and world-class universities tightly linked to industry.
India, by contrast, is still debating frameworks while renting foreign compute.
The way forward is not more summits — it is brutal, expensive, and unglamorous execution.
AI experts say that we need sovereign-scale compute infrastructure measured in exaFLOPs, not pilot clouds. We need multi-billion-dollar AI research missions tied directly to universities and startups. We need semiconductor fabrication that actually ships advanced chips, not policy announcements. We need open national datasets for health, agriculture, mobility, and governance to train world-class models. We need safety and ethics bodies with real technical teeth, not cultural slogans. And we need to move from API consumption to model creation.
If we can marry our vast user base with real technological depth, we can indeed shape inclusive AI for the developing world. But until then, the Delhi Declaration remains what it currently is: a beautifully worded diplomatic milestone floating over a thin technological foundation.
The AI Summit was not meaningless. It strengthened India’s soft power, elevated Global South voices, and framed AI as a development tool rather than just a profit engine. But calling it a turning point in global AI leadership is premature.
Right now, India is the world’s fastest-growing AI consumer — not its builder.
And no amount of declarations can compute a future.
P. Sesh Kumar is a retired Indian Audit and Accounts Service (IA&AS) officer of the 1982 batch who served in senior audit functions under the Comptroller and Auditor General of India (CAG), including as Director General of Audit in the CAG’s office. He is also an author of books and commentary on public sector auditing, accountability, and governance, drawing on his extensive experience in financial oversight and institutional reform.
