PRAJYOT MAINKAR
On the evening of February 19, 2026, at New Delhi’s Bharat Mandapam, Prime Minister Narendra Modi stood on stage holding hands with some of the most powerful people in AI. Sam Altman of OpenAI, Dario Amodei of Anthropic, Sundar Pichai of Google and many others. It was a spectacle on the stage, but it did not quite land that way.
Altman and Amodei, whose companies had run dueling advertisements at each other just weeks earlier, noticeably failed to join the hand-holding circle as ordered. Altman later said he was “confused.” The image went viral immediately, not as the symbol of global AI unity that PM’s team had envisioned, but an unintentional metaphor for big ambitions, shaky planning, and two of the world’s most powerful AI labs refusing to cooperate even for a photo opportunity. This was the India AI Impact Summit 2026.
For five days at Bharat Mandapam, India hosted what was billed as one of the largest assemblies on AI in history, with more than 250K registered visitors from over 88 nations. Investment pledges exceeding $200 billion were announced. There was also a New Delhi Declaration adopted. India broke a Guinness World Record for the most pledges received for an ‘AI responsibility campaign’ in
24 hours, with 250,946 valid pledges received. This mission was a joint effort from the Ministry of Electronics and Information Technology (MeitY) and Intel India to foster a trustworthy, human-centric AI ecosystem.
The country showed up, broke a Guinness World Record for ‘AI responsibility pledges,’ and got some very expensive photos. But the key question remains “what, concretely, did any of this achieve?”
Art of the grand stage
To understand the India AI Impact Summit, you first need to understand Vibrant Gujarat.
In 2003, when Modi was Chief Minister of the prosperous western state, he launched what would become a defining feature of his political brand: the biannual investor summit. Every two years, business leaders from across India and around the world would jet into Gandhinagar, sign memoranda of understanding worth eye-watering sums, pose for photographs, and fly home. The Vibrant Gujarat Global Summit ran through edition after edition. An astonishing 8,662 MoUs worth $243 billion had been signed at a single summit in 2009. By 2011, that figure stepped to $462 billion in a single two-day event. The numbers kept growing. The fanfare
never diminished.
After becoming the Prime Minister in 2014, Modi took the template with him and scaled it nationally. The AI Impact Summit is that instinct applied to the defining technology of this era, positioned as India’s bid to claim a seat at the global AI table alongside Washington and Beijing.
What was actually announced
Numbers from February’s summit are, in fairness, not small. Reliance Industries and its telecom arm Jio pledged $110 billion over seven years to build AI and data infrastructure. Adani Group announced that it plans to develop $100 billion of renewable-powered AI data centres by 2035. The two largest Indian companies have made commitments worth $210 billion, which was widely reported and rightly so. Microsoft confirmed the numbers by stating it was on course to invest $50 billion across the Global South by 2030, building this figure on $17.5 billion already committed to India. Google announced a $30 million AI for Government challenge and a $30 million fund for AI for Science. Blackstone led a $600 million equity investment in Neysa, an Indian AI cloud startup. OpenAI signed a partnership with the Tata Group and became the first customer of the TCS HyperVault data centre unit. Anthropic opened an office in Bengaluru, confirming India as its second-largest market.
The New Delhi Declaration was adopted by 89 countries and international organisations, including the US, China, and Russia – a diplomatic achievement of genuine note. India also formally joined the Pax Silica – the US-led semiconductor supply chain alliance, which includes Japan, South Korea, the UK, and Israel. Under the IndiaAI Mission, the government has revealed plans to add 20,000 GPUs to the existing public compute pool of 38,000 in India.
One announcement that received less attention than it deserved: the government separately earmarked $1.1 billion for a state-backed venture capital fund specifically targeting Indian AI and advanced manufacturing startups – a direct bet on domestic builders rather than foreign data centre builders.
Signal, spectacle and the gap
Back in 2017, Congress leader Shaktisinh Gohil, in a press conference in New Delhi, released what he said was official Gujarat government data collected through Right to Information filings. The Gujarat government had claimed that from 2003 to 2015, Vibrant Gujarat summits had generated investment MoUs worth Rs. 84.55 lakh crore – a figure Gohil noted was roughly equivalent to India’s entire GDP at the time. Industrial approvals data from Gujarat’s own Industries Commissionerate, however, showed actual approved investment over 33 years at Rs. 9.51 lakh crore. The gap between promise and reality was not marginal. It was structural.
The Gujarat government’s own chief secretary, when asked about MoU realisation ahead of the 2022 summit, offered a notably candid assessment: roughly 70% of MoUs signed at Vibrant Gujarat summits get realised over time, he said- acknowledging the gestation periods can stretch for years. Independent analysis and opposition researchers have cited far lower figures, and even if one accepts the government’s own 70% figure, that means roughly three out of ten rupees announced at these summits evaporate.
This is not a problem unique to India. Investment summits worldwide suffer from the gap between intent and execution. The $200 billion target Minister Vaishnav announced for the next two years is ambitious, and we should look forward to seeing how this plays out, especially for a country whose total annual FDI inflow was around $50 billion in FY2025.
Style over substance?
There is a problem here, and it is not just India that has it. Many world governments have found out that saying ‘AI’ in a press release does not cost anything, and it makes people happy. This is called ‘AI washing.’ It is when people attach AI to any idea, policy or promise, even if it does not really have anything to do with AI.
India’s IndiaAI Mission, approved in March 2024 with a Rs. 10,372 crore (approx. $1.25 billion) budget over five years, is real. The 38,000 GPUs it has deployed are real. The subsidised compute rates of Rs. 65 per hour for start-ups and researchers represent a genuine policy intervention. But context matters. The U.S. is expected to see hyperscalers alone spend over $700 billion on AI infrastructure this year. The $1.25 billion five-year budget for the entire IndiaAI Mission is approximately what a single large American data centre costs to build. Even with private sector pledges factored in, India is entering an enormously capital-intensive race from a position of significant structural disadvantage.
One of the hard-hitting lines I came across was by Udith Sikand, senior emerging markets analyst at financial research firm Gavekal. When interviewed by CNBC during the summit week, he quoted “India is making splashy attempts to kickstart its belated AI push, but it is doing so primarily by offering headline-grabbing sops without addressing many of the underlying difficulties of actually doing business in India.” That is a pointed critique, and one that deserves to sit alongside
the celebration.
What India actually needs
It does not mean that the AI aspirations of India are misguided. They are, in fact, sound, but the journey from aspiration to realisation involves several uncomfortable truths.
The compute gap is real. India’s 38,000 public GPUs, soon to be 58,000, represent a meaningful national resource. But U.S. export controls currently cap GPU sales to India under a ‘Tier 2’ classification – a constraint that limits procurement of the most advanced chips. Without resolution of that geopolitical friction, India’s ability to build frontier AI infrastructure will remain constrained regardless of how many summits it hosts.
Although India is one of the top performers in AI skill penetration talent pipeline-wise, it is also struggling with a chronic brain drain situation. The engineers who drive OpenAI, Google DeepMind, and Anthropic in California are predominantly Indian. The question is not whether India has talent. The question is whether it can create the conditions to keep that talent or attract it back. Competitive salaries, research infrastructure, and an ecosystem where genuinely ambitious technical work can happen domestically are prerequisites. Summit photo opportunities are not.
And then there is the start-up funding gap that Anirudh Suri, founding partner of the India Internet Fund, identified plainly during summit week: venture capital and private equity money for Indian AI entrepreneurs has not arrived in proportions matching the rhetoric. Backing early-stage Indian AI founders is a different commitment, and one that has not yet materialised at scale.
The optimist’s case
It would be unfair not to acknowledge what these events genuinely accomplish. The New Delhi Declaration is a diplomatic outcome that took real work to achieve. Altman’s disclosure that India now accounts for over 100 million weekly active ChatGPT users – second only to the U.S. – is a fact that changes how every AI company on earth thinks about product investment and localisation. Anthropic opening a Bengaluru office matters not because of a press release, but because it creates a local presence that changes hiring patterns, research partnerships, and policy engagement. Microsoft president Brad Smith’s comment that India’s engineering talent pool means the country can be “a place where models are developed” is not empty flattery – it reflects a serious calculation. These relationships and signals are real. They were built, at least in part, at the summit. Signal, in geopolitics and in markets, sometimes precedes substance. The question is whether India uses the attention it has purchased to build the actual foundations – compute, governance, talent retention, startup capital – that would make these pledges stick.
The India AI Impact Summit achieved real things. The show was spectacular. The hard work starts now.
