What AI summit means for India: Driving faster innovation and execution

India's ambition to ensure AI-led economic growth and social good rests on accelerated innovation and execution

There were more gun-toting security men than robots, more carbon-based life forms than “intelligent’’ silicon-based ones. A Chinese robodog paraded as a local innovation resulted in a Noida-based university getting evicted. 

 

Yet, inside Bharat Mandapam in New Delhi, away from all the noise, the India Artificial Intelligence (AI) Impact Summit 2026 was an AI showcase with a geopolitical signal. In attendance were more than a dozen heads of state, including France’s Emmanuel Macron and Brazil’s Luiz Inácio Lula da Silva, alongside some of the most influential leaders in AI — Sundar Pichai of Alphabet, Sam Altman of OpenAI and Dario Amodei of Anthropic.

 

At the close, 88 countries and international organisations endorsed the Delhi Summit’s declaration, vowing to use AI for “economic growth and social good”. Laying out India’s AI ambitions, Prime Minister Narendra Modi unveiled the MANAV framework — Moral and Ethical Systems, Accountable Governance, National Sovereignty, Accessible and Inclusive AI, and Valid and Legitimate Systems — positioning India as the moral and demographic anchor of the AI age.

 

India’s AI ambition rests largely on small language models (SMLs) with a complementary focus on large, specialised foundational models (LLMs) to tackle population scale challenges in healthcare, agriculture, sustainability, education, public services delivery and so on.

 

At the fourth edition of the global AI gatherings — the first hosted by a Global South nation – the message was unmistakable: India does not intend to be a spectator in the AI-led world.

 

The numbers underline the pivot. Union IT Minister Ashwini Vaishnaw said India expects to attract $200 billion of AI investments over the next two years; Reliance Industries Chairman Mukesh Ambani pledged ₹10 trillion toward AI infrastructure; Gautam Adani’s group signalled plans to invest $100 billion by 2035 in AI data centres and computers.

 

Google, OpenAI, Microsoft and others shared plans to tap into India’s AI ambitions and serve the wider South Asia region. Microsoft is investing $50 billion this decade to boost AI adoption in the Global South. AI startup Anthropic and Infosys will build AI agents for the telecoms, financial services, manufacturing and software development. Google is setting up America-India Connect, an international subsea gateway, as part of its “full stack” AI ecosystem in India. Three new subsea paths connecting India with Singapore, South Africa and Australia, and four strategic fibre-optic routes will also be established.

 

A decisive shift

 

India has also signed on to Pax Silica, a US-led technology alliance to secure semiconductor and AI supply chains, joining Japan, South Korea and the UK. Anku Jain, managing director of chipmaker MediaTek India, said, “India’s AI ambition is entering a decisive phase. The projection of over $200 billion investments in AI and deep tech over the next two years signals strong momentum across the ecosystem.” 

 

There was significant presence and contribution from companies outside Big Tech. French major Dassault Systèmes is embedding AI deeply into its platforms that companies use for design, simulation and manufacturing. Deepak NG, managing director-India, Dassault Systèmes, said: “For India, AI is not just another technology wave but is becoming a core infrastructure — much like power, roads or the internet.”

 

At the other end of the spectrum, home-grown startups Sarvam AI, Gnani.ai and BharatGen unveiled sovereign AI systems in an attempt to build alternatives to Big Tech-dominated models. Bengaluru-based Sarvam AI launched two large language models — a 30-billion-parameter system and a 105-billion-parameter mixture-of-experts (MoE) model — trained from scratch in India. MoE is like a panel of specialists, with queries being handled by the right specialist instead of the whole panel.

 

Bengaluru-based Gnani.ai, founded by former Texas Instruments engineers, unveiled Vachana TTS, a text-to-speech system capable of cloning human voices across 12 Indian languages using less than 10 seconds of reference audio — built for low-bandwidth, high-volume public deployments. IIT Bombay-led BharatGen introduced Param2 17B, a multilingual model optimized for Indic governance, healthcare and education use cases. It is backed by around ₹988crore funding from the IndiaAI Mission. 

 

User, builder, shaper

 

Sundar Pichai, CEO, Alphabet (parent company of Google) said, “India is in a unique position at this moment. It has a chance to play a big role in all three — as a user, as a builder and as a shaper of AI.”

 

India trails the US and China in frontier AI models, scale and compute density, but few countries can test and operationalise AI at a billion-user scale.

 

“We are not trying to replicate Silicon Valley. We are building AI that solves India-scale problems at India-scale economics,” said Ganesh Gopalan, CEO of Gnani.ai. 

 

Deployment of AI models will focus on areas like education, healthcare, diagnostics, agricultural advisory, addressing learning disabilities, climate change and public service delivery. The objective: population-scale impact of AI.

 

Anvesh Ramineni, partner, Endiya Partners pointed out that India’s AI ambitions centre around four aspects: sovereignty, global AI leadership, economic productivity and AI for social good. “However, we are still building capabilities in foundational AI research and sovereign large-scale model development.” 

 

Ajay Modi, investment director, Piper Serica VC fund, which has a portfolio of 31 startups, said: “India’s AI ambition is not to win the models arms race, but to deploy at scale,” citing use-cases like underwriting, risk, compliance and operational workflows.

 

For global industrial firms, India’s AI moment is tied to manufacturing growth. India has been late in the global manufacturing scene but has an opportunity to leapfrog it. Dassault Systèmes is embedding AI-powered “virtual companions” — Aura, Leo and Marie — into its platform, allowing Indian manufacturers, including small businesses, to simulate factories and products digitally, before physical rollout.

 

India’s semiconductor ambitions — catalysed by US-led  Pax Silica and domestic fabs — are meant to close the loop between sovereign AI models and machines.

 

While generative AI dominates headlines, some entrepreneurs are looking at robots. Gokul NA, founder of Bengaluru based robotics startup CynLr, argued that physical AI — object intelligence and general-purpose robotics (which aims to make robots intelligent, to handle multiple tasks) — may ultimately eclipse software AI in economic value. 

 

To be sure, globally, the competition is fierce.

 

The US dominates frontier model research. China commands computer scale and state coordination. Europe is crafting regulatory guardrails. India, with 800 million internet users and a multilingual, mobile-first population, offers the world’s largest living laboratory for AI deployment.

 

India trails the US and China in frontier research intensity and doesn't have LLMs like ChatGPT, Anthropic Claude or Gemini. It lacks the hyperscale compute clusters of Silicon Valley and the centralised execution of China. The semiconductor ecosystem is nascent.

 

The race is not for the largest AI model. But one that can embed intelligence into every transaction, every clinic, every factory floor and so on. Outside the Bharat Mandapam reality bites, and how! Former UK PM Rishi Sunak joked that AI can do many things, but can’t fix Delhi’s traffic problems. Inside, though, it is attempting to manufacture intelligence to solve real problems. In that contrast — between chaos and code — lies India’s AI ambition.   

Mission AI 

• New Delhi Declaration endorsed by 88 countries and global organisations
• India unveils MANAV Vision for human-centric AI governance
• Over $250 billion in AI investments pledged
• Indigenous AI models showcased
• National AI compute programme to add 20,000 GPUs
• Big Tech step in with Google America-India Connect, OpenAI joint AI research programmes with Indian institutes,  Microsoft’s $50bn investment in India’s AI ecosystem

 

Source: India AI Summit 2026. List is indicative, not exhaustive

 

The writer is a New Delhi-based independent journalist