Abu Dhabi, UAE, 29 Sept 2026: As global AI enterprises and governments prepare to converge for Ai Everything Abu Dhabi from 6-7 October 2026 at ADNEC Centre, the conversation has shifted decisively towards nation-scale AI infrastructure – how AI has moved beyond the remit of CIOs to national AI missions and policymakers.
Ramanan Ramanathan, Mission Governing Board Member of the IndiaAI Mission, Government of India, addresses this through a keynote on “Redefining Digital Power – India’s Agentic AI Leap for the Global South,” carrying a message aimed squarely at how the AI era’s winners and losers will actually be decided.
“The next phase of AI will not be won simply by whoever has the largest model,” Ramanan said ahead of the event. “It will be won by whoever can translate intelligence into trusted, affordable and inclusive impact at scale.”
It is a deliberate departure from the frontier-model race that dominates most AI headlines – and one Ramanan argues India is well placed to lead. “The ultimate measure should not be how many AI pilots a government has, but whether AI improves the quality, speed, accessibility and cost of public services while maintaining public trust,” he reiterated.
Lessons from deploying AI at 1.4 billion population scale
Few governments anywhere have had to design AI systems for a population as large or as linguistically diverse as India’s, and Ramanan was candid about the learnings from this scale. “An AI system that works well for thousands or millions of users cannot simply be multiplied to 1.4 billion people,” he said.
“At population scale, infrastructure, language, affordability, reliability, cybersecurity, inclusion and human oversight have to be considered together” – including designing for linguistic and cultural diversity rather than assuming an English-centric model will serve everyone.
That scale also changes how governments approach AI as it becomes more autonomous. “The moment AI moves from answering questions to taking actions – trust, auditability, and accountability and human supervision become as important as intelligence,” Ramanan said, advising that governments must move from “innovate first and regulate later” towards continuous testing, evaluation, guardrails and human accountability.
A blueprint for government and industry collaboration
Ramanan also sees the greatest opportunity in treating AI infrastructure as an ecosystem, jointly built with governments, the private sector and academia.
“Government creates the rails, standards and demand; the private sector brings speed, innovation and capital; and academia pushes the frontier and helps build talent,” he said, describing this as especially critical for sovereign AI infrastructure, where nations need resilience and strategic capability “without unnecessarily duplicating everything themselves.”
He extended that logic beyond India’s borders, framing the country’s collaboration with the UAE, and with the wider Global South, as an opportunity rather than competition. “The UAE and India both have an opportunity to be important bridges between advanced AI capability and the needs of emerging economies,” he said. “Let us build AI that is not only intelligent, but also responsible, accessible and capable of creating meaningful human and economic impact.”
Ramanan’s remarks are grounded in a national programme with major capital and infrastructure behind it. The IndiaAI Mission, launched by the Government of India with an outlay of ₹10,372 crore (US$ 1.5 billion), aims to build a comprehensive ecosystem that fosters AI innovation and industry collaboration for inclusive national development.
Under the Mission, more than 38,000 GPUs have been onboarded onto a common compute facility and made available to Indian startups and academic institutions at affordable rates.
On the show-floor: Indian AI companies lead international participation
Representing one of the largest international tech participations at the event, several Indian AI companies will present breakthrough solutions developed for financial services, enterprise software, conversational commerce and public services.
Gallabox, a business messaging and conversational AI platform with offices in India, the UAE and the US, is launching its AI Voice Agents to handle inbound and outbound phone calls in natural speech. This is in addition to its AI Chat Agents for managing the customer lifecycle across WhatsApp, Instagram and web chat. “We built a voice agent that behaves the way a well-trained call handler would, not a rigid IVR menu,” said Karthik Jagannathan, CEO and Co-founder. The company has raised US$5 million to date and says its platform has processed more than 2 billion conversations across 40-plus countries.
Gujarat-based ERP technology and AI solutions company Serpent Consulting Services is delivering Odoo, ERPNext, SAP, and Zoho implementations globally, while introducing Enterprise Automation Framework built directly into ERP software. “The challenge is no longer access to AI technology; it is operational readiness,” says CEO Jay Vora, pointing to “fragmented enterprise data, disconnected software platforms, unclear governance, and limited internal AI expertise” as the principal barriers organisations face when integrating AI into everyday operations.
Jay further added, “Our solution connects business functions into one intelligent ecosystem where AI continuously assists employees and management in making informed decisions while automating repetitive tasks.”
Providing AI-based video intelligence for real-time monitoring, Tarsyer Insights has built a plug-and-play solution that integrates with existing camera infrastructure without additional hardware. “Most AI video systems work fine in one store or one site during a demo. The hard part is making them work the same way across five hundred sites,” says Venky Nayar, founder & CEO.
“The real challenge was never building an accurate AI model. The real challenge is making it work reliably everywhere – in factories, ports, stores, and warehouses where the internet isn’t always stable and companies don’t want their data leaving the premises.”
Tarsyer says it has converted more than 400,000 cameras into intelligent analytical tools across over 20,000 sites in seven countries.
