As India races toward an AI-driven future, one truth is becoming unmistakable: the nation’s leadership won’t be defined by the number of GPUs it owns, but by how intelligently and inclusively it builds its digital foundations. Few leaders understand this shift as clearly as Neelakantan Venkataraman, Vice President and Global Head of Cloud, AI and Edge Computing Business at Tata Communications, who sits at the intersection of cloud, connectivity, and AI innovation.
At a national level, initiatives like the IndiaAI Compute Platform, which has already empanelled 34,000+ GPUs for academia and research, signal India’s intent to democratize compute access. Yet, as Venkataraman emphasizes, the real breakthrough will come from shared, federated frameworks, strong public-private models, and deep collaboration across telecoms, hyperscalers, and data centers—ushering in an era of AI-as-a-Service that lowers entry barriers for innovators across the country.
In this exclusive interaction with Express Computer, Venkataraman unpacks how India can move from AI adoption to AI influence—building sovereign, secure, and interconnected AI infrastructure that could help the country emerge as a global benchmark by 2030.
Some edited excerpts:
The global GPU shortage has created a bottleneck for AI development, especially for startups and academia. How do you see India navigating this challenge in the next two years?
The global GPU shortage has made one thing clear: access to compute cannot depend on hardware availability alone. India’s opportunity lies in building resilient and intelligently managed compute ecosystems that can scale AI workloads seamlessly across cloud, edge, and data center environments.
Tata Communications has built a full-stack cloud platform spanning Infrastructure -as-a- Service (IaaS) to Platform-as-a-Service (PaaS), offering both GPU-based and general compute capabilities. Our GPU stack, developed in partnership with NVIDIA, delivers the power needed for advanced AI workloads.
However, we go beyond infrastructure- we enable essential platform capabilities such as data management, databases, and key management services – giving enterprises the flexibility and control they need to build, deploy, and scale digital solutions seamlessly.
To make AI development accessible, we launched AI Studio, an end-to-end workspace where enterprises, startups, and data scientists can build and deploy models without worrying about infrastructure management. It supports fractional GPU allocation, pre-integrated tools, governance, and AI/ML Ops capabilities ensuring seamless experimentation.
In today’s AI-driven and data-intensive world, this breadth is critical for enterprises looking to innovate quickly and securely.
Initiatives like the IndiaAI Compute Platform, which has already empanelled over 34,000 GPUs for academic and research use, are important steps. As this ecosystem matures, the emphasis will shift from simply having more GPUs to distributing access more equitably, enabling thousands of smaller teams, researchers, and startups to participate in AI development at scale.
Do you believe India’s AI innovation potential is being limited more by infrastructure or by ecosystem readiness — and why?
The greater challenge today lies in ecosystem maturity. While compute capacity and digital infrastructure are expanding rapidly, the ecosystem still needs stronger collaboration between enterprises, startups, academia, and policymakers to accelerate adoption and innovation at scale.
At Tata Communications, we believe enabling this ecosystem means going beyond just infrastructure. Tata Communications Vayu cloud promotes a hybrid cloud model, giving customers the flexibility to run workloads on-premise or modern, cloud-native environments. This allows enterprises to harness the scalability, resilience and cost efficiency of the cloud while modernizing at their own pace and maintaining control over critical workloads.
Through initiatives like AI Studio, we’re providing enterprises and startups access to fractional GPU resources, pre-trained models, and secure development environments — lowering barriers to entry for AI experimentation.
With hyperscalers dominating compute capacity, what new collaborative models can make AI infrastructure more equitable and accessible?
Hyperscalers are central players in the AI ecosystem, but solving accessibility challenges will require more collaborative, hybrid approaches. One way forward is through partnerships where each layer of the digital stack connectivity, cloud, and edge works together to scale capacity and availability.
Tata Communications, for instance, enables enterprise and cloud connectivity through secure, high-performance networks that help organizations scale workloads in hybrid and multicloud environments. By offering interoperable platforms and connectivity that tie compute and data closer together, such collaborations can make infrastructure use more efficient and help more innovators participate in the AI economy.
You’ve often spoken about the convergence of cloud, edge, and connectivity. How can this triad help democratize access to AI compute and data resources in India?
The convergence of cloud, edge, and connectivity represents the foundation of India’s next AI leap. In a country as geographically and economically diverse as India, AI workloads can’t depend solely on centralized cloud resources.
Edge computing allows us to bring compute closer to the source of data be it in a factory, retail store, or farm which reduces latency, lowers costs, and enhances privacy. Cloud provides elasticity and scalability, while secure connectivity ensures that both environments communicate seamlessly.
This triad enables an AI model to be trained in the cloud, refined at the edge, and deployed securely across networks unlocking innovation in every geography. We have been building this connected fabric to ensure that access to compute and intelligence isn’t limited by location or scale.
Could India’s journey toward AI sovereignty hinge less on owning GPUs and more on creating shared AI infrastructure frameworks—federated, interoperable, and secure?
The real challenge lies in how the GPUs are governed and utilized. India’s AI journey can benefit more from building shared, federated, and secure AI infrastructure frameworks than from simply acquiring resources.
Federated architectures, for example, allow AI models to be trained across multiple institutions without moving sensitive data, enabling responsible and privacy-conscious development. Interoperability ensures that compute and data resources — whether in public clouds, private clouds, or edge environments — can work together seamlessly.
Robust security frameworks provide the trust necessary for organizations to collaborate safely.
In this way, AI sovereignty is less about isolation and more about enabling trusted collaboration, creating an ecosystem where innovation can thrive within India while remaining connected to global advancements.
How can partnerships between telecoms, data centers, and hyperscalers evolve into AI-as-a-service ecosystems that lower entry barriers for innovators?
We see this evolution already unfolding. AI-as-a-Service will thrive when infrastructure, connectivity, and platforms converge under a single, interoperable framework. Each stakeholder; telecoms, data centres, and hyperscalers brings a unique value: scale, proximity, and reach. The opportunity is in how we connect them.
At Tata Communications, our AI Cloud and Vayu Cloud form the foundation of this collaboration. Vayu’s multi-cloud fabric integrates hyperscaler environments, while our AI Cloud provides sovereign, high-performance GPU infrastructure optimized with Direct Liquid Cooling (DLC). On top of this, AI Studio acts as a collaborative layer, offering fractional GPU access, integrated AI tools, and a secure sandbox for experimentation.
How can the public and private sectors co-create AI sandboxes or open testbeds that replicate real-world environments for experimentation?
Public-private collaboration is fundamental to building real-world AI test environments. Governments can define the guardrails, compliance, data governance, and ethical use. Private players can contribute the platforms, compute resources, and connectivity to simulate production-scale environments.
Tata Communications’ role in this ecosystem is to bridge policy with infrastructure helping design environments where innovators can test safely, at scale, and in compliance with data localization norms. Such sandboxes will be crucial for sectors like healthcare, mobility, and logistics, where experimentation requires both high fidelity and high trust.
What lessons can India draw from countries like Singapore or the EU that are promoting sovereign AI clouds or AI commons?
These initiatives aim to ensure that data sovereignty and interoperability coexist. They reflect a broader push for digital and AI self-reliance, aligned with their own regulatory and societal priorities. For India, the emphasis should remain on adapting these lessons to our unique digital context. With our robust digital public infrastructure and trusted digital frameworks, India has the opportunity to shape a sovereign, inclusive AI ecosystem that promotes innovation and accountability in equal measure.
Looking ahead to 2030, do you believe India could emerge not just as a user but a shaper of global AI infrastructure paradigms?
India is already shaping global conversations around digital equity and secure connectivity, and the same potential exists in AI infrastructure. In next 5 years, India could stand out not for the size of its compute capacity but for how effectively it builds an inclusive digital foundation, one that blends cloud, edge, data governance, and innovation seamlessly.
India has the building blocks to influence global standards in AI infrastructure standards built on interoperability, openness, and accountability rather than mere scale.