The rapid rise of AI is changing the scale and design of digital infrastructure, with higher compute densities, GPU deployments and resilient power becoming central to datacenter expansion. In this exclusive conversation with DT, Sunil Gupta, Co-Founder, CEO and MD, Yotta Data Services, discusses Yotta’s AI infrastructure strategy, investments, sustainability efforts and India’s opportunity to build sovereign AI capabilities.
Q: What is your current capacity and expansion roadmap till 2030?
A: Artificial Intelligence is driving the biggest transformation the data center industry has seen in decades. Industry estimates suggest global demand for data center capacity will grow at an annual rate of 19 to 22 percent through 2030, driven largely by AI workloads. That growth is redefining what digital infrastructure needs to deliver. Higher compute densities, large-scale GPU deployments, energy efficiency and resilient power infrastructure are becoming fundamental design requirements rather than future considerations.
For India, this is a defining opportunity. The next phase of growth will depend on creating sovereign, scalable and sustainable infrastructure that supports enterprises, researchers, startups and government alike. That is the direction in which the industry is moving, and it is the approach we have adopted at Yotta.
Our expansion strategy is centered on building integrated AI infrastructure rather than standalone data center capacity. Yotta is India's largest and only vertically integrated Sovereign AI Infrastructure Platform, combining hyperscale AI-ready data centers, sovereign cloud (Yntraa), AI cloud (Shakti), GPU infrastructure and managed AI services under a unified architecture. We operate AI-ready facilities across Navi Mumbai, Greater Noida and GIFT City. Our Navi Mumbai campus has a planned power outlay of 2 GW, while Greater Noida is designed to scale to 350 MW. Together with Shakti Cloud and our sovereign cloud platforms, this creates an ecosystem that enables organisations to build, train and deploy AI applications on infrastructure hosted in India.
Today, we have more than 10,000 NVIDIA GPUs, including H100 and L40S GPUs, supporting enterprise AI, sovereign AI initiatives, foundation model developers and research institutions. As demand for AI infrastructure continues to accelerate, we expect to scale to more than 85,000 NVIDIA GPUs. Yotta plans to invest an additional $6 billion to expand its artificial intelligence infrastructure. We are also targeting a total capital raise of $1.2–1.5 billion through the combined pre-IPO and IPO process and have already secured over $150 million as part of the pre-IPO round.
We see this as an important step in helping India transition from being predominantly a consumer of global technology to becoming a producer of AI infrastructure, compute and intelligence.
Q: What are your biggest challenges: power, land, or funding?
A: Building infrastructure at this scale requires long-term thinking. Reliable power, suitable land, connectivity, engineering capability and execution discipline are becoming increasingly important as AI deployments become larger and more power intensive. Our latest AI facilities have been specifically engineered for large-scale NVIDIA GPU clusters. For high-density AI environments, Direct-to-Chip and Immersion Cooling technologies significantly reduce cooling energy consumption while supporting extremely high rack densities required for AI training and inference. Liquid cooling systems operate in sealed closed-loop configurations with minimal water loss while improving cooling efficiency and reducing overall facility energy consumption.
Q: How are you balancing growth with sustainability goals?
A: We source the bulk of our power from green energy sources, including hydro, solar and biomass wherever possible. Yotta D1 currently operates on 100 percent green energy at its current load, while NM1 operates with 80 percent green energy. Our facilities are supported by internationally recognised sustainability certifications and energy management systems because growth and sustainability have to progress together.
Q: Are we heading towards overcapacity or is demand still under-served?
A: I believe the industry is still addressing an expanding opportunity rather than approaching overcapacity. Despite generating nearly 20% of the world's data, India hosts only about 3% of global data centers. The real challenge is not whether enough capacity will be built. It is whether the industry can build the right infrastructure, with the right power, engineering capabilities and execution, quickly enough to support the next wave of AI innovation. This will be extremely important to ensure that India moves from the experimentation phase to production phase. That is where India has an opportunity to lead, and it is the opportunity we are investing behind.”
𝐒𝐭𝐚𝐲 𝐢𝐧𝐟𝐨𝐫𝐦𝐞𝐝 𝐰𝐢𝐭𝐡 𝐨𝐮𝐫 𝐥𝐚𝐭𝐞𝐬𝐭 𝐮𝐩𝐝𝐚𝐭𝐞𝐬 𝐛𝐲 𝐣𝐨𝐢𝐧𝐢𝐧𝐠 𝐭𝐡𝐞 WhatsApp Channel now! 👈📲
𝑭𝒐𝒍𝒍𝒐𝒘 𝑶𝒖𝒓 𝑺𝒐𝒄𝒊𝒂𝒍 𝑴𝒆𝒅𝒊𝒂 𝑷𝒂𝒈𝒆𝐬 👉 Facebook, LinkedIn, Twitter, Instagram
