Yotta Data Services is preparing a substantially larger AI infrastructure push, with plans to invest about $7 billion by FY27 as demand for high-performance compute continues to reshape India’s data centre and cloud market. The company’s current expansion plan includes deploying 40,000 NVIDIA Blackwell GPUs over the next four months and scaling total GPU capacity to roughly 85,000 chips by the end of FY27.
The update builds on Yotta’s earlier commitment to deploy 20,736 liquid-cooled NVIDIA Blackwell Ultra GPUs, a programme the company positioned as one of Asia’s largest AI supercluster deployments. That earlier plan involved investment of more than $2 billion and was expected to go live by August 2026. The latest expansion signals a move from a single large deployment to a broader AI compute platform strategy, with GPU infrastructure becoming a core commercial line rather than a supplementary data centre service.
Yotta has also raised about $150 million at an estimated valuation of $3.9 billion, with the capital intended to support AI and cloud infrastructure expansion. The company has been preparing for a possible public listing and has attracted attention from investors because of the demand-supply mismatch in AI compute, particularly for enterprises that need GPU access but cannot build or operate large-scale clusters on their own.
The company’s infrastructure strategy is not limited to domestic workloads. A large portion of its planned capacity is expected to serve overseas demand, reflecting the global scramble for AI compute and India’s emerging role as a possible alternative location for GPU-backed cloud services. That international orientation could help Yotta balance domestic sovereign AI requirements with export-oriented infrastructure revenue, although execution will depend on chip supply, power availability, cooling capacity, customer commitments and data localisation requirements in different markets.
For Indian enterprises, the buildout comes at a time when AI adoption is moving from experimentation to deployment across banking, IT services, healthcare, manufacturing, retail and public-sector use cases. Local access to advanced GPU capacity can influence model training, fine-tuning, inference economics and data governance. It also gives Indian cloud and data centre operators a stronger role in the AI stack, a field that has been dominated by hyperscalers and specialist GPU cloud firms in the US and Europe.
The expansion places Yotta in a more strategic segment of India’s digital infrastructure market. The company is no longer competing only on colocation and cloud hosting; it is positioning itself in the infrastructure layer for frontier AI, enterprise AI workloads and sovereign compute demand.
