Tata Communications To Pour Up To $450 Mn To Build AI Connectivity Infrastructure

Tata Communications plans to invest $400-450 Mn (around ₹4,350 Cr) to build 10,000 km of AI-ready connectivity infrastructure across India in a bid to capitalise on rising demand for data centre connectivity and compute infrastructure.

The investment will be used to build a new high-speed fibre network connecting major data centres and AI hubs, allowing transfer of large amounts of data faster with minimal delays. The network will also connect to Tata Communications’ existing global network.

The company plans to add around 6 petabits per second (pbps) of network capacity under the programme, taking its total capacity to approximately 7.25 pbps once the expansion is completed. The proposed capacity addition is scheduled between FY30-36 and will be financed via operating cash flows.

Its existing network capacity utilisation stands at around 70%-80%.

Tata Communications said the investment will help it cater to demand from hyperscalers, AI platforms, cloud providers, global capability centres (GCCs), enterprises and digital-native businesses.

The network is designed to support the high capacity, reliability and low latency required to move data between computing infrastructure and data centres.

The first phase will focus on the Mumbai-Pune-Hyderabad-Chennai corridor, one of India’s key AI and hyperscale growth routes. Subsequent phases will extend connectivity to emerging data centre clusters in other parts of the country as well.

The expansion is expected to connect India’s key AI and data centre hubs with international cable landing stations, strengthening connectivity between domestic infrastructure and global AI hubs and allowing for large volumes of data to be shared quickly and reliably between computing facilities. 

“The AI economy needs more than compute. It needs a network built for speed, scale and resilience,” said Tata Communications’ CPTO Rupesh Chokshi.

E2E Networks To Add Compute Capacity

The announcement comes as Indian infrastructure providers step up investments to support growing demand for AI training and inference workloads.

Earlier today, cloud and AI infrastructure provider E2E Networks announced plans to expand its AI factory capacity by nearly 59%, adding around 5.4 megawatts (MW) of capacity through NVIDIA HGX B300 systems and other Blackwell GPUs.

The proposed expansion will take E2E Networks’ AI factory capacity to approximately 14.6 MW. The company said the additional capacity will support large-scale model training, post-training, production inference, agentic AI and multimodal applications.

The company is seeing demand from foundation model builders, AI-native startups, enterprises, higher education and research institutions, and public-sector programmes. It also highlighted growing demand for AI infrastructure hosted in India, particularly among organisations operating in regulated and data-sensitive environments.

“India’s AI builders are moving beyond the pilot stage. They now are actively consuming AI tokens in production at previously unheard of scales,” said Tarun Dua, founder and managing director of E2E Networks.

This capacity expansion comes on the back of heightened demand for E2E Networks’ GPU services. Its operating revenue jumped 411% to ₹223.8 Cr in the second quarter of FY27 from ₹43.8 Cr in the year-ago quarter.

It posted a consolidated net profit of ₹72.7 Cr during the quarter under review, reversing a net loss of ₹13.5 Cr a year earlier. It also signed a multi-year cloud GPU contract worth approximately ₹1,000 Cr during the quarter. 

The post Tata Communications To Pour Up To $450 Mn To Build AI Connectivity Infrastructure appeared first on Inc42 Media.