AMD and TCS to Deploy 'Helios' AI in India

- The "Helios" platform is a rack-scale AI architecture featuring AMD's Instinct MI455X GPUs, next-generation EPYC "Venice" CPUs, and Pensando Vulcano networking chips, all integrated with the open-source ROCm software ecosystem. - This collaboration will establish an AI-ready data center blueprint with a capacity of up to 200 megawatts, which is powerful enough to run tens of thousands of high-performance AI servers simultaneously. - The project is a key part of India's "sovereign AI" initiatives, which aim to develop and train AI models domestically, reducing reliance on foreign infrastructure and ensuring data residency. - TCS is executing this initiative through its subsidiary, HyperVault AI Data Center Limited, which was established in 2025 to build secure, gigawatt-scale AI infrastructure for hyperscalers and enterprises. - The partnership places AMD in direct competition with Nvidia, which has also been expanding its presence in India through collaborations with companies like Yotta, Larsen & Toubro, and E2E Networks to build out AI cloud infrastructure using its GPUs. - This deal builds upon a previous strategic collaboration announced in January 2026 between TCS and AMD to co-develop industry-specific AI solutions and modernize hybrid cloud environments. - India's broader AI ambitions are backed by the "IndiaAI Mission," a government initiative infusing over $1 billion into the nation's AI ecosystem to enhance compute capacity and foster the development of sovereign AI datasets and models. - The country's data center capacity is projected to grow from about 1.5 gigawatts to over 10 gigawatts by 2030, with the government aiming to attract $200 billion in data center investments in the coming years.