As India progresses with its artificial intelligence goals, the discussion is moving away from the AI models to the infrastructure that supports them. As hyperscalers continue their growth, issues around computing accessibility, data sovereignty, local manufacturing, and sustainable infrastructure continue to dominate the discourse. In an interview with Tech Achieve Media, Priya Krishnamurthy, Director, Altos India, comments on the importance of democratizing AI infrastructure for India’s next growth phase, the importance of the company’s ‘Make in India’ initiative, and the requirements for India to transform from an AI consuming country to an AI infrastructure and innovation country on the global stage.
TAM: Acer India wants to become a global AI hub, but infrastructure remains a bottleneck. What gap is Altos hyperscalers and existing OEMs not addressing yet?
Priya Krishnamurthy: India has made significant progress in building AI-ready infrastructure, but one of the biggest gaps today is accessibility and democratization of compute. Large hyperscalers and global OEMs have built substantial infrastructure. However, much of that capacity is still concentrated within a limited ecosystem. It is not easily accessible to researchers, educational institutions, startups, or public-sector innovation initiatives at an affordable scale.
At Altos, we believe the next phase of India’s AI journey will depend on enabling broader access to AI infrastructure, not just creating capacity. Our focus is on building locally integrated AI infrastructure that can be deployed faster, managed securely within India, and tailored for Indian enterprise and sovereign AI requirements. We also see a strong need for infrastructure that balances performance with sustainability and energy efficiency, especially given India’s power and resource constraints.
TAM: You’ve positioned your servers as “Make in India” AI servers. How much of the value chain is genuinely local today? and what’s the roadmap to deepen localisation?
Priya Krishnamurthy: Today, over 50% of the server components by value in our AI servers are locally manufactured or sourced in India, which qualifies them under Class One manufacturing norms. Beyond components, a significant part of the engineering, integration, and manufacturing activities is being executed in India through our local teams and EMS partners.
Currently, around 80–85% of the activities involved in bringing these systems to market, including engineering, manufacturing, and integration, are India-centric. However, critical components such as CPUs and GPUs are still globally sourced because India’s semiconductor ecosystem is still evolving.
Our long-term roadmap is to deepen localisation progressively as India’s component ecosystem matures. We are already seeing investments in areas like memory and storage, and with stronger government support and scale, we expect India to build greater self-reliance in high-performance computing infrastructure over the next few years.
TAM: Which sectors in India are showing the strongest immediate demand for dedicated AI infrastructure?
Priya Krishnamurthy: We are currently seeing strong interest from education and research institutions, government-led initiatives, manufacturing, BFSI, healthcare, and enterprises working with sensitive or proprietary data.
A major driver is the growing need for data sovereignty and secure AI deployments within India. Organisations increasingly want AI workloads to run either on-premises or within trusted private cloud environments rather than relying entirely on external public cloud platforms.
There is also growing demand from enterprises building domain-specific AI models, where dedicated infrastructure offers better performance, lower long-term costs, and stronger control over data privacy and compliance requirements.
TAM: Do you see India evolving primarily as an AI consumption market, or can it become a true AI infrastructure and innovation exporter? What needs to change for that?
Priya Krishnamurthy: India has the potential to become far more than just an AI consumption market. The country has the talent base, developer ecosystem, and scale to emerge as a global hub for AI innovation and infrastructure.
However, for that to happen, India must continue investing in three critical areas: advanced compute infrastructure, local manufacturing capabilities, and broader access to AI resources. Affordable and scalable AI infrastructure needs to reach universities, startups, research institutions, and enterprises more widely.
At the same time, stronger domestic capabilities in areas like semiconductors, GPUs, and advanced electronics manufacturing will be important in reducing dependency on global supply chains over the long term. Public-private collaboration and policy support through initiatives like the IndiaAI Mission will also play a key role in accelerating this transformation.
TAM: How critical are partnerships with India’s system integrators and cloud providers to your go-to-market strategy?
Priya Krishnamurthy: These partnerships are extremely important to our strategy because AI infrastructure today is not just about hardware. Customers require a complete ecosystem that includes compute, software stacks, deployment expertise, cloud integration, and long-term support.
We are working closely with system integrators, cloud providers, and software partners to ensure customers can deploy AI workloads efficiently and securely. These partnerships also help us address industry-specific requirements and accelerate adoption across enterprise and government environments. India’s AI ecosystem is evolving rapidly, and collaboration across infrastructure providers, software developers, and service partners will be essential to scaling adoption effectively.
TAM: The Indian government is pushing initiatives like the IndiaAI Mission. How aligned is Altos with these national ambitions, and do you expect policy support to accelerate adoption?
Priya Krishnamurthy: Our approach is closely aligned with India’s broader sovereign AI and digital infrastructure vision. The IndiaAI Mission is an important initiative because it aims to make advanced AI infrastructure more accessible to a wider ecosystem, including researchers, startups, educational institutions, and enterprises.
At Altos, our goal is to support this vision by building AI infrastructure that enables organisations to deploy and manage AI workloads securely within India, whether through on-premises environments or private cloud deployments.
We also believe policy support will play a significant role in accelerating adoption. Government incentives and manufacturing initiatives are already helping strengthen India’s local technology ecosystem, and continued support can further accelerate investments in AI infrastructure, domestic manufacturing, and compute accessibility.
TAM: What does success for Altos in India look like? Would it be market share, local manufacturing scale, or becoming foundational to India’s AI stack?
Priya Krishnamurthy: For us, success is not defined by just one metric. Market presence and manufacturing scale are important, but the larger goal is to become a meaningful contributor to India’s AI ecosystem.
We want Altos to play a foundational role in enabling India’s AI ambitions by providing scalable, secure, and locally integrated AI infrastructure. If Indian enterprises, research institutions, startups, and government initiatives can build and deploy AI more effectively using infrastructure developed and integrated in India, that would be a significant achievement for us.
At the same time, scaling local manufacturing and strengthening India’s AI infrastructure ecosystem remain key priorities in our long-term vision.
TAM: If you had to be honest, what is the single biggest challenge that could slow India’s AI ambitions?
Priya Krishnamurthy: The biggest challenge is balancing rapid AI infrastructure growth with the realities of resource availability and ecosystem maturity. AI infrastructure requires enormous compute power, electricity, cooling, and advanced supply chains, and India still faces structural constraints in areas such as power availability and access to critical components like GPUs.
At the same time, compute infrastructure must become more accessible and affordable for a broader ecosystem. If advanced AI resources remain concentrated within a small segment of the market, adoption at scale could slow down.
Going forward, India’s success in AI will depend on how effectively the country expands infrastructure, strengthens domestic manufacturing capabilities, improves access to computing, and builds sustainable, energy-efficient systems that can scale responsibly over the long term.
