By Vijay Kasi, Partner, Kearney and Abhinav Priyanshu, Senior Operations Manager, Kearney
India is entering a new phase of AI readiness. Microsoft’s latest Hyderabad cloud region adds another local deployment option for cloud and AI workloads. AWS, Google Cloud and Oracle already have established India footprints. Together, these investments signal a market that is moving beyond a simple question of whether cloud is available. The more consequential question is whether Indian enterprises are ready to redesign work around AI.
The infrastructure story is real, but it is not the main story
Microsoft’s launch of its India South Central cloud region in Hyderabad is an important milestone for India’s AI ambitions. The new infrastructure matters. It gives enterprises more choice in where to place workloads, greater scope to design for resilience, and more local capacity for data-intensive applications and AI inference. For regulated sectors, it can also strengthen options around data location, controls and continuity. But a new cloud region does not make an organisation AI-ready by itself.
Indian public-cloud spending is forecast to reach $17.5 billion in 2026, up 28.1 percent year on year. Infrastructure-as-a-service spending is expected to grow 40 percent, while platform-as-a-service spending will reach $6.4 billion. The investment is being driven by demand for AI-ready infrastructure, application modernization and more scalable technology models. This is a clear sign of momentum.
India has a distinctive opportunity to lead in AI enabled operations
India’s advantage is not limited to its technology talent or its growing cloud footprint. It is also operational. The country hosts large shared-service organizations, global capability centres, technology delivery teams and process-intensive businesses that can become laboratories for AI-led work redesign. In these environments, AI can do more than automate a discrete task. It can reshape how an end-to-end process is executed, controlled and improved.
Consider procurement, finance operations, customer service, supply chain planning, software engineering or risk operations. Traditional automation has focused on structured, rules-based activities. Generative and agentic AI extend the frontier to language-heavy, analytical and judgment-adjacent work: triaging exceptions, synthesizing information, preparing recommendations, checking policies, coordinating actions and continuously monitoring signals.
The prize is not simply a lower cost-to-serve. It is faster cycle times, better decisions, fewer handoffs, stronger control and more capacity for teams to focus on complex customer, commercial and strategic work.
Four moves for Indian CEOs
| Executive move | What it means in practice |
| Start with work, not tools | Select a small number of high-value, end-to-end processes. Break them into tasks, identify where human judgment is essential, and redesign the sequence of work around the most effective executor: human, rule-based automation or AI agent. |
| Treat released capacity as a leadership choice | Make an explicit decision about whether capacity will be removed, reinvested in growth and service, or used to improve control and resilience. Without this choice, benefits are usually absorbed into the organisation. |
| Build governance into the workflow | Move beyond policy documents. Embed data access, approval thresholds, human oversight, traceability and exception management into the design of AI-enabled processes. |
| Make cloud and AI economics visible | AI is consumption heavy. Establish FinOps discipline across compute, models, data movement and vendor commitments – and measure value at process level, not simply by the number of pilots or licences deployed. |
What this means for India’s cloud agenda
The public-cloud conversation in India is evolving from migration to workload placement. Enterprises will increasingly decide which AI workloads belong in a public cloud, a private environment or a hybrid setup – based on performance, data sensitivity, economics and resilience. They will also use multiple providers where there is a clear business case, rather than allowing accidental multicloud complexity to develop over time.
This is why additional local infrastructure is relevant. It expands the feasible choices. It can support faster, more resilient and more compliant deployment of selected workloads. But it does not answer the harder questions: Which processes should be re-engineered? Which decisions should be augmented or automated? Which skills will be required? How will value be measured? And how will risk be governed at scale?
India has the ingredients to become a leading market for AI-enabled operations: a large and sophisticated technology workforce, process scale, mature digital ecosystems and a rapidly expanding infrastructure base. The next source of differentiation will be leadership discipline. Enterprises that treat AI as a work-redesign and operating-model agenda will build an advantage that is difficult to replicate. Those that treat it as a collection of tools may accumulate activity without capturing the value.
The question for Indian CEOs is no longer whether AI infrastructure is arriving. It is whether the enterprise is prepared to redesign around it.