‘AI is moving from assistive to autonomous; from copilots to agents’

The next generation of enterprise software will not just have AI features, but will behave like an operating model: autonomous, resilient, and sovereign, Kalyan Kumar, Chief Product Officer at HCLSoftware, told DH.

HCLSoftware is the software business division of IT services company HCLTech. The company displayed its latest enterprise solutions at the India AI Impact Expo 2026, and it shared how it orchestrates experience, data and operations to take AI from pilots into production outcomes.

Kumar said AI is moving from assistive to autonomous — from copilots to agents that can initiate and complete work end-to-end.

“Our focus is building what we call a self-driving enterprise capability, but with governance-by-design so autonomy scales without losing accountability,” he said. That means three solution thrusts:

Sovereign-by-design AI stacks: (Choice of deployment: on-prem, private cloud, sovereign infrastructure), because enterprises are rethinking what must remain under their control

Data-first intelligence: Untangling data from apps, mastering metadata, and creating the orchestration layer that makes AI reliable in complex enterprises

Agentic Services as Platform (ASaP): Codifying repeatable services (ops, security, marketing, workflow) into software + AI, so outcomes are consistent, measurable, and scalable

According to Kumar, the summit is a turning point because it is explicitly about impact — moving from AI experimentation to measurable outcomes.

India’s role in this AI era is evolving from being known primarily as a talent hub to becoming a trust-and-scale hub — a place that can operationalise AI for real problems, across languages, industries, and populations. India’s experience in building large-scale digital public infrastructure is increasingly seen as a blueprint for deploying technology responsibly at national scale, he added.

India is putting real weight behind AI capacity — compute, data centres, and sovereign approaches — because the next phase of AI is not just about models, it is about who controls the stack, where the data lives, and how governance is built in. “India’s opportunity is to be the place where AI becomes trustworthy, affordable, and deployable at scale — not just impressive in a lab,” the chief product officer said.

“Our ‘India play’ is not just about building in the country — it’s about enabling sovereign, compliant AI for Indian enterprises. For example, we’ve announced a sovereign AI stack approach with Sify that keeps enterprise data within national borders and under Indian jurisdiction — especially relevant for BFSI, healthcare, and government,” he said.

Talking about clients, Kumar said clients’ interest in AI solutions is higher than ever, and that it is also more mature. Clients are asking tougher questions now: “What’s the ROI (Return on investment)? How do we secure it? How do we govern it? How do we keep control of data and continuity?”

“We’re seeing many organisations hit the same wall: they can run pilots, but struggle to scale. And the blocker is rarely which model — it’s the enterprise reality of fragmented data, integration complexity, and missing governance. We’re also seeing a strong shift to hybrid and private AI stacks for sensitive workloads — because as AI becomes core to operations, choice and control become board-level requirements,” he explained.

Kumar said that they are not sprinkling AI on top of legacy environments, they are building a unified engine that orchestrates Experience, Data and Operations (XDO), so AI decisions are grounded in real enterprise context. He pointed out that customers see benefits in five areas:

Faster time-to-value: Move from pilots to production with repeatable patterns and orchestration

Governed autonomy: AI agents that can act, but with guardrails, auditability, and accountability

Sovereignty and compliance: Control across IP, data, technology, and commercial continuity — especially critical in regulated industries

Reduced complexity for regulated AI: a managed ‘sovereign foundation + applications’ approach, so enterprises don’t have to assemble the stack themselves.

Security built in: Embedding security and remediation into the life cycle (not bolted on at the end)

The real benefit is not ‘AI inside the product’ — it is AI you can trust, govern, and run at scale without losing control, he added.