As businesses confront an era defined by soaring data volumes and the rapid advance of artificial intelligence, many leaders are reassessing what truly underpins a resilient digital strategy. For Srikanth Katuri, Group CIO, Omega Healthcare, the answer is unequivocal: data must be the organising axis around which modern enterprises build, innovate and govern. His perspective cuts through the noise, emphasising not only the strategic value of data but also the operational discipline required to manage it responsibly. In this conversation, he offers measured, pragmatic insights into how organisations can harness data and AI to drive clarity, reduce risk and accelerate intelligent transformation.
You come from one of the most critical domains, data and data privacy. How do you view the growth of data and its connection to AI within organisations today?
Data has been at the centre of all conversations for the past decade: technical, regulatory, and strategic. I documented this in my Swivelberger Framework, which states that data must be the axis around which infrastructure and applications revolve. That remains true today and will for years to come.
Because we work extensively with PHI and PII, our responsibility is significant. We launched a Data Security Posture Management strategy, starting with understanding what data we hold, where it resides, and why it exists. More data increases risk and liability, so reducing our data corpus is essential. With over 20 years of global data and multiple acquisitions, the journey is large but necessary.
Once the data is classified and understood, how do organisations move toward extracting insights? Are we ready for that?
Insight extraction is possible only after proper classification, and that process is challenging. Each company interprets data differently, based on internal practices cultivated over decades. Even when aligning with global standards like ICH, business teams may have differing views.
We developed ML algorithms to automate classification, but what is “correct” versus what is “accurate” can differ significantly. Our discovery process has been ongoing for 90–120 days, and we’re still uncovering new data. AI is essential here. Without it, navigating decades of contracts, regulatory changes, and staff transitions becomes impossible. Legal teams may not even have access to old agreements defining data retention. So yes, AI-driven discovery and classification is not just useful; it’s necessary.
Organisations today deal with massive data growth across countries. How critical is automation in handling this complexity?
Automation has become a fundamental necessity in managing the scale and complexity of modern data ecosystems. We are already seeing significant progress, with Level 1 tasks substantially automated and Level 2 tasks slowly moving in that direction. However, the industry is still building the skills required to fully leverage AI in these workflows. Teams are learning how to structure the right queries, validate responses, and integrate AI-driven outcomes into their operations. It is a long journey, but the trajectory is clear: over the next two years, a large portion of operational work will shift toward automation, leaving only the highest-value decisions and oversight to human teams.
Do you see organisations moving toward more autonomous and agentic AI systems in the near future?
We are absolutely moving towards a future dominated by agentic AI. One major catalyst is the emergence of what analysts are calling “digital nations,” where countries and regions assert stronger jurisdictional control over their data. Nations like India or regulatory regions like the EU increasingly insist that their data cannot be freely used for training external models. This shift forces enterprises to build AI systems that are localised, context-aware, and capable of operating within strict data boundaries. Agentic AI becomes central in such an environment because these systems can adapt to regional constraints while delivering tailored outcomes. With India alone representing 1.4 billion data-rich citizens, the move toward data sovereignty is accelerating, and AI must evolve to meet it.
There is constant debate around the maturity of AI. Is AI still early, is it a bubble, or is enterprise AI already real?
AI in the enterprise is very real and has been present for many years, long before generative AI captured mainstream attention. Technologies like RPA, predictive engines, and intelligent automation have been quietly powering enterprise workflows for over a decade; we had more than 1,400 bots running well before the rise of modern LLMs. What is changing now is the shift toward agentic AI, where systems can autonomously perform specialised tasks with minimal intervention. In the coming years, smaller, domain-focused language models will become more prominent, and many AI capabilities will move closer to the edge. Routine tasks, such as summarising communications or triggering follow-up actions, will become fully automated. AI is not a future concept; it is a present reality that is rapidly maturing and embedding itself deeper into enterprise operations.
As enterprises begin adopting AI at scale, what is the biggest takeaway for leaders and teams?
My biggest advice is simple: don’t limit your learning to AI tools, talk to people. Speak with colleagues, industry peers, and even competitors. Real-world conversations offer context, nuance, and practical insights that no AI system alone can provide. I make it a point to attend every event where data or AI is discussed because something new always emerges. A hundred industry conversations will give you far better answers than a handful of AI prompts. Stay curious, stay connected, and stay informed.