India’s data centre boom is still in its early stages, power infrastructure is less constrained than most major markets and engineering talent is already here in depth — but the question is whether Indian enterprises can move fast enough to turn those advantages into actual AI deployments at scale, said a top global executive.
Adam Judd, who oversees F5’s business across Asia Pacific, China and Japan as senior vice president of sales, says that while the US and China moved first, the fundamentals now point squarely at India for the next wave of AI infrastructure build-out.
“The next 12 to 18 months are going to boom here in India,” Judd told TechObserver.in’s Mohd Ujaley, pointing to the convergence of data centre tax incentives, available power infrastructure and sovereign AI ambitions as factors that most global observers are still underweighting.
Edited Excerpts:
India is one of F5’s fastest-growing markets and a major engineering hub. How does buying behaviour here — especially among banks, telcos and digital-native firms — differ from the rest of APCJ when it comes to AI and security? Are Indian enterprises leapfrogging or playing catch-up?
The US and China were first to market — there’s still very much an AI race there around models, deployments, data centres, GPUs. But when I look at the fundamentals of India, a few things sit clearly in its favour.
First, power. The country is only at about 50 per cent utilisation. There is no shortage right now. In many countries, including the US, power availability to run GPUs is a real concern.
Second, the government’s tax-free holiday on data centres is a clever and critical initiative, not just for Indian companies and hyperscalers, but for markets in Europe, the Middle East and Africa that are looking at India for disaster recovery and secondary data centre resiliency.
India has probably been slower on GPU adoption, but that does not matter. The fundamentals are extremely robust. Add to that one of the most inquisitive engineering markets in the world, arguably the largest development talent pool globally and it is natural that AI gravitates here. India is already the second-largest market for ChatGPT and for Claude.
The next 12 to 18 months are going to boom here because of the data centre build-outs, the absence of power restrictions, the skill sets, and government-backed investment.
Breach costs are rising fast and APIs now lead the risk landscape. When you sit across from CFOs being asked to fund yet another security investment, how do you frame the ROI? Has the industry moved beyond selling fear?
It is one of those things where you are damned if you do and damned if you do not.
Certain verticals have taken the lead. Across APCJ, the BFSI community is moving quickly. Government organisations have moved fast in certain areas. Digital-native companies, of course, are moving. Anyone seeing productivity gains through AI is moving.
Most AI applications have started internally, and I think that is a good thing — it’s become a sandbox, a testing ground for how organisations actually use AI. Companies have generally been quite wise about how they bring AI applications into production. They use hyperscalers to test, to determine whether an application has life and legs and whether there’s an ROI.
What they’re quickly working out, especially in this new world of agentic AI, is that the real costs are around tokens, not GPUs. And those costs are impossible to control if workloads are sitting outside your environment.
So we’re seeing companies trial AI in the public cloud and then rapidly productise it in a hybrid multi-cloud environment. That’s where they regain control of costs and gain real visibility into ROI — because they control their own tokenisation.
What is the single biggest operational challenge enterprises face as they scale AI initiatives?
Complexity debt. I have heard it from enterprises across Asia multiple times. The challenge is not just simplifying existing infrastructure. More importantly, in the world of AI, it’s about freeing up people. Costly, important people, who should be focused on innovation rather than babysitting infrastructure.
When you’re worried about digital natives using AI to replicate your business, you cannot afford to have your best engineers buried in operational complexity. You need to free up innovation inside the enterprise.
India is increasingly a Global Capability Centre hub, with many centres now handling core product engineering and R&D. What is F5’s India team actually building that goes into global products and how do you ensure local talent is seen as design leadership, not just execution?
We have never treated our engineering team in India as anything other than leading-edge development.
We have over a thousand engineers in Hyderabad and Bengaluru. Our first major product infusion from India was our AI-powered WAF — something we have been talking about heavily over the last couple of months. Managing traditional WAFs is difficult. Using AI, and more importantly the networking effect of the neural networks we have built across the world as one of the largest WAF providers, creates incredible value for customers who are now using AI-driven tools for application security.
Expect all of our products to become AI-powered or AI-enabled over time. And a lot of that innovation is going to come right here from India.
