Shadow AI, infra gaps challenge enterprise AI rollout in India: Nutanix execs

Indian enterprises are accelerating their adoption of artificial intelligence (AI), but governance risks, infrastructure readiness gaps and organisational silos are emerging as key hurdles as companies move from experimentation to large-scale deployment, according to the latest Enterprise Cloud Index (ECI) report released by Nutanix on Thursday.

The eighth annual global study, conducted by Wakefield Research, surveyed IT decision-makers on trends in cloud adoption, containerisation and AI deployment. The findings indicate that while AI is driving rapid infrastructure modernisation worldwide, Indian organisations are moving particularly quickly toward containerised architectures even as CIOs and CISOs grapple with governance, data security and operational complexity.

Shadow AI emerges as a governance risk

One of the most pressing concerns for enterprise technology leaders is the rise of “shadow AI”—the use of AI tools by employees outside formal IT oversight.

Speaking at a select media roundtable on the report, Daryush Ashjari, CTO (APJ), Nutanix, said shadow AI is rapidly emerging as one of the most significant governance risks for enterprises globally.

“Shadow AI—the use of AI tools by employees outside official IT oversight—is rapidly emerging as one of the biggest governance risks for enterprises globally. As employees experiment with generative AI tools outside formal IT oversight, organisations face growing exposure to data leakage, compliance breaches and intellectual property risks,” Ashjari said.

The report found that 73% of organisations in India have already encountered AI applications or agents being deployed outside IT functions, while 96% of IT executives believe such usage creates business risk. This level of concern is higher than the global average of 87%, indicating that Indian enterprises are particularly sensitive to the potential data security and compliance risks associated with unregulated AI adoption.

“Shadow AI is quickly becoming one of the most significant governance challenges for enterprises as AI adoption accelerates across business functions,” Faiz Shakir, VP and Managing Director, India & ASEAN, Nutanix, told TechCircle in an exclusive interaction.

“Enterprises should pivot toward standardised, secure platforms that allow for visibility and responsible adoption. By breaking down silos, IT can move from being a bottleneck to an enabler, providing the structured environments employees clearly crave for AI innovation,” he said.

Ashjari further suggests that enterprises will need stronger governance frameworks and secure platforms that allow innovation while ensuring visibility and control.

Containers, data sovereignty shape AI infrastructure

Beyond governance concerns, the report highlights how AI is reshaping enterprise infrastructure strategies. Globally, 87% of organisations say AI is accelerating their adoption of containers, as companies seek faster and more scalable ways to run modern applications and AI workloads.

“India has emerged as the global frontrunner in the shift toward modern, containerised architectures, outperforming every other market in its commitment to this transition,” Shakir said.

According to the survey, 97% of Indian organisations expect container adoption to increase over the next three years, the highest rate among markets surveyed, while 82% say they are already building new applications using containers.

“Whilst AI is clearly the engine driving this modernisation, the real challenge for Indian enterprises isn’t just adopting the technology, it’s breaking down the organisational silos that currently hinder execution. To truly capitalise on the potential of AI agents and autonomous tools, organisations must build a unified, secure foundation that bridges the gap between IT and business units,” he added.

Containers—software packages that bundle application code and dependencies—are increasingly forming the backbone of AI-ready architectures because they allow applications to run consistently across hybrid environments, from on-premises data centres to public cloud platforms.

At the same time, infrastructure decisions are increasingly influenced by data sovereignty considerations. The report found that 82% of Indian organisations consider data sovereignty a high priority when deciding where to run applications and infrastructure, while 57% say they need to keep infrastructure within a single country, either on-premises or through a local cloud region.

This growing emphasis on localisation reflects both regulatory requirements and rising customer expectations around how sensitive data is stored and processed. However, Shakir noted that the shift does not indicate a retreat from cloud adoption.

“Organisations are adopting hybrid infrastructure strategies that allow them to maintain control over sensitive data while supporting modern applications and AI workloads,” he said. “The ability to balance compliance, portability and operational flexibility will remain central to infrastructure decisions.”

AI pilots outpacing infrastructure readiness

Despite strong momentum in AI initiatives, the study suggests many organisations are still grappling with infrastructure readiness challenges.

Across the Asia-Pacific and Japan (APJ) region, enterprises are gradually shifting from AI experimentation toward revenue-generating deployments, though several structural barriers remain. Many AI pilots are still confined to experimentation stages because existing infrastructure cannot support the elastic, governed workloads required for AI agents at scale, Ashjari said.

At the same time, enterprises are being pushed to move beyond basic automation toward agentic AI systems capable of reshaping business processes and enabling entirely new services and revenue streams.

However, the region’s diversity—including language complexity, varying levels of infrastructure maturity and stringent data sovereignty requirements—means that Western AI deployment models cannot always be replicated directly across APJ markets.

In India specifically, infrastructure readiness remains a major concern. Nearly 81% of enterprises believe their current on-premises infrastructure is not fully prepared to support intensive AI workloads, highlighting a gap between strategic AI ambitions and operational capability.

“The transition from AI experimentation to full deployment has revealed a significant readiness gap,” Shakir said.

“To bridge this, companies must prioritise unified, data-intensive platforms capable of running containerised workloads at scale while meeting data sovereignty requirements. The winners will be those who establish a secure, consistent hybrid foundation that allows them to scale AI initiatives effectively without compromising security or compliance.”

The findings underscore the expanding role of CIOs and CISOs in shaping enterprise AI strategies, as organisations seek to align governance frameworks, infrastructure investments and business teams to support secure and scalable AI deployment.