Corporate conversations have shifted drastically over the past decade, moving from the cloud to software-as-a-service, and now landing squarely on artificial intelligence. However, while AI has become a non-negotiable boardroom mandate, a staggering 77 to 87 percent of enterprises across the Asia Pacific and Japan (APJ) region admit their infrastructure is not fully ready to support on-premises AI workloads.
Addressing this critical bottleneck was the focal point of Nutanix’s Q2 FY26 media roundtable, which highlighted real-world deployment struggles, a landmark $250 million investment from AMD, and a surge in customers seeking safe harbor from shifting industry tides.
The Edge AI Reality and the Infrastructure Fix
The rush to deploy AI is no longer confined to centralized data centers; it is rapidly moving to the edge, creating unique logistical nightmares. During the roundtable, specific regional use cases illustrated this mounting pressure.
In Australia’s mining sector, organizations must process complex occupational health and safety analytics before a miner ever enters a shaft—all with zero IT staff on site.
In India, the microbanking boom requires AI inferencing across thousands of remote branches plagued by limited connectivity and a lack of IT talent.
Meanwhile, Thailand’s retail market is demanding AI analytics directly on the shop floor near the goods and consumers to drastically reduce latency.
Globally, 82 percent of IT leaders admit their on-premise infrastructure simply isn’t ready for these workloads. To close this gap, Nutanix announced a new integration with Pure Storage (now EverPure) and FlashArray. By marrying high-performance NVMe storage with the Nutanix Cloud Platform, enterprises can finally build infrastructure that is truly AI-ready.
The impact of modernizing this architecture is already visible in the field. Take GC Biopharma, a company dedicated to creating plasma proteins and rare disease treatments.
Requiring massive stability and performance for drug development, the company relied on the Nutanix Kubernetes Platform to modernize its manufacturing systems, specifically praising the platform’s open standards for reducing vendor lock-in.
AMD’s Quarter-Billion Dollar Bet on Open AI
To further combat the global infrastructure deficit and maintain this open-standard philosophy, Nutanix has secured a massive strategic investment from AMD. AMD is investing up to $250 million into Nutanix, consisting of $150 million in common stock purchases and up to $100 million dedicated to co-engineering. The ultimate goal of this partnership is to construct an open, full-stack agentic AI platform.
Jay Tuseth, Vice President and General Manager for Asia Pacific and Japan, Nutanix, emphasized that enterprise customers are actively demanding open systems and choice in their infrastructure.
“We wanted to work with both NVIDIA, the market leader from that perspective, but also AMD, which is a strong contender in that same market,” Tuseth noted.
Because AMD features both a GPU and CPU model—both critical for training and inferencing large language models (LLMs)—this integration will allow organizations to download an LLM and create agentic AI applications within hours using a common control plane.
The Looming Threat of “Shadow AI”
As business units race to deploy these technologies, they are increasingly bypassing IT departments entirely. Deploying cloud-based AI is often just a “credit card swipe-away” for non-IT employees, leading to a dangerous phenomenon dubbed “Shadow AI”.
This rogue IT behavior is colliding with a wall of strict regional regulations. Companies are currently wrestling with Singapore’s MAS guidelines mandating strict AI oversight , Australia’s Privacy Act demanding high levels of capability transparency , and India’s tough data protection frameworks that dictate exactly where data can be stored and who can see it.
Daryush Ashjari, Chief Technology Officer and Vice President of Solutions Engineering Asia Pacific & Japan, Nutanix, explained that the solution is not for IT to become more rigid. Look no further than the University of Canberra.
Previously, different university departments were buying their own equipment, creating a sprawling Shadow AI governance nightmare. By implementing a unified Nutanix self-service portal, IT transformed from a “blocker” into an enabler, allowing researchers to safely and securely spin up their own virtual machines and Kubernetes infrastructure inside an approved sandbox.
Broadcom Defectors Drive Record Financial Growth
This urgent need to modernize AI architecture securely, combined with shifting industry dynamics, is reflecting clearly in Nutanix’s balance sheet. The company reported $723 million in revenue for Q2 FY26, achieving double-digit year-over-year growth. Notably, Nutanix secured over 1,000 new customer wins in the quarter, marking its highest customer intake in nearly eight years and pushing its global base past 30,000.
A significant catalyst for this growth is an industry-wide exodus from VMware following Broadcom’s acquisition. Tuseth pointed out that disruptive changes to the VMware partner ecosystem, combined with a forced, mandatory migration to VCF9 by the end of next year, have created a compelling event for customers. Enterprises across the public sector, healthcare, retail, and financial services are now actively seeking predictable, long-term partners.
Looking Ahead: The Value of Agentic AI
Ultimately, the race to implement AI isn’t currently driven by immediate Return on Investment (ROI); it is fueled by speed, urgency, and aggressive board imperatives.
However, the tangible business value remains clear: by automating mundane, labor-intensive tasks with agentic AI, organizations can free up critical budget and redirect resources toward further innovation.
As enterprises navigate this complex hybrid reality, the guiding principle for modern IT leadership remains straightforward, as Tuseth noted: experimentation is good, but fragmentation is bad.
Microsoft announces US$1 Billion investment in Thailand
Disrupt Health Impact Fund launches HealthTech Outlook 2026
