Over the past two years, I have seen the conversation with customers change fundamentally. The question used to be whether AI would become relevant to their operations. That question has been answered. The conversation now is about whether their infrastructure is ready to support it. But behind every AI application is something less visible and just as important: infrastructure.

That is where data centres come in.

Today, data centres are not just storage and compute facilities. They are becoming the backbone of a much larger digital shift. As AI adoption grows, the expectations from infrastructure are changing with it. We now need facilities that can support higher densities, faster growth, and more demanding workloads without losing reliability or efficiency. BCG’s research on the sector reflects this shift clearly, with generative AI and hyperscale demand emerging as major drivers of growth.

In practical terms, AI-readiness is not one single feature. It is a way of thinking. It starts with design. It continues through planning, execution, and operations. It means asking the right questions early: how much density will the facility need to support, how will cooling perform at scale, how flexible is the architecture, and how ready is the campus for change over time?

These are not theoretical questions anymore. They are part of the real conversation between customers and infrastructure partners.

AI workloads are also changing the technical requirements of modern data centres. Unlike traditional enterprise applications, AI training and inference environments require significantly higher computing densities powered by advanced GPU clusters.

This increases both power consumption and heat generation, placing greater demands on electrical infrastructure, cooling systems, and overall facility design. This is not a future problem. We are engineering for it now. High-density rack configurations, liquid cooling systems, and advanced thermal management are no longer niche considerations — they are becoming baseline requirements for any facility that wants to remain relevant. At Sterling and Wilson Data Center, we have been investing in exactly these capabilities, because we believe the window to build the right foundation is now, not after your customers have already outgrown you.

What is also becoming clear is that AI is pushing the industry to think more carefully about efficiency. The more powerful digital workloads become, the more important it is to build responsibly. That means smarter energy use, better cooling strategies, and infrastructure that is designed to perform over the long term. The energy
implications of AI and data centre growth are now being discussed at the highest levels across the industry.
For the industry, this is not a concern. It is an opportunity. AI-ready facilities will need to balance performance, scalability, efficiency, and sustainability while remaining flexible enough to support future technological advancements. The companies that solve this challenge effectively will help define the next generation of digital
infrastructure.

Every industry transformation creates a new standard. AI is creating that standard for digital infrastructure now. The future data centre will need to be more adaptable, more efficient, and more future-ready than what we have known before. It will also need to be delivered with confidence. Customers are looking for partners who understand that infrastructure is not just about capacity. It is about readiness, resilience, and the ability to support what comes next.

I have spent considerable time thinking about what this moment means for an infrastructure company like ours. AI is not a passing trend that we accommodate at the edges. It is reshaping the core of what we build and how we build it. The companies that take this seriously early — that redesign their approach rather than retrofit it — will be the ones that earn long-term customer trust. That is the real race.

By Prasanna Sarambale, CEO, Sterling and Wilson Data Center

Reference:
ï‚· https://www.bcg.com/publications/2025/breaking-barriers-data-center-growth
ï‚· https://www.brownadvisory.com/us/insights/data-center-balancing-act-powering-sustainable-ai-growth

  • Defect Liability Period (DLP) [12 months] – bank guarantee, incident management, and transition to O&M.

  • Operations & Maintenance (O&M) [36-60 months] – Resource allocation, structured L1-L3 incident management, and long-term infrastructure optimization.
  • Contract Award – LoA issuance, detailed contract signing, branch/subsidiary formation, BG submission, and insurance processing.

  • Detailed Design & Engineering – Design validation, detailed engineering, software-based modeling, client approvals, regulatory compliance, value engineering & product optimization.

  • Procurement – Vendor evaluation, technical compliance checks, competitive global sourcing, vendor selection via reverse auction, and commercial negotiations.

  • Construction (Site Infra Build) – Civil work, substructure, superstructure, facades, fenestrations, and interiors.

  • MEP Implementation – Electrical, HVAC, ELV, fire fighting, plumbing, structured cabling.

  • Testing & Commissioning – Third-party agency appointment, script development, sequence of operation (SOO), L1-L5 testing, ISAT, and RFS (Ready for Service).
  • Marketing Activities [3 months] – Market identification, customer segmentation, enquiry sourcing, networking & events, referrals, and media outreach.

  • Pre-Bidding [1 month] – Client due diligence, commercial & financial terms, technical scope identification, risk analysis, migration strategy, PQ activities (client registration), go/no-go decision.

  • Bidding [1-2 months] – Contract evaluation, technical & execution strategy, value engineering, sourcing plans, project planning, pricing & bid proposal preparation.

  • Contract Negotiation [2 months] – Finalizing technical, commercial, and execution terms, contract duration, and client expectations.