The AI boom is, in effect, fueling the data center boom. The widespread adoption of AI at both personal and corporate levels has fundamentally reshaped expectations for digital infrastructure. As demand for computing power surges, hyperscalers and neocloud providers continue to scale data center capacity. In turn, data centers must become more resilient to withstand the operational and strategic risks that lie ahead.
Traditionally, data center resilience has been measured largely by uptime, with five-nines availability serving as a key benchmark. Maintaining this level of service continuity requires robust power redundancy, cooling systems and connectivity offerings to ensure uninterrupted operations.
However, what was considered a resilient data center before the AI boom may no longer be sufficient in an AI-driven world. Resilience is no longer about staying online. It is about ensuring that the infrastructure supporting AI can continue to perform, adapt and scale despite an increasingly complex and uncertain risk environment.
Combining redundancy with strong governance and risk management
Resilience must now be designed in from the outset—not addressed as an operational consideration after construction. It must be embedded throughout the data center life cycle, from site selection, design and construction through commissioning and ongoing operations.
Over time, data centers can become less resilient as operating environments, technologies, infrastructure and risk exposures evolve. New workloads may place additional demands on power and cooling systems, while changes in the external risk environment can introduce vulnerabilities that were not present when a facility was first designed.
The next generation of data center resilience will therefore need to consider not only whether a facility can remain operational, but also how effectively it can anticipate, absorb, adapt to and recover from a broader range of risks.
While data center stakeholders are doing everything they can to keep pace with the exponential growth of AI and cloud computing, the industry continues to face persistent and emerging challenges. Climate-related risks, geopolitical uncertainty, supply chain disruption and labor shortages can all affect the ability of data centers to maintain operations, reinforcing the need for a more proactive approach to resilience.
Assessing resilience through scenario testing and loss simulations
Scenario testing and loss simulations are essential to understanding how data centers would respond to extreme but plausible events. These exercises can expose vulnerabilities that may not be apparent under normal operating conditions, allowing organizations to identify critical dependencies, test response protocols and strengthen preparedness before a disruption occurs.
Operators must continue to conduct regular reviews of their facilities, challenge existing assumptions and reassess their risk exposures. Automation can improve efficiency, monitoring and operational consistency, but it cannot replace human judgment. Maintaining resilience therefore requires continuous testing, learning and adaptation—not simply achieving a resilient design at the point of construction.
Resilience cannot be treated as a one-time exercise. The most resilient data centers will be those that treat resilience as a dynamic capability: one that evolves alongside the technology, operating environment and risks they are built to support.
Our Global Data Centers & Digital Infrastructure Practice (opens a new window), launched in July 2026, combines risk strategy, analytics, engineering insight and market expertise to support clients throughout the asset life cycle and ongoing risk management. For more information, visit our Data Centers (opens a new window) pages or contact us at enquiry.asia@lockton.com (opens a new window).