Data centers face escalating risks from cyber-physical systems (CPS) and operational technology (OT) vulnerabilities, which can trigger facility-wide outages rather than traditional breaches. These systems, managing essential functions like power and cooling, are becoming prime targets, with compromises frequently leading to disruptive outages. The reliance on connected operational infrastructure is accelerating due to artificial intelligence (AI)-driven demand, significantly increasing the potential cost of an outage. Facilities that train and run AI models are now strategic assets, making their uptime an economic security and public safety issue. Uninterruptible power supplies (UPS), power distribution units (PDUs), cooling systems, environmental sensors, and building management platforms are routinely connected, often through remote access tools and legacy protocols not designed for modern threats. When this layer is compromised, it typically manifests as an outage, such as unexpected shutdowns, false readings, delayed alarms, or cooling degradation that triggers thermal throttling. Traditional IT security frameworks have not evolved to address this cyber-physical environment, creating a gap in ownership, visibility, and prioritisation where outages are frequently born. The rapid design, construction, and commissioning of data centers, particularly in markets experiencing hypergrowth like India, create conditions where governance and security validation struggle to keep pace. Many new vendors may lack mature secure software development life cycles (SDLC), and speed-driven mistakes, such as misconfigured network segmentation, exposed ports, and unchanged default credentials, become common.