Designing Buildings Is Only the Beginning — AI Is Transforming How They Perform
When a new building is handed over from the contractor to the owner, it often marks the end of a long journey — years of planning, engineering and construction culminating in a completed asset.
But for asset owners and operators, this moment marks the beginning of something else entirely.
From Day Two, buildings must perform exactly as they were designed: to be efficient, reliable and sustainable, not just in year one, but in year five, ten and beyond. Systems must cool, ventilate, light and secure spaces every day, often for decades. Ensuring that buildings continue to deliver the efficiency they were designed for is where the true value of infrastructure lies.
This is where artificial intelligence (AI) is emerging as the next frontier in building operations.
Modern buildings already run on sophisticated building management systems (BMS) that control utilities, water, lighting, air-conditioning and thermal comfort. Yet these systems traditionally operate within fixed parameters and rely heavily on human intervention to monitor performance.
AI changes this dynamic by layering intelligence on top of existing BMS infrastructure. Instead of replacing systems, AI aggregates and analyses data across them — turning operational data into insights that help operators optimise building performance continuously across the entire lifecycle of an asset.
For asset owners, this matters because the majority of value in assets is realised after construction. More than 60% of a building’s lifecycle cost occurs during operations and maintenance, making operational optimisation one of the most powerful levers for protecting long-term asset value.
The SJ Digital Experience Centre at SJ Campus showcases how digital ecosystems can drive sustainability, efficiency and long-term performance. The Centre integrates SJ’s most advanced tools and platforms, including digital twins, AI-driven analytics, virtual and augmented reality to unlock value across the entire asset lifecycle.
Yet despite the growing interest around AI, adoption across the built environment remains relatively early. Industry surveys by Deloitte and McKinsey suggest that only around 30–40% of built environment companies globally are actively deploying AI solutions, with many still in pilot or experimentation stages rather than full-scale deployment.
Where we are seeing the most practical application of AI today is in operations and maintenance, particularly when layered on top of existing building systems.
Take air-conditioning as an example. In many buildings, cooling systems account for 70–80% of operating energy consumption. Even buildings that were originally designed to be energy efficient tend to experience operational drift over time.
Maintenance cycles may be skipped due to budget pressures. Temperature settings are frequently adjusted because occupants feel spaces are too hot or too cold. Gradually, these small operational changes compound, and energy consumption begins creeping upward until the system is no longer operating efficiently.
AI helps set things right.
By analysing operational data continuously, AI optimisation tools can detect inefficiencies and recalibrate systems automatically, restoring performance and ensuring buildings continue to deliver the efficiency they were designed for. In doing so, AI helps protect asset value and safeguard return on investment for owners over the long term.
It also mitigates growing operational risks. Facilities managers today face labour shortages, rising utility costs and increasingly complex building systems. AI can automate routine monitoring tasks and correct operational lapses, allowing human teams to focus on higher-value activities such as investigating anomalies or managing system performance across large portfolios of buildings.
Perhaps most importantly, AI is also enabling a new approach to facilities management: outcome-based operations.
When data from sensors, surveillance systems and building platforms is aggregated into a common environment, AI begins to connect patterns across systems. For example, an indoor air quality sensor detecting odour levels in washrooms could also signal rising CO₂ levels that help diagnose ventilation issues — or even help validate whether a triggered fire alarm is legitimate.
When infrastructure systems begin communicating with one another in this way, buildings move beyond traditional task-based operations toward a more dynamic, data-driven model of performance management.
And that is where the real transformation begins.