When JLL released its report “Real estate’s AI reality check: 90% of companies piloting, only 5% achieved all AI goals,” building operators paid a lot of attention to the delta between the use of artificial intelligence and the return on investment from it.

According to the report, the three ways building operators are using AI are mostly around data-related workflows, portfolio optimization and energy management. 

Notably, the report found that the number of real estate companies running AI pilots went from 5% to 92% in three years. Yet only 5% of survey respondents said they’re realizing gains from their investments.

Given that AI is still in its pilot phase, building operators having a high interest in the technology but realizing low ROI shouldn’t be a surprising result. 

Human interactions are critical

What stood out in the report is the way AI is being piloted. Arguably, one critical data set is missing: the human experience in buildings.

Understanding the human experience is key to ensuring collaborative and productive workplaces. AI workplace data shows the importance of the irreplaceable human connection in the office: having access to physical environments such as labs, handling high security data, engaging with one another in dialog and solving problems in a cross-functional way. None of these can be easily replicated when employees aren’t on-site, working together. 

As it evolves, the workplace must be designed around these critical, high-impact interactions. Doing this requires assessments grounded in data that reflect how people use and experience space.

There is no lack of tools with data showing every facet of building management. And foot traffic, occupancy and other data provide building operators with baseline insights. However, understanding how humans interact in a space is critical to the data foundation that drives AI. This intel comes from sensors — physical AI — that gather data. 

In commercial real estate, physical AI datasets can be captured without compromising individual privacy. This is possible using thermal-based sensors that detect movement and infer actions based on body heat while ensuring anonymity. Ideally, the physical AI data integrates with other facility management and real estate platforms for a comprehensive picture of a property.

Physical AI

Here are two examples of how physical AI is being deployed in the workplace:

Global medical technology manufacturer. The company was considering leasing additional space due to the need for more administrative workstations. Before they made the investment, they installed thermal sensors that combine AI with body heat sensing technology. The goal was to understand how the lab and workspaces were being used.

Without capturing any personal data, the sensors uncovered that the lab was 30% underused. This led to an office redesign without having to incur unnecessary life sciences fit-out costs, which average $846 per square foot, according to data from Cushman and Wakefield. Additionally, the sensors helped the medical technology company identify which instruments were most frequently used, therefore enabling more efficient capital planning.

Global software provider. The company was trying to boost its 15% workstation use rate. A return-to-office mandate only increased office attendance to 45%. When the company looked at the data from the AI sensors, they saw workers reserving large conference rooms for focused work or phones calls and using the cafe for late afternoon meetings. They also saw a lot of coffee badging — people swiping their badges to show they’re in the office but leaving quickly to do most of their work at home.       

The software company reconfigured the office to accommodate the workforce. The new layout provided additional, smaller meeting spaces, flexible and portable pods for private calls and focused work, and more efficient maintenance crew schedules for the café after the traditional lunch hours.

Within two weeks, usage was up to 72%. Along with a boost to productivity, the facilities management team improved efficiencies in operations and energy consumption based on aggregate data from the sensors about human behavior.

AI strategies

This rise of physical AI is redefining how companies manage commercial real estate while transforming buildings from passive environments into active participants in productivity, compliance, sustainability and energy efficiency while supporting smarter decisions about building operations.

The key to realizing ROI from AI investments lies in the quality of the underlying data. This principle is not lost on many facility teams: they’re bringing together complex data sets about nearly every aspect of building management, giving them the right foundation for making AI work. But it’s helpful to see the data sets will be incomplete without insight from physical AI.