“The Real Promise of Operational AI” – Trust, Accountability and Understanding AI for the Built Environment

In episode 181 of the Asset Champion Podcast, host Mike Petrusky speaks with Daniel Stonecipher, Independent Advisor for AI Infrastructure & Built Environment Platforms, about the growing role of artificial intelligence in facilities, maintenance, and asset management. Drawing on his work advising organizations, investors, and executive teams, Daniel explains why the real promise of operational AI extends far beyond building automation. Instead, he believes AI can help create more capable and confident organizations with stronger operational oversight and decision-making. Together, they explore one of the most important questions facing facility and asset management leaders today: when can AI be trusted? Daniel argues that while AI-powered pilots continue to demonstrate impressive analytical capabilities, organizations must carefully validate results, establish accountability, and ensure human operators maintain appropriate oversight. He explains why confidence leads to trust, trust leads to dependence, and why facilities leaders need to focus less on what AI can do and more on how it improves business outcomes.

Agenda

• Understanding why trust, accountability, and confidence matter in AI adoption
• Balancing AI recommendations with operational context and human expertise
• Moving from successful AI pilots to organization-wide implementation
• Identifying practical steps facility leaders can take when evaluating AI initiatives

What you need to know: Facility and maintenance takeaways

Takeaway 1: AI is most valuable when it improves decision-making, not just analysis

Daniel explains that many organizations are seeing strong results from AI-powered pilots that summarize data, identify patterns, and generate recommendations. However, operational value comes from improving decisions, not simply producing more information.

While AI can help analyze maintenance history, identify asset issues, and process large amounts of operational data, facilities leaders still need to evaluate recommendations within the context of business priorities, building conditions, operational constraints, and risk tolerance.

“The real promise of operational AI is not just smarter buildings by themselves. I think it goes way beyond that. I think it’s more confident, capable operating organizations that have a better understanding and better control over how they operate,” says Daniel.

Takeaway 2: Operational context determines whether AI recommendations can be trusted

A recommendation can be technically correct while still leading to the wrong decision if critical context is missing. Daniel points out that facilities operate in dynamic environments where asset conditions, maintenance history, occupancy patterns, operational priorities, and business objectives continually change.

For example, an AI system may recommend replacing a pump based on sensor data. Before acting, an experienced facility manager would also consider whether the sensor is reliable, whether recent maintenance was performed, whether redundancy exists, and what operational impact could result from taking the asset offline.

Organizations should view AI recommendations as valuable inputs into decision-making, not automatic actions.

“A building doesn’t operate inside a prompt window. AI may understand the pattern, but facilities leaders still need to understand what changed in the building, the asset, the occupancy, or the priorities behind the decision,” says Daniel.

Takeaway 3: Trust develops through testing, validation, and accountability

According to Daniel, trust is not something organizations grant automatically. Instead, it develops through a progression of confidence, trust, and eventually dependence.

Facilities teams must first verify that AI systems consistently produce reliable results using both test data and real-world operating conditions. As confidence increases, organizations become more willing to trust recommendations and integrate AI into workflows.

However, accountability remains with people, not algorithms. Leaders must understand who owns outcomes, who verifies recommendations, and what safeguards exist when decisions affect operations, safety, or compliance.

“The prediction is only one input. The operational decision requires context, and context is where facility professionals create value.”

Maintenance management insights

• AI adoption should focus on measurable business outcomes rather than technology capabilities alone.
• Human expertise remains essential when evaluating operational recommendations generated by AI.
• Facilities leaders should clearly define who is accountable for decisions influenced by AI systems.
• Trust in AI develops through repeatable performance, validation, and real-world operational experience.
• Facilities professionals should continually ask what happens if an AI recommendation is wrong and who owns the outcome.

Do a deep dive into more asset management insights by exploring all Asset Champion Podcast episodes. Watch the full video here: https://www.youtube.com/playlist?list=PLSkmmkVFvM4H3pwnlU2AuqynuRDpvnh4J