Evolving Into What? The Rise of the Virtual Building Whisperer

Monday Live! | October 26, 2026 | Evolving Skill Sets: Evolving Into What?

October’s Monday Live! discussions have explored the changing skills behind smarter buildings. This week’s conversation challenged a deeper assumption: we keep talking about evolving, but do we actually know what we are evolving into?

The answer may lie less in acquiring new skills and more in capturing the expertise we already have, making it trustworthy, and putting it to work.

A recent article on turning AI into a revenue stream reinforced a basic reality: AI needs something valuable to work with. In building automation, that means reliable data and decades of operational knowledge, much of it still locked inside people’s heads.

The question is how to preserve that intelligence before it walks out the door.

Learning Without Breaking Things

What if a 48-page equipment manual became an interactive training simulator?

That is already being explored. One AutomatedBuildings contributing editor is using AI to transform technical documentation into a working simulation, with subject-matter experts validating the results. The process is documented in From a 48-Page Manual to a Tested Simulator in One Evening.

The opportunity goes beyond making manuals more engaging.

Pilots learn in flight simulators. Cybersecurity professionals practice decisions through tabletop exercises. Why shouldn’t building operators experience the consequences of a poor control decision before it affects an occupied building?

Imagine adjusting a chiller plant, watching efficiency collapse, diagnosing the mistake, and trying again. No damaged equipment. No uncomfortable tenants. No emergency call.

The challenge is accuracy. A simulator that teaches the wrong lesson is worse than no simulator at all. Human expertise remains essential to validating what AI produces.


AI Has to Earn Its Energy

Fresh observations from Greenbuild brought the conversation back to value.

Buildings generate enormous amounts of data, but collecting more information is not the same as creating better outcomes. Every metric needs a purpose, and increasingly, an economic justification.

One phrase captured the mood: AI has to earn its energy.

A live exercise at Greenbuild turned attendees into a human knowledge graph, demonstrating how relationships between information can reveal more than isolated data points. The discussion also highlighted LEED v5’s emphasis on measured post-occupancy performance.

But better-connected data does not automatically produce trust.

When asked whether they trusted AI in building operations, participants expressed substantial hesitation. Most wanted continued human oversight.

Technicians raised more practical questions:

  • Who wrote this document?
  • Who trained the model?
  • Where did that value of 72 come from?
  • Can we trace the answer to a reliable source?

These are questions of provenance. AI can only be as dependable as the information, assumptions, and rules behind its decisions.

It may eventually help identify conflicting documents, missing records, and questionable sensor values. But people must first establish what constitutes a trustworthy answer.

That connects directly to September’s Monday Live! theme, trust in autonomous buildings.

The uncomfortable truth is that much of the knowledge we want AI to learn has never been properly documented.


The Building Whisperer Is Retiring

Every experienced operations team knows someone who can walk into a mechanical room, hear a change in equipment behavior, and understand what is wrong.

These are the building whisperers.

They carry decades of operational experience, equipment history, practical judgment, and an intuitive understanding of how systems interact.

In many ways, they are already running sophisticated simulations in their heads.

But they are retiring.

Last week’s discussion of the Great Generational Handoff explored why simply transferring responsibilities to younger workers will not preserve that knowledge.

Buildings have become too complex, documentation too inconsistent, and experienced personnel too scarce.

A building can be viewed as an extraordinarily complex robot: interconnected sensors, actuators, equipment, software, and human decisions operating together.

Meanwhile, intelligence is moving deeper into devices and controllers. Tasks once requiring extensive equipment knowledge may now begin with scanning a QR code.

That makes some work easier. It does not eliminate the need to understand what the system is doing.

Nor can we assume experienced professionals will always be available to approve every AI recommendation.

The future cannot depend on having a building whisperer review every decision.


The Data Chasm Beneath Our Buildings

Before AI can help manage buildings intelligently, it needs to know what is actually there.

That sounds obvious until someone checks the records.

A property may list 53 units when the actual count is 50 or 56. Equipment documentation may be outdated. Maintenance histories may be incomplete. Critical knowledge may exist only in an operator’s memory.

Property transactions expose the problem particularly well.

Financial records, leases, and revenue receive intense scrutiny. The physical condition and operational history of the building may receive far less attention. Buyers and sellers do not always have incentives to uncover every hidden issue.

This is the data chasm separating today’s building operations from tomorrow’s autonomous systems.

Collecting more data alone will not bridge it. The information must be organized, connected, verified, and maintained.


Digital Building Profiles: A Practical Bridge

One promising direction is the digital building profile.

Rather than treating a building as one enormous, undocumented system, break it into identifiable assemblies and components: walls, windows, doors, roofs, mechanical equipment, and controls.

Each assembly carries information that matters throughout its life:

Who manufactured it? Who maintains it? Who pays when it fails? What is covered under warranty? When does that coverage expire?

Those assemblies connect into systems, creating a more useful picture of the asset.

This builds on the earlier October discussion of composable buildings, where standardized building blocks could make complex systems easier to describe, connect, and manage.

The technical possibility is compelling. The business incentive is harder.

A renovated lobby is easy to see and market. A reliable digital asset record is less visible, even when it may deliver greater long-term value.

Not every owner faces the same pressures. Institutional facilities, laboratories, and other complex properties may have stronger reasons to invest in reliable building information than owners focused primarily on near-term transactions.

The practical advice was to retain available information, use AI to help organize and reconcile it, and prioritize according to financial and operational risk.

A failing roof membrane above a major grocery tenant may deserve attention long before dozens of less consequential data gaps.

The objective is not perfect data everywhere. It is reliable information where decisions carry real consequences.

From Reactive Operations to Proactive Intelligence

The conversation ultimately returned to governance.

Autonomy is not a software feature that can simply be switched on. It is a progression built on trustworthy information, defined responsibilities, and systems capable of operating within clear boundaries.

Some people hesitate to make operational changes because they fear breaking something. Others rely on experienced intuition to intervene when things go wrong.

Neither approach scales particularly well.

A more mature model would combine structured building knowledge, simulation, AI-assisted analysis, and human accountability.

Operators would spend less time hunting for missing information and more time making informed decisions.

Maintenance could shift from responding to failures toward anticipating them. Asset managers could connect equipment condition to financial exposure. New technicians could learn from scenarios built around real operational experience.

That is a more meaningful evolution than simply adding AI tools to existing workflows.


The Virtual Building Whisperer

The most interesting idea of the session was also its simplest.

What if we could capture the knowledge of an experienced building operator and make it available long after that person retires?

Not as a collection of PDFs or an unreliable chatbot, but as a continuously improving body of verified building knowledge, connected to equipment, operational history, simulation, and decision-making.

A virtual building whisperer.

It would not replace human judgment. It would preserve and extend the expertise that makes good judgment possible.

Getting there requires provenance, structured data, simulation, governance, and a business model that rewards maintaining reliable information.

The skills conversation, then, is not simply about what tomorrow’s technicians must learn.

It is about what the industry must stop losing.

For another perspective on the professionals needed to make this transition, see Wanted: Cyber-Physical Engineers/Architects.

Coming Up

An C4SB.org town hall is scheduled for November 5 at 3:00 p.m. Central, focusing on cybersecurity, AI, and the digital building profile working group.

Planning is also underway for the AHR 2027 Chicago education sessions. Explore the early session preview.

Next Monday begins a new Monday Live! theme.


Play

What Is Monday Live!?

Monday Live! is a free weekly industry roundtable, held Mondays at 3 p.m. ET, where building automation professionals explore the technologies, challenges, and ideas shaping smarter buildings.

Find participant profiles, presentation decks, and past sessions at mondaylive.org.


Further Reading

October’s Monday Live! Series

Related Reading

LinkedIn
Twitter
Pinterest
Facebook