AI is moving into real estate through the front door, the loading dock, and the mechanical room. The meaningful change is not another chatbot added to a property management platform. It is the redesign of complete workflows, from a leasing inquiry to move-in, from a maintenance request to resolution, and from a construction issue to a documented decision.
McKinsey estimates that AI and automation could unlock as much as $550 billion in annual value across the real estate value chain. The opportunity is significant, but the number is not a guarantee. It depends on whether owners and operators rebuild the way work moves through their organizations.
For Bay Area property owners, developers, facility directors, contractors, and investors, the question is no longer whether AI belongs in real estate. The question is where it can improve cost control, schedule reliability, building performance, and service without removing the judgment that complex properties require.
The real opportunity is workflow redesign
Alex Wolkomir, Ankit Kapoor, and Vaibhav Gujral of McKinsey describe the next phase of AI adoption as a shift from isolated tools to domain-level transformation. A domain is a connected area of work with a clear owner, a defined process, and measurable outcomes. Leasing, asset management, construction administration, and maintenance are all examples.
That distinction matters.
A standalone AI tool might summarize a lease, draft an email, or classify a maintenance ticket. Those tasks may save time. But the larger gain comes when AI connects the entire chain:
- A tenant submits a request.
- The system identifies the issue and its urgency.
- It checks the asset history and warranty information.
- It assigns the right vendor or building engineer.
- It schedules access.
- It updates the tenant.
- It confirms completion.
- It logs the result for future planning.
That is not a single automated task. It is a redesigned operating process.
McKinsey’s real estate research identifies leasing and asset management as high-potential domains for this kind of change. The same thinking applies to construction and building operations. McKinsey’s real estate analysis frames the opportunity around productivity, revenue improvement, and more informed decisions.
AI can support every stage of a construction project
Construction projects produce large volumes of information. Drawings, specifications, RFIs, submittals, schedules, invoices, inspection reports, change orders, meeting notes, and field photographs all carry pieces of the project story.
The problem is that those pieces often live in separate systems. A project manager may spend hours locating the current drawing, confirming the status of an RFI, or matching an invoice to a budget line. AI can help organize and connect that information.
Useful applications include:
- Schedule risk identification. AI can review project data and flag activities, approvals, or procurement items that threaten the schedule.
- RFI and submittal coordination. Systems can classify open items, identify overdue responses, summarize related documents, and route decisions to the correct reviewer.
- Cost tracking. AI can compare committed costs, invoices, allowances, and approved changes against the project budget.
- Progress documentation. Field photographs and reports can be organized by location, date, trade, and project phase.
- Scenario planning. Project teams can compare sequencing options when labor, materials, inspections, or design decisions change.
Construction technology providers such as Autodesk describe AI applications that connect risk, documents, schedules, and field information. The tool is only one part of the equation. The project team still owns the decision.
AI can identify a pattern. It cannot walk the site, understand a subcontractor’s constraint, or judge whether a proposed recovery plan is practical under actual Bay Area conditions. That is where experienced project management remains essential.
Maintenance ticketing becomes a service workflow
Maintenance is one of the clearest places to apply workflow-based AI because the process is repetitive, measurable, and tied directly to tenant experience.
A conventional ticketing process may involve several manual handoffs. Someone receives the request, determines what it means, checks availability, finds a vendor, contacts the tenant, and later closes the ticket. Every handoff creates an opportunity for delay or incomplete information.
An AI-supported process can assist with:
- Categorizing requests by trade and urgency.
- Recognizing possible life-safety or water-intrusion concerns.
- Reviewing equipment history before dispatch.
- Checking service-level agreement requirements.
- Grouping recurring issues by building or asset.
- Drafting tenant updates.
- Escalating unresolved work.
- Creating reports for owners and property managers.
The goal is not to let software decide that a boiler failure is routine. The goal is to help the right person see the problem sooner and act with better information.
For larger properties, AI can also connect maintenance tickets to capital planning. If the same rooftop unit generates repeated service calls, the issue may no longer belong in the repair queue. It may belong in a replacement forecast.
The U.S. Department of Energy’s guidance on Energy Management Information Systems describes how building data, fault detection, diagnostics, and maintenance systems can support better operations. DOE has also reported median energy savings among campaign participants using energy information and fault-detection systems. Read the DOE summary.
Building operations benefit from better signals
AI is particularly useful when a building produces more data than a human team can review manually. Heating and cooling systems, access controls, utility meters, elevators, pumps, lighting systems, and indoor environmental sensors can all produce operating signals.
A properly configured system may identify:
- A fan running outside its normal schedule.
- A temperature pattern that suggests a failing valve.
- A water meter reading that does not match expected occupancy.
- A recurring comfort complaint on one floor.
- An energy spike after a control sequence changed.
- Equipment operating longer than its design schedule.
This is where predictive maintenance and automated fault detection can support a preventive program. The system points to the abnormal condition. A building engineer or contractor investigates, confirms the cause, and decides what action is appropriate.
That human review is not a weakness. It is the control that keeps a data signal from becoming an expensive false alarm.
Owners should also ask how the system handles missing or inaccurate data. A model trained on incomplete asset records can produce confident but incorrect recommendations. Before purchasing a new platform, confirm that the building has current equipment inventories, consistent naming conventions, reliable meter data, and documented service histories.
Asset management moves from reporting to judgment
Asset managers often spend too much time assembling information and too little time interpreting it. AI can reduce the manual work involved in collecting rent rolls, operating statements, capital budgets, lease data, maintenance histories, and market information.
A redesigned asset management workflow might:
- Pull operating data from approved systems.
- Compare actual performance with the budget.
- Identify unusual changes in income or expenses.
- Review open work orders and deferred capital needs.
- Flag lease expirations and renewal decisions.
- Prepare a draft owner report.
- Route exceptions to the asset manager for review.
McKinsey’s discussion of agentic AI places emphasis on systems that can coordinate several steps while keeping people in control. The firm’s operating model analysis describes a future in which AI agents handle data gathering and routine execution while humans retain authority over material decisions.
That division of labor makes sense. An AI system can surface a falling net operating income trend. An experienced asset manager must decide whether the cause is weak leasing, deferred maintenance, a capital project, a vendor problem, or a temporary market condition.
Tenant service still needs a human center
AI can answer routine questions, schedule appointments, send reminders, and provide updates. It can help tenants receive faster responses, especially outside normal business hours.
But tenant service is not just a response-time metric. Residents, restaurant operators, office tenants, and nonprofit organizations need to know that someone understands the effect of a building problem on their work or daily life.
The strongest model combines automation with clear escalation:
- AI handles routine questions and intake.
- Staff review safety, access, payment, legal, and accommodation matters.
- Contractors receive complete work information before arriving.
- Tenants receive a named point of contact for complex issues.
- Every action is logged for accountability.
That model protects service quality while reducing administrative drag.
Data is the foundation, not an afterthought
AI cannot compensate for disorganized records. If one system calls an asset “RTU-03,” another calls it “Rooftop Unit Three,” and the service history uses a third label, the resulting analysis may be incomplete.
A practical data foundation includes:
- A current asset register.
- Standard names for rooms, equipment, units, and locations.
- Clear ownership of source data.
- Access controls based on role.
- Document version control.
- Retention rules for project and tenant information.
- A record of when AI outputs were reviewed or approved.
The NIST AI Risk Management Framework organizes responsible AI practice around four functions: Govern, Map, Measure, and Manage. Real estate organizations can apply that structure without building a large technology department.
Governance means deciding who can use AI and for what purpose. Mapping means understanding the people, systems, and risks involved. Measuring means tracking accuracy and operating results. Managing means correcting problems and changing controls as conditions evolve.
Start with one load-bearing workflow
The worst way to adopt AI is to buy several tools and hope they create a strategy. Start with one workflow and one measurable outcome.
For a property owner, that might be:
- Reduce maintenance ticket triage time.
- Improve preventive maintenance completion.
- Shorten monthly reporting preparation.
- Reduce unresolved RFIs.
- Improve visibility into change-order exposure.
- Identify recurring equipment failures.
- Increase the percentage of tenant requests resolved on the first visit.
Map the current process from intake to completion. Mark every handoff, duplicate entry, approval, and delay. Separate routine work from decisions that require professional judgment. Then test AI in a controlled environment while keeping the existing process available.
The reported gains can be meaningful. McKinsey materials cite potential improvements of 10% to 30% across selected performance measures, along with examples of 60% to 80% reductions in financial reporting time. Those figures describe potential results in specific workflows, not a promise for every property.
The foundation has to carry the load. Clean data, clear ownership, practical controls, and an accountable project team matter more than a flashy demonstration.
The competitive advantage will be internal
Public AI tools are increasingly accessible. The lasting advantage will come from how well an owner organizes its own information and connects it to real operating decisions.
A property’s maintenance history, construction records, tenant service patterns, capital plans, and vendor performance data are difficult for competitors to copy. That internal record can help an organization understand which assets fail, which contractors perform, which building improvements pay back, and where tenant service breaks down.
For Bay Area real estate, that advantage will come from combining technology with local judgment. Buildings have different construction eras, seismic conditions, operating constraints, tenant needs, and regulatory obligations. AI can make the information easier to use. Skilled people still have to decide what to do next.
Atlas Premier Services & Consultants supports owners and project teams through construction management, design-build delivery, commercial construction, and property maintenance. The practical lesson is simple: treat AI like building infrastructure. Plan it, govern it, test it, and connect it to the work people must complete.
Ready to move your project from concept to completion?
Contact Atlas Premier Services and Consultants today.
Atlas Premier Services and Consultants
Strategic Solutions. Trusted Execution.
Lake Merritt Plaza
1999 Harrison Street, 18th Floor
Oakland, CA 94612
Phone: (800) 994-9028
Email: info@atlas-premier.com
Disclaimer: This content is for general informational purposes only and does not constitute legal, financial, engineering, construction, regulatory, or other professional advice. Reading this content does not create a client or contractual relationship with Atlas Premier Services & Consultants. Because every project and property is different, consult qualified professionals regarding your specific circumstances. Atlas Premier Services & Consultants makes no warranties regarding the accuracy or completeness of this information and is not responsible for third-party content or references. Testimonials, examples, and case studies are illustrative only and do not guarantee similar results.