Tyler Hochman is the founder and CEO of FORE Enterprise, a premier AI solutions architect and a Forbes 30 Under 30 honoree.
AI has entered commercial real estate (CRE) in full force. Nearly 90% of CRE investors, owners and landlords are already integrating some form of AI into their operations. Yet for many firms, the returns have been disappointing—not because AI lacks potential, but because the generic tools being deployed were not built for this industry with the training data needed to succeed.
Most tech companies offering AI tools fall into one of two camps: those that push the frontier by building new models and expect clients to adapt to their specific needs, and those that take existing technology and tailor it to specific workflows and use cases. In commercial real estate, the second approach is the one that actually delivers results.
Where Volume And Time Diverge
Many tech companies are eager to build tools to support CRE needs. But solving the costly, time-consuming problems commercial property managers face daily requires more than a bird’s-eye view of the industry. Tech companies need to understand the industry from the inside out.
Generic tools can handle some of the predictable parts of property management, such as lease renewals, rent collection and maintenance requests. But automating those tasks does not necessarily free up the hours that matter most. The real drain, from my observations, lies in the exception, not the rule: the tenant who stops paying and disputes the default clause, the compliance obligation buried in page 187 of a lease stack, the co-tenancy carve-out nobody flagged at signing. These unusual situations, or what I call “edge cases,” can pull property managers away from their regular work and typically take the most time to resolve.
Without genuine immersion in CRE operations, tech companies build in a vacuum, often unaware of the edge cases that determine where AI can deliver the highest return. A system designed to automate the most common tasks, rather than the most costly ones, will not reduce total workload in any meaningful way. The goal for tech companies should not be automating more. It should be automating smarter and targeting the greatest inefficiencies first.
The Challenge Of Building What You Have Not Lived
Our team developed a suite of AI agents purpose-built for commercial property management. Early on, we recognized that a tool earns its value in the field, not the lab, which is why we partnered with a CRE firm to pilot, pressure-test and refine it against real-world conditions.
We learned that no two commercial real estate documents are alike. Leases shift based on local law, property type, tenant and financing structure. Even a routine transaction can follow entirely different rules depending on the market and the specific terms negotiated in each individual lease.
A model trained on generic data handles predictable cases well enough. But the moment it encounters jurisdiction-specific nuances, non-standard clause structures or fact patterns that don’t appear in public datasets, the error rate climbs. And in CRE, errors are costly. The only way to account for that complexity is through robust training data.
Beyond showing us the day-to-day work, our partnership gave us access to the full spectrum of a national portfolio—training data we could not have assembled independently—including the messy, exception-heavy files that do not appear in public datasets. That exposure is the difference between a tool that handles the easy 90% and one that holds up under the other 10%, where the real liability lives.
Why Tangible Outcomes Beat Efficiency Arguments
A common approach for tech companies building AI for CRE is to tackle problems with a clear value gain. Investments with a visible before-and-after—money saved, costs cut, a loss prevented—are easier to measure and therefore easier for potential clients to justify.
Property tax appeals are a compelling example. A property is assessed at a set value, and that assessment determines how much the owner pays in taxes. If an AI tool can identify valid grounds to challenge that assessment and reduce the appraised value, the property owner pays less. The tool can eliminate the third-party intermediaries that typically handle appeals and allow owners and managers to capture much of the upside themselves. The financial impact is immediate and direct.
AI investments targeting inefficiencies can be more difficult to greenlight because the gains are harder to quantify. But they are equally important. Whether the ROI is immediate or gradual, the question tech companies should build around is the same: Which problems are consuming the most time and generating the most unnecessary cost?
Starting With Understanding, Not Automation
Off-the-shelf AI platforms are built for the average use case, which means they are optimized for no one’s use case in particular. Before tech companies can help CRE firms automate anything, they need an understanding of where time is actually going: which tasks consume the most hours and which carry the clearest financial or operational return. Once tech companies understand those priorities, they can tailor their AI solutions around the edge cases where complexity is concentrated. The tool should not be designed around the bulk of the problem. It should be designed around the hardest part.
Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?
