Every major real estate firm is running AI pilots. Most are discovering the same thing: the model works fine. The data does not.

This is not a surprise to anyone who has spent time inside institutional real estate data environments. The problem was always there. AI did not create it. AI exposed it.

Here is what the problem actually is: your organization has accumulated decades of institutional knowledge. Definitions, hierarchies, relationships between entities. That knowledge lives in spreadsheets, in the heads of long-tenured employees, in the informal conventions that different teams developed independently. It is not in your data.

The meaning problem

Take a simple example. Your asset management team tracks net operating income one way. Your acquisitions team tracks it another. Your fund reporting team has a third definition that aligns with LP reporting conventions. All three are internally consistent. None of them agree with each other.

When an AI model queries your data to answer a question about NOI, which definition does it use? The answer is: whichever one it finds first. And it will not flag the discrepancy. It will produce an answer with full confidence.

That is not an AI problem. It is a meaning problem. The AI is doing exactly what it was designed to do. It is running on data that was never designed to support it.

What solving it actually requires

You do not fix this by buying a better model. You fix it by building a semantic layer: a structured, governed representation of your institutional knowledge that AI systems can read without ambiguity.

This is harder than it sounds. It requires entity resolution across your source systems, a knowledge graph that captures relationships between properties, portfolios, leases, and counterparties, and governance processes that keep the layer current as your data changes.

None of that is glamorous. But it is the prerequisite for every AI use case that actually matters: agent workflows, autonomous reporting, decision support, portfolio analytics. Without it, you are running fast on a foundation that cannot hold the weight.

The firms that get AI right are not the ones who found the best model. They are the ones who built the layer underneath it.

• Tagged as:
• AI
• ai adoption
• AI in Real Estate