DATA & COLLABORATION · ANALYSIS
From Real Estate Data to Real Estate Intelligence
What MLS organizations need before AI can make listing information more useful: reliable sources, shared standards and clear permission.
Access is the starting point
A listing database can tell a professional what is available. The harder work is deciding what that information means for a particular client: which properties deserve attention, what a comparison leaves out and what needs to be checked locally.
The interview behind this article explores AI as an additional tool for that work. Summarizing records and finding patterns could save time, but a useful summary still needs a dependable source and professional review. Faster output is not evidence that a conclusion is correct.
Keep the source in view
Before adding an analytical tool, an MLS can ask familiar questions. When was the listing updated? Which organization supplied it? How does a correction reach the people using the information? Can the result be traced back to the underlying record?
These checks matter when information moves between systems. An old price or a missing qualification should not become more convincing simply because a tool presents it fluently. Data quality work remains part of the service, not a step that ends when a feed connects.
Give systems a common language
The RESO Data Dictionary defines shared resources, fields and values. The RESO Web API provides a standardized way to transport real estate data. Together, they can reduce the translation work needed when different systems exchange information.
Standards do not supply the listings or grant permission to use them. RESO directs data recipients to the relevant MLS and its licensing arrangements. A technically compatible connection still needs an agreed purpose, authorized recipients and appropriate use conditions.
Cross borders without losing local context
A client considering another country needs more than a longer property list. Local market practices, the meaning of a field and the limits of the available information still matter. Shared definitions help systems communicate; they do not make every market identical.
GDX describes its model as a federated exchange of authorized public listing data, with source organizations retaining control over their rules and permissions. That is its stated operating model, not a claim that every feed is active or that GDX supplies the AI analysis discussed in the interview.
Agree on the use before expanding access
A public listing is not, by itself, a blanket authorization to redistribute every photograph or description, or to use the material for AI training. Those uses need to be assessed against the applicable rights, licenses and agreements.
For a proposed connection, the practical questions are specific: which fields may travel, who may receive them, what attribution must remain and how changes or withdrawals will be handled. These are operational checks for review with the source organization, not legal advice about a particular dataset.
Start with one useful connection
A practical starting point is one authorized use case with a clear benefit for brokers or agents. Check the source records, document the permitted use, test the connection and decide how people will report errors. Evaluate the results before widening access.
That is the opportunity considered here: better ways to use reliable information while keeping its source and limitations visible. AI may help with interpretation. It does not remove the need for local knowledge, accountable decisions or human relationships.
The conversation behind this article
This analysis develops themes from GDX Talks with Jeremy Crawford of FMLS, hosted by Natalie Ghazaley. It is an editorial adaptation, not a transcript or an article authored by the guest or host. The source interview was published on 20 August 2026.
Sources and factual qualifications were reviewed on 2026-09-28. Related readings: article index · article sitemap.