HousingWire expands AI strategy to housing data and news

HousingWire is expanding its artificial intelligence strategy by making its housing market data, mortgage information and industry analysis accessible through AI platforms including ChatGPT and Claude.

HousingWire executives previewed the technology on Tuesday during the company’s third annual AI Summit at the George W. Bush Presidential Center in Dallas.

“When we started this event three years ago, the conversation that we had planning content was, how do we weed through all the noise to get to actionable insight?” HousingWire CEO Clayton Collins said.

That task has become easier as companies have moved from discussing potential AI applications to deploying tools in their businesses, Collins said.

HousingWire has spent much of the past year preparing its own data infrastructure for AI applications, including organizing housing market data, mortgage data, mortgage rates, RealTrends Verified rankings, and other information available through the company and its partners.

The effort was necessary to make the data usable both internally and through new products, Collins said.

“Anyone who’s initiated and implemented AI initiatives understands the importance of having data structured the right way,” Collins said.

The company has used the infrastructure to give reporters, editors and podcast hosts faster access to information while also providing its technology team with a foundation for developing new AI-powered products.

The HousingWire Intelligence product, which was in beta mode at the time of the summit, is scheduled for a full release within the next week.

During the summit, Holden Page, HousingWire’s senior vice president of technology, demonstrated how the company’s data can be incorporated into an AI workflow through a model context protocol (MCP), an open standard that lets AI applications like Claude or ChatGPT securely connect to outside data, tools and systems.

“The way that we’re doing that is the model context protocol, which is basically just the way that we describe our data to AI and helps you query it without having to go through a data scientist,” he explained.

In the example presented by Page in front of the audience, a user asked Claude to run a weekly analysis of the Minneapolis-St. Paul market and identify where sellers were positioned. Rather than requiring a user to write structured query language (SQL) or work directly with raw databases, the MCP allows Claude to access HousingWire data and produce an analysis based on the available context.

Claude can then generate an artifact, such as a dashboard, showing where a lending team should focus its efforts. The system also can incorporate HousingWire news and analysis, giving users additional context for interpreting market conditions.

Page said HousingWire plans to make the MCP available through the Claude and ChatGPT stores. A beta version also is available through Snowflake.

The company’s broader goal is to make its content and data available wherever housing professionals work, Collins said.

“We’re going to build for what you need,” Collins said. “We’re going to build for the information that helps you make better decisions faster. …. We’re not building for ourselves. We’re building for our subscribers. We’re building for leaders in mortgage, real estate and homebuilding.”

Collins said the company plans to gather feedback from professionals in the mortgage, real estate and homebuilding spaces as it develops its AI solutions, including through events and direct conversations with industry participants.

“This is not a conference where we had to worry about cutting through the noise of vaporware,” Collins said. “These are people who have actually implemented real AI solutions into their businesses.”

This article was written by Sarah Wolak and generated with the assistance of HousingWire Automation, then reviewed by a HousingWire editor before publication.