How homebuyers are using AI to speed up the hunt

Austin Ledet and his fiancée closed on a Northwest Washington, D.C., condo in April, a two-level duplex in a rowhouse. But instead of relying exclusively on his real estate agent during his house hunt, he turned to artificial intelligence apps to evaluate neighborhoods and compare homes on the market.

After finding one home he thought was perfect, Ledet asked the AI app Claude to analyze the inspection report. It told him that he would need to make $80,000 to $120,000 in repairs. While his Compass Real Estate agent Traci Johnson thought the AI was estimate was high, she agreed with some of the analysis, and that ultimately persuaded Ledet to walk away from the home.

“I think the human in the loop is arguably the most important aspect,” said Ledet, an IT consultant. “AI is helpful for the quantitative aspects of home buying, but for the qualitative stuff you have a real estate agent who knows the area well.”

AI is rapidly changing home buying, with buyers and agents using it to research homes, analyze inspections, visualize renovations, and streamline communication.

For example, buyers might give ChatGPT a laundry list of requests, from number of bedrooms to yard size to specific neighborhood, to see what’s on the market that matches their criteria. In addition to ChatGPT and other general AI apps, specialized home-buying AI programs are entering the field. Homes AI by Homes.com bills itself as “your consultative home shopping guide.” Buyers can assess homes side by side so they can, for example, compare kitchen photos of several houses. They can also answer more specific questions than the general AI apps, such as “What three-bedroom houses are for sale near a Starbucks?” or “Will my king-size bed fit in the primary bedroom?”

Real estate company the Agency DC Metro has HouseMe.ai, a program that can help buyers narrow a search with information on comparable sales, schools, and neighborhood vibe. It promotes its plain English search by instructing users to “ask the question the way you’d ask a friend who knows the market — and get a useful answer in seconds.”

In July, a national survey by Bright MLS of people who plan to purchase a home in the next one to two years found that nearly 40% of respondents said they would be fully comfortable using AI alone to search for a property, and about 37% said they feel comfortable using AI for financial calculations, such as an affordability analysis. Many said they are OK with using AI but also want a human to review the AI report.

Johnson says AI research can change the dynamic with clients because they are more informed, ask more sophisticated questions, and expect real estate agents to understand AI as well. Johnson said she uses AI before a client signs with her, first recording their conversations (with permission, of course) and then using Claude to summarize the discussion and draft an email.

“This helps clients feel heard and nothing gets forgotten, while buyers receive a concise summary of priorities,” she said.

Early in the home-buying process, AI can help identify the trade-offs buyers might need to identify between their budgets and their dream homes, Fairfax County, Va., real estate agent Sherry Rahnama said.

“It encourages them to think of their future needs rather than just their wants. And then it can model different scenarios for them to show them what’s possible within their budget,” she said. “So, they may need a three-bedroom home for a growing family, but they want a finished basement as well as a pool, and AI can help them sort through this.”

McLean, Va., Remax agent Tony Cammarota uses AI to organize buyers’ must-haves and filter out unsuitable listings. He says it also saves money and time on his end, as AI can create appealing brochures, draft emails, and handle some other marketing tasks.

Virtual staging is another popular application. An agent can photograph an empty room and use AI to add furniture, rugs, and artwork, helping prospective buyers visualize what the space could look like. Some programs allow buyers to place photos of their own furniture in the rooms to help them imagine what it might be like to live there. AI can also show what a house in need of renovation could look like after improvements.

“Most people find homes on the internet first before they visit, so we’re able to easily put homes in their best light,” he said.

Unsurprisingly, AI has pitfalls, agents said.

“People aren’t really verifying facts. They use it as an all-knowing omniscient source, and it’s not. They are taking everything AI spits out as if it’s golden,” Johnson said. “Market value is what a consumer will pay for it. When talking about comps, the identical home next door may have upgrades and updates, whereas the one on the market does not, but AI often doesn’t distinguish.”

Rahnama noted that AI can provide only information that’s already on the web.

“I get a lot of calls from people who are looking for a particular listing, say 123 Main St. Then when I go and look, I see that site that they found it on was never updated. So it’s really not even available,” she said.

She acknowledged that AI capabilities are growing at a lightning pace. “I still think that we are far from an agent being replaced. Buying a home is one of the biggest financial and emotional decisions someone’s going to make,” she said.

“AI can definitely help make it faster, gather the information, and save some time,” she said. “It can analyze data, but it can’t replace the experience, the negotiation skills, the local market knowledge, things like that. You still need those trusted advisers who will continue to play a critical role in helping you navigate the process, and those are things AI can’t do.”