Real estate

business the speed and agility it needs to focus on what's important.”

Artificial intelligence is far from new, having come into being about the time Elvis recorded “Blue Suede Shoes.”

Scientists believed in the 1950s that machines could be programmed to think like humans, but computers lacked the power for significant machine learning at the time.

Nevertheless, AI use has been growing steadily in the past three decades. AIpowered chatbots came out in the late 1980s.

Zillow began using AI algorithms in 2006 to create its Zestimates of home values, and virtual assistants like Siri surfaced in 2011.

Until recently, however, AI was limited to techies and coders.

That changed early last year with the release of “general purpose generative AI,” which allows users to give machines instructions via natural language prompts. Now, virtually anyone who can speak English (and many other languages) can program computers to analyze data, write reports or create modified images without having any technical expertise.

“GenAI really changes the game, it really democratizes the access to data in the sense that the user can ask pretty much any question,

(and) the database can answer,” Or Hiltch, chief data and AI architect for commercial real estate brokerage JLL, said during a company webinar in September.

JLL researchers identified at least 500 companies around the world that are developing AI-powered products and services for real estate.

During a demonstration, Hiltch used JLL's “generative pre-trained transformer,” or GPT tool, to convert an Excel spreadsheet into a ranked list of comparable properties for an apartment building. The process took seconds.

AI will help commercial real estate brokers increase their efficiency in the face of rising costs and falling profits, said Laurent Grill, an executive with JLL Spark, an affiliated company.

“The reality is we're living in an industry that's facing a lot of challenges,” Grill said. “It's more about how AI is going to empower us to be better in our job and service our clients.” Industry uses

Like JLL, other companies are embracing AI.

Compass has invested more than $1.5 billion in developing an AI-assisted technology platform its agents can use to search for listings, arrange showings, create brochures, write social media posts and produce newsletters and videos.

“ChatGPT sits on this massive Compass database,” said Mark McLaughlin,

Compass' chief real estate strategist. “We can say, create a property description for 123 Main St. Boom! There it is.”

Agents still must use their knowledge and expertise to perfect the description, but “they don't have to start from scratch,” he said.

“AI and ChatGPT will significantly reduce, if not eliminate, the routine and the mundane,” McLaughlin said. “It will take the place of (repetitive tasks) and give that time back to … reinvest into my client relationships.”

In addition, Compass uses AI to look for homeowners who are likely to sell their homes by analyzing such signs as marriages, job changes or children leaving the nest.

Other new AI products include a tool by Irvinebased Revive that uses photos to assess the costs and benefits of renovating a home before putting it up for sale. In seconds, the tool shows how much a renovation can increase a home's value.

“AI is creating some great opportunities on how we can (make) it more than just numbers on a sheet,” said Dalip Jaggi, Revive Chief Operating Officer. Homeowners “are able to contextualize in their mind a little bit better, going, `Oh, wow. I could see my million-dollar home could be worth $1.4 million if my house looked like that.' ”

In December, Rocket Homes — a home-search

company affiliated with Rocket Mortgage — unveiled a new application that helps home shoppers find homes while they're driving about.

Using the buyers' preferences and a list of homes they already looked at, the program displays listings worth considering on a buyer's Apple CarPlay while they're driving.

“You can continue (home shopping) as you drive,” said Sam Vida, Rocket Homes' president and chief product officer. Cutting costs, calories

AI use is rising among Compass agents, McLaughlin said.

When asked at sales meetings three months ago how many used AI, about a third of the agents raised their hands, McLaughlin said. In February, about 60% of the hands went up.

Florida broker Cyndee Haydon, who teaches agents how to improve their ChatGPT prompts, said she uses the tool to make emails more personal and sincere. She also dropped her email marketing company, saving $7,500 a month, after learning to use AI for that task.

“I told them I did not need their help because I could do that myself,” Haydon said.

Sacramento-area agent Rachel Adams Lee not only uses ChatGPT to write listing descriptions, she even asked it to create a healthy, low-calorie meal plan. She lost 31 pounds.

“You can use it for everything

from weight-loss tips to writing a difficult email to a client,” Adams Lee said.

Estrada, the Beverly Hills agent who uses AI “for everything,” has experimented with applications that virtually furnish, or “stage,” a home for showings. Another application uses AI to show buyers how a tired, dilapidated home might look after remodeling.

“They're still not there yet. I don't think they have a wow factor,” Estrada said of these new applications. But they're getting better.

Nevertheless, AI has its drawbacks.

“When we move so quickly to dependency on those systems, we lose that institutional memory on how to actually execute the things,” said Linsell, the real estate coach. “There's a lot of people who are jumping on this bandwagon and ditching the old ways very quickly.”

Professionals still need to be wary of perpetuating the age-old dangers of redlining and fair housing violations.

“The problem with these tools is there is no internal moral compass. It is just input and output,” Linsell added.

And the accuracy of a technology that once told a New York Times reporter to leave his wife (whom he loved) needs to be doubleand triple-checked before shipping results off to clients.

There's also a danger of relying too heavily on

AI for a “people business” like real estate, Minnesota broker Teresa Boardman wrote last October in Inman, the online real estate news outlet.

“AI is artificial and not at all intelligent. It isn't kind or friendly, or even all that interesting. Customers do not look forward to interacting with it,” Boardman wrote. “A good, experienced real estate agent knows all sorts of things that AI just doesn't get. Like the fuzzy logic that is used to determine how much a house will sell for.”

If agents aren't careful, Estrada added, AI could lead to plagiarism, or worse: housing deepfakes.

One agent posted a photo showing a backyard with a pool and a fire pit without disclosing that those amenities were added using AI.

“You don't want to catfish buyers and sellers,” Estrada said.

However it's used, AI will only get bigger in the future, and it “will change the way that real estate professionals communicate, how properties are marketed, and much more,” predicted Linsell, the real estate coach.

“Task-based AI applications will also likely revolutionize the back-end operations of the real estate market,” he said. “Right now, a person is responsible for inputting and verifying every single measurable (fact) included on a real estate listing. Very soon, that will be done by smart machines.”