Fake video walk-throughs, a loft that seems to grow between shots, and staircases that abruptly end are turning up in property listings as artificial intelligence tools get put to work by agents and brokers.
Imagine scrolling through listings for a house in Franklin, Tennessee, and tapping a vertical clip that looks like a glossy social-post tour. Rooms appear spacious, a four-poster bed occupies a staged master, bottles line a lavish wine cellar, and a freestanding tub sits beneath soft lighting. A smiling agent appears in the corner, narrating in a calm voice as an apparently steady camera pans through each room. It feels flawless. The reality: the house was empty, the furniture and finishes were virtually inserted, the agent’s voice and facial expressions came from text prompts, and no actual video camera captured the footage.
That sort of content is now easy to produce, according to Alok Gupta, a former product manager at Facebook and a software engineer at Snapchat who cofounded AutoReel, an app that converts listing photos into motion clips. “exactly that, at home, in minutes,” Gupta says. He adds that AutoReel users are creating between 500 and 1,000 new listing videos every day, with agents across the United States and in markets like New Zealand and India turning still images into short films to promote properties.
Tools such as AutoReel sit alongside familiar generative systems like OpenAI’s ChatGPT and Google’s Gemini, and together they are changing routines in a sector where visual presentation matters a great deal. Dan Weisman, director of innovation strategy at the National Association of Realtors, reports wide uptake among professionals. “I’ve been at a few conferences over the past few weeks, and just anecdotally speaking, we’ll ask out of 100 people in the audience how many are using AI, and I’d say 80 to 90 percent of people raise their hand,” Weisman says. “We are seeing this huge uptick in people using it.”
That quick acceptance raises questions for buyers and renters who rely on listing media to judge a home’s condition and character. For many consumers, a home search is the single largest purchase decision they will make, and augmented photos or invented scenes can mislead before a prospective buyer even steps inside.
A homeowner in rural Michigan who asked that her surname not be used, and who monitors local listings to follow the market for her own property, recently spotted an example of what the technology can do. “About two or three weeks ago was the first time that I’ve truly seen actual pictures of a house fed through AI,” Elizabeth says. She first noticed an odd yellow cast on images, a tint that has become a common tip-off for algorithmically altered photos to the point that third-party tools now promise to “UnYellow” such images.
“As I was scrolling through the photos, I noticed that some things just weren’t making sense. There were stairways leading to nowhere,” she says. “In general, it just looked cartoonified.” Elizabeth then found an earlier posting for the same property and compared the originals with the edited set. Cabinets vanished in the altered kitchen shots, pavement in the yard turned to grass, and window sizes shifted dramatically. She posted both sets to a popular Reddit thread that catalogues small annoyances; the post drew more than 1,200 comments and a stream of readers calling the edits misleading.
“I think this is misleading. It’s distorting the features of the house,” Elizabeth says. Realtors have long used wide-angle lenses and careful staging to present spaces attractively. AI editing can push those efforts into images that change structural cues or add features that were never there.
Outrage over questionable listing content has appeared across social platforms, including complaints about a New York apartment listing in which a tiny loft was presented as a master bedroom, and a Detroit house whose curb face received a digitally altered roof in its marketing images. Requests for comment from the agent attached to the New York listing went unanswered. A different agent whose phone number was listed on the Detroit ad said the edited images were created by a broker and posted before everyone involved had vetted them; she said she could not confirm whether AI was used.
Some industry figures shrug at secrecy concerns and point to cost savings. Jason Haber, a licensed realtor and cofounder of the American Real Estate Association, argues that automated staging and copywriting replace services that once required days and hundreds of dollars. “Why would I send my photos of an empty room to a virtual stager, have them spend four days and send it back to me at a charge of 500 bucks when I can just do it in ChatGPT for free in 45 seconds?” Haber asks. He adds that virtual renderings have been part of the market for two decades and that the people who once made a living producing those renderings will need to shift roles. “We’ve done virtual renderings for 20 years, so the fact that you can just do it now on AI, there was a whole cottage industry of virtual renderings and those people are now looking for a new job.”
In his role representing more than 22,000 members, Haber stresses that agents should disclose when they use synthetic media the same way they disclose virtual staging. The National Association of Realtors has warned that the legal territory around using AI-created imagery remains “murky,” and the group’s code of ethics bars the use of images that mislead an audience.
Haber says some patterns give away lazy or unvetted use of chatbots. He warns that ChatGPT “almost always” inserts a familiar filler word into listing descriptions, producing phrases such as “in a prime location,” “in the heart of the city,” and “between two other homes” when agents rely on the bot’s default prose. He posted on social platforms that spotting these canned turns of phrase can indicate copy pasted directly from an AI tool. If an agent lets machines write everything, he says bluntly, the result is a homogeneous product. “If you become just a toll taker, you’re not a really good agent. You have no differentiation, you’re not creative, you’re just another agent,” Haber says.
Gupta frames AutoReel’s appeal in business terms. As social platforms dominate attention, short video that holds a viewer’s gaze can be a powerful marketing asset, he says. AutoReel, Gupta claims, can shave “$500 to $1,000” off the price of a production and cut the delivery time from many days to a matter of minutes or hours. “When we started this two years ago, we kind of got a no from customers,” he says. “In 2024, they started saying ‘tell us more.’ And then this year, they’ve been asking How do I get started?”
Not every service provider is sold. Nathan Cool, a real estate photographer with an instructional YouTube channel that approaches 100,000 subscribers, has tested AI-driven tools while keeping a skeptical eye on the outcomes. He says shooting a vertical video alongside standard listing photos remains a quick, low-cost addition to his services, even though some productions can be larger and pricier.
The risk of visual hallucination — where the AI invents objects, fixtures, or spatial relationships — is a constant concern. Gupta says AutoReel is trained on millions of real property videos and that the team refined motion patterns to favor zooms over rotations, a choice meant to reduce the chance the model will generate non-existent elements. In a private test using a set of real listing photos, that approach prevented obvious fabrications. When Gupta uploaded the heavily edited images Elizabeth had found, though, AutoReel inserted a fake couch into a living room.
Cool points to audience fatigue with synthetic clips in social feeds, and to the higher stakes of house hunting. “People that want to buy a house, they’re going to make the largest investment of their lifetime,” he said. “They don’t want to be fooled before they ever arrive.”
Industry rules and local laws around disclosure have not yet settled. Regulators and trade groups are debating whether, when, and how agents must flag AI-altered media, and courts may eventually decide what counts as deceptive advertising. For now, agents juggling new tools, photographers adjusting services, and shoppers sorting fact from fiction are all adapting in real time as these technologies spread.