Robert Levine didn’t hire a listing agent when he decided to sell his South Florida home recently. Instead, he opened a chat window and let artificial intelligence guide the entire process.
NBC 6 reported on March 11 that the married father of three relied on ChatGPT to handle nearly every aspect of selling his home in Cooper City, Florida — from early planning and pricing strategy to marketing, staging decisions, and curb-appeal improvements. And it worked very well.
AI became a kind of digital transaction coordinator for Levine. It helped him create marketing materials, draft his listing description, design an open house handout, and navigate the process of getting the home onto the multiple listing service (MLS). It even suggested the optimal day to go live.
Within the first 72 hours, Levine had five offers. He still held an open house that Saturday. By Sunday morning — just five days after listing — the home was under contract. Even that contract, Levine said, was generated with help from ChatGPT.
Levine did bring in a real estate attorney to review the final paperwork. But beyond that, he said the AI handled most of the heavy lifting. Cost savings were a major motivator. Levine estimates that by using AI, he saved about 3 percent of the total sale price, which was a “meaningful amount of money.”
Still, Levine isn’t predicting the end of real estate agents. Instead, he sees AI as another tool reshaping how homes are bought and sold. And for consumers willing to learn, the barrier to entry may be lower than expected.
‘Time is not on your side’
Levine’s decision to sell his property using ChatGPT instead of a real estate agent has sparked online debate in the industry, but one commentator argues the real disruption isn’t what most people think.
In a recent analysis, real estate strategist Rob Hahn said the viral story is being misread as an example of AI replacing agents. In reality, he argues, the transaction followed a familiar playbook: a for-sale-by-owner deal supported by a flat-fee MLS service.
“This was an assisted-FSBO sale. One happens every single day in the US,” Hahn writes. “NBC 6 News would have been more accurate to title the story, ‘Man uses Beycome’s Enhanced Package to sell his Florida home,’ but that would not have drawn eyeballs and attention. People are reacting because of AI’s involvement.”
Hahn frames the more meaningful shift as the emergence of AI as a functional stand-in for many of the tasks traditionally handled by agents. Tools like ChatGPT can now assist with pricing research, marketing copy, timelines and transaction coordination, effectively acting as a “digital employee” for sellers willing to take a more hands-on approach.
“What the Levine property example tells us is that an MLS-only company + AI + real estate lawyer is all you need today to sell a house,” Hahn writes.
But his sharpest critique is reserved for the MLS itself. As more listings flow through third-party platforms and syndication channels, he argues the MLS is increasingly functioning as little more than a distribution pipeline.
“If the MLS becomes simply a gateway to syndication as the Beycome website suggests, then quite frankly, its days are truly numbered,” Hahn writes.
While Hahn stops short of predicting the immediate disappearance of real estate agents, he suggests the timeline for significant change may be shorter than many in the industry expect. “Let me end by reiterating that time is not on your side,” Hahn concludes. “You don’t have years to figure something out. You have months.”
