Americans Are Using ChatGPT to Price Their Homes-Why That Could Backfire
Artificial intelligence is rapidly reshaping the U.S. housing market, with buyers and sellers increasingly turning to tools like ChatGPT to determine a home’s value.
But industry experts warn that while these tools can offer quick answers, they may also introduce costly mistakes-especially when users rely on them to make high-stakes pricing decisions.
A recent example shows how AI can actively disrupt real estate deals.
In an Instagram post, titled "ChatGPT Just Below up My $50M Deal," celebrity broker Ryan Serhant described how a $50 million transaction nearly collapsed after both the buyer and seller consulted ChatGPT for advice.
The seller asked whether to accept the agreed price, and was told the home was worth more. Meanwhile, the buyer asked if he was overpaying and was told the opposite-that the price was too high.
Both parties walked away from the deal's logic, threatening months of negotiations.
Serhant explained the core problem: "AI can model a market. It can't model a deal."
That distinction gets to the heart of why AI-driven pricing, particularly in high-value transactions, can horribly backfire.
When AI Advice Collides With Reality
The rise of generative AI has made it easier than ever for consumers to access housing data, from comparable sales to neighborhood trends. But unlike traditional valuation tools or professional agents, chatbots generate responses based on patterns in text-not real-time understanding of a property or transaction.
Kamini Lane, the CEO of Coldwell Banker Realty, told CNBC that this can lead to systematically flawed pricing guidance.
"Artificial intelligence is trained to be sycophantic…[it's] more likely to give you the price that you want versus the price at which a home is going to sell for," Lane said.
That tendency can reinforce biases on both sides of a deal. A seller hoping to maximize value may get reassurance that their home is worth more, while a buyer seeking leverage might be told they are overpaying, leading to incompatible expectations.
In some cases, that mismatch can derail negotiations entirely.
A Real-World Example of the Risks
Beyond near-collapse scenarios, AI use has already been linked to financial losses.
In Florida, homeowner Robert Levine used ChatGPT to guide nearly every step of selling his home, including pricing and strategy. The property attracted multiple offers and ultimately sold for more than some agents had initially suggested.
However, real estate professionals who later analyzed the transaction argued that the AI-led approach failed at a critical stage: Negotiating the best outcome. According to one assessment, the seller may have missed the opportunity to fully capitalize on a bidding war, potentially costing Levine between $75,000 and $225,000 in additional value.
The episode highlights a key limitation of AI: While it can certainly help generate interest, it cannot guide the complex, human element of deal-making.
The Missing Context Problem
Industry leaders consistently stress that housing markets are shaped by far more than data points. Local knowledge, buyer psychology, and even informal "off-market" information play a significant role in determining price.
"One of the most important things that agents can see…is what's up and coming," Lane told CNBC, pointing to neighborhood trends and emerging design features. "That is something that no AI tool is ever going to be able to aggregate in the same way."
Similarly, analysts note that general-purpose AI tools often miss key details such as renovations, micromarket dynamics, and the motivations of buyers and sellers.
Without that context, even well-intentioned advice can prove misleading.
Why Mispricing Matters
If a home is priced too high based on overly optimistic AI guidance, it may linger on the market, leading to price cuts and reduced buyer interest. If priced too low, sellers risk leaving significant money on the table.
"AI is more likely to give you the price that you want…and the risk for that is time when you put your home on the market at a price that it's not going to sell for," Lane told CNBC.
AI Is a Tool, Not a Decision-Maker
Despite the risks, real estate professionals are not dismissing AI outright. Instead, they emphasize that it should be used as a starting point rather than a final authority.
AI can synthesize large amounts of data quickly, helping buyers and sellers understand the basic market conditions. But experts argue that it lacks the nuance needed to navigate one of the most complex financial transactions most people will ever make.
The broader lesson from both the Serhant deal and the Florida sale is the same: AI can augment decision-making, but it cannot replace human judgment.
As more Americans turn to ChatGPT for answers, that distinction may become increasingly important, especially when hundreds of thousands, or even millions, of dollars are at stake.
Contact Newsweek editors on this story: Ben Kelly and Sam Wilson.
2026 NEWSWEEK DIGITAL LLC.
This story was originally published June 22, 2026 at 6:32 AM.