Real estate broker Ryan Serhant saved a $50 million [1] New York City penthouse sale after a ChatGPT prompt nearly derailed the transaction.
The incident underscores the tension between the adoption of generative AI and the necessity of human oversight in high-value financial agreements. While AI tools can streamline communication, inaccuracies in their output can erode buyer confidence and jeopardize multi-million dollar deals.
Serhant said a prompt generated by ChatGPT produced inaccurate information that threatened the stability of the sale. The broker intervened to correct the misinformation and preserve the deal, which was valued at $50 million [1].
According to Serhant, the experience illustrates that AI cannot yet replace human judgment in high-stakes transactions. The ability to navigate the nuance of a luxury real estate deal requires a level of precision and relationship management that current AI models lack, especially when the cost of a mistake is measured in millions of dollars.
This event occurred earlier this week, with details emerging in reports published on June 16, 2026 [1]. The situation serves as a cautionary tale for professionals in the luxury sector who are increasingly integrating AI into their workflows to manage client interactions, and property descriptions.
Serhant's intervention prevented the buyer from walking away from the New York City property. The broker said that while AI is a powerful tool for efficiency, the final verification of facts must remain a human responsibility to ensure the integrity of the contract.
“A ChatGPT prompt nearly derailed a $50 million New York City penthouse sale.”
This incident highlights a critical 'trust gap' in the deployment of generative AI within professional services. In luxury real estate, where precision is paramount, the tendency of AI to produce plausible but incorrect information—known as hallucination—creates a liability risk. As high-net-worth individuals increasingly interact with AI-augmented agents, the role of the human professional is shifting from a primary information provider to a critical auditor of AI-generated content.
