Luxury home prices in San Francisco are climbing rapidly as buyers enriched by the artificial intelligence industry compete for limited inventory, often paying millions above asking prices. Local agents describe the market as rational rather than chaotic, driven by high liquidity and a shortage of available homes.
According to a market report from Sotheby's International Realty, the vast majority of residential transactions in the city exceeded their listed prices during the third quarter. Specifically, 87 percent of homes closed above their asking price. The following month saw the highest average sale price relative to list price of the year, with homes trading at 131 percent of their initial values.
Compass agent Helena Zaludova, who focuses on the north end of the city, stated that the current bidding wars are a logical response to market conditions. She cited "unprecedented liquidity" and an influx of wealthy buyers competing for a constrained supply. The Agency's Laura Pallin agreed, noting that San Francisco faces a persistent inventory issue that is unlikely to change soon.
The shift in buyer behavior has upended traditional pricing strategies. Compass agents Mike Simonsen and Janel Evans-Anderson noted in a September report that while entry-level homes typically face bidding wars, luxury buyers usually have more negotiating power. This fall, that dynamic has reversed. Compass agent Ali Healy facilitated the sale of a Noe Valley home that closed for more than $2 million over its asking price last month. The $9.1 million deal followed an asking price just under $7 million, with the seller accepting an AI buyer's offer only seven hours after the listing went live. Healy said she did not intentionally underprice the home but was surprised by the magnitude of the overbid.
Buyers are increasingly impatient, according to Zaludova, who observed that luxury buyers who waited on the sidelines in 2024 and 2025 are now acting immediately. In one recent transaction, a buyer who had previously lost out on another property told Healy that they did not care if the price reached $8 million or $9 million, knowing they were competing against colleagues in the tech sector.
Brokers are adapting their pricing methods in response. Zaludova relies on detailed data of every sale in her segment and market intelligence from other agents to set accurate list prices. Her approach has yielded mixed results; she recently sold the longtime Pacific Heights home of late conductor Michael Tilson Thomas for exactly her listing price of $9.9 million. However, in the spring, she represented a seller in North of Lake Street who listed a seven-bedroom mansion at $12 million and sold it for $13.5 million about two weeks later.
In the luxury condo market, Compass agent Justin Goldberg prices listings two to three months ahead of expected market movements. He finds that condos near AI office hotspots sell for about asking or 5 percent above. About 30 percent of his sales are to second-home buyers, many of whom are AI investors looking to maintain proximity to their work. Goldberg noted that these buyers are willing to pay premiums now because they anticipate a more aggressive market environment after upcoming initial public offerings.
AI giants Anthropic and OpenAI are preparing for IPOs, which are expected to bring new wealth into the market. A July report from Redfin estimated that potential windfalls to current and former employees of these companies could allow them to buy nearly one-third of all homes in San Francisco. OpenAI CEO Sam Altman has delayed his company's public listing to 2027 amid safety concerns, but Anthropic's IPO date remains undetermined.
Despite the uncertainty surrounding these financial events, agents do not expect a slowdown in residential sales. High construction costs continue to limit new supply, while demand remains strong due to liquidity and market confidence. As Zaludova summarized, if confidence and liquidity persist while supply remains constrained, the current trend is likely to continue.