San Francisco has had tech booms before. The dot-com era, the IPO wave of 2019, the pandemic run-up that pushed median house prices to $2 million before the correction. Each cycle had its own character and its own fingerprint on the market. The current one is different in ways that show up clearly in the data, and understanding what’s driving it matters whether you are buying, selling, or just trying to make sense of what’s happening around you.
The number that tells the story most directly: in the first half of 2026, roughly 144 San Francisco homes sold for at least $1 million over their asking price, compared with approximately eight in the same period a year earlier. June 2026 alone accounted for 44 of those sales, the most in any single month since Compass began tracking this figure.
Where the Money Is Coming From
Earlier tech wealth cycles were driven primarily by salaries and stock appreciation that played out gradually over years. The current cycle has a different mechanism. The AI sector has generated concentrated bursts of liquidity through secondary share transactions, in which employees sell equity in private companies before a public offering. Reports from major financial outlets suggest hundreds of employees at leading AI companies have participated in secondary transactions worth billions of dollars in aggregate over the past two years.
The significance of secondary liquidity is that it converts paper wealth into purchasing power quickly and in large amounts. A buyer arriving at the market with proceeds from a secondary transaction is operating differently from a buyer whose wealth is still tied up in unvested shares or a stock portfolio they are reluctant to sell. The money is available, the decision is already made, and the only question is where to re-invest it. San Francisco residential real estate has been one of the primary answers.
This wave is distinct from the more gradual wealth accumulation that drove previous cycles. The buyers it produces are not waiting for interest rates to fall and not particularly sensitive to what the market did last month. They are transacting from a position of financial certainty, which changes how they behave in a competitive offer situation.
What the Numbers Look Like
Compass’ June 2026 data makes the picture concrete. The median house sales price came in at $2,128,000, up 24.8% year over year, just 3.3% below May’s $2,200,000, which was an all-time high. June was the second-best month this city has ever recorded. Price per square foot reached $1,230 in June, a new all-time high, up 22.3% year over year and up 28% since January alone. The average home sold at 125.2% of its asking price. June’s 44 homes selling at $1 million or more above asking — combined overbidding of just over $60 million across the city — was the most active month on record for that metric.
The condo market is running its own recovery in parallel. The median condo price reached $1,225,000 in June, up 6% year over year. Closed condo sales hit 314 for the month, up from 123 in February. Condos are now selling in 37 days on average, down 38.3% from 60 days a year ago and the fourth consecutive month of improvement. The gap between the house and condo markets is narrowing in real time.
Savvy Buyers
The buyers I am working with and competing against in this market share a few consistent characteristics. They have done significant research before engaging. They are not in a rush in the general sense, but they move decisively when the right property appears. Many are transacting all-cash or with minimal financing dependency. Some are buying a primary residence they intend to hold long-term rather than a speculative investment.
The AI sector has produced a specific buyer profile that did not exist in the same form in previous cycles: principals, researchers, and early employees at companies that have grown very quickly, who have accessed liquidity through secondary transactions or compensation events and are now treating real estate as the logical next step. These buyers are concentrated in San Francisco and the immediate Bay Area because that is where their companies and networks are. Remote work didn’t change that for this group the way it did for earlier tech workers who dispersed to Austin, Miami, and other markets.
There is also a secondary effect. Other buyers who are not themselves in the AI sector are moving earlier than they otherwise would, anticipating that upcoming IPOs and additional liquidity events will bring more competition into the market. Some buyers are making offers now specifically because they expect the pool to expand further.
The Supply Side of the Equation
The AI demand surge is hitting a market with almost no room to absorb it. San Francisco closed June with 590 active listings across houses and condos combined, against 546 properties already under contract, with an absorption rate near 92% and just over a month of supply. It’s the tightest this market has been heading into summer since 2020.
Homeowners who refinanced or purchased at pandemic-era rates are reluctant to sell into a higher-rate environment, which keeps a significant portion of the housing stock off the market entirely. Mortgage rates spent June near 6.5%, continuing a range that has held for roughly three years. New construction in San Francisco remains constrained by the same geographic and regulatory factors it always has been. The Compass data is also showing a “mansion shortage,” meaning more demand for large, high-end houses than there is supply. That’s pushing buyers toward whatever is available and keeping upward pressure on prices even as the number of transactions remains relatively modest.
SF real estate is in a compressed environment where more buyers are competing for a shrinking pool of available homes. That combination produces the overbidding levels, the accelerated pace of sale, and the frequency of all-cash offers that the June data shows. The demand is real and the supply is truly limited.
The Longevity Question
The honest version of this analysis has to include the risk. The current demand surge is heavily concentrated in a single sector. If AI investment cycles, if anticipated IPOs disappoint, if the secondary transaction market dries up, the specific buyer profile driving this market could pull back. San Francisco has experienced sector-driven corrections before, and anyone who has been in this market long enough has seen what happens when a concentrated source of demand retreats.
The counterargument, which I find more compelling over any reasonable time horizon, is that San Francisco’s supply constraint does not depend on what the AI sector does. The city will not develop significantly more housing regardless of economic conditions. The geographic and regulatory constraints that have kept inventory low for decades are not going away. When demand recedes, prices may adjust at the margin. The floor, however, tends to hold in a way that more buildable markets cannot replicate.
The buyers who have historically done well in San Francisco are the ones who bought with a long horizon and held through volatility. The current moment rewards preparation and decisiveness, but the underlying case for the market does not begin or end with any single economic cycle.
If You Are in the Market Right Now
For buyers, preparation has to precede the search. Full pre-approval is a minimum. Understanding your ceiling, your contingency posture, and your timeline before you are in a live offer situation is not optional in this environment. The buyers losing offers are overwhelmingly the ones making those decisions reactively rather than in advance. The condo segment, where overbidding is present but less extreme than the house market, is still offering more breathing room for buyers who can be flexible on property type.
For sellers, this is the strongest positioning environment in years, and the data supports acting on it. Inventory is near its tightest level since 2020. The buyers in this market are serious and motivated. A well-prepared, accurately priced property in the right condition is drawing the kind of competition that produces outcomes sellers have not seen since the spring of 2022.
If you want to talk through what this market means for your specific situation, I am happy to walk through it. Let’s talk.
Frequently Asked Questions
How is the AI economy affecting San Francisco home prices?
The AI sector has generated significant concentrated liquidity through secondary share transactions and compensation events, producing a wave of buyers who are rate-insensitive, highly capitalized, and motivated to transact. In the first half of 2026, roughly 144 San Francisco homes sold for at least $1 million above asking price, with 44 of those occurring in June alone, the most in any single month on record. The median house price reached $2,128,000 in June, up 24.8% year over year, per Compass data.
Who is buying luxury homes in San Francisco right now?
The primary driver is buyers with AI sector liquidity: principals, early employees, and researchers at leading AI companies who have converted equity into purchasing power through secondary transactions or compensation events. A secondary group consists of buyers moving earlier than planned in anticipation of additional competition from upcoming liquidity events.
Is the San Francisco real estate market in a bubble?
The current market is heavily influenced by a concentrated demand source, which introduces real risk if the AI sector cycles. The structural counterargument is that San Francisco’s supply constraint is geographic and regulatory, meaning the floor for values tends to hold regardless of what drives demand in any given cycle. Long-horizon buyers have historically navigated this dynamic well.
What does AI wealth mean for the San Francisco condo market?
The condo market is participating in the current surge and accelerating. The median condo price reached $1,225,000 in June, up 6% year over year. Condos are selling in 37 days on average, down 38.3% from a year ago, with closed sales more than doubling since February. The competitive environment for condos remains less extreme than for houses, making it a more approachable entry point for buyers who can be flexible on property type.
How long will the AI-driven real estate boom in San Francisco last?
The direct demand from AI sector liquidity events will follow those events. Upcoming IPOs and secondary transaction activity are expected to sustain the current buyer profile through the near term. The structural factors — supply constraint and durable location demand — support pricing regardless of the specific demand source.