NEW YORK, 24 AUG 2026 — New York has passed the San Francisco Bay Area as the largest technology talent market in the United States, with 394,300 tech workers against 375,730, according to CBRE's annual analysis.
It is the first time New York has held the position in the history of that survey. The gap is about 18,570 jobs, or roughly 5 per cent.
Two movements, not one
The lead changed because New York grew while the Bay Area shrank.
The Bay Area shed non-AI technology roles. New York added roles driven by artificial intelligence and by finance, which in New York are increasingly the same hiring line: banks, asset managers and insurers building AI capability in-house rather than buying it.
That composition matters more than the headline. A market can lead on headcount while trailing on the work that sets direction, and by most measures of frontier AI activity — model labs, research concentration, venture funding into foundation models — the Bay Area still leads comfortably.
What a headcount survey does and does not measure
CBRE is a commercial real estate firm and its tech-talent work is built to inform leasing decisions. That shapes what it counts.
It measures where technology workers are employed, which is a proxy for office demand. It does not measure output, capability, revenue, or where decisions get made. A market can gain 20,000 roles maintaining systems for regulated industries and lose 200 that would have built the next platform, and the survey will record a gain.
The number is still interesting, but as a measure of employment concentration, not technological leadership. Most coverage conflates the two.
The composition question that decides whether it lasts
AI roles reportedly make up close to a third of New York technology listings. That is a striking share and it cuts two ways.
If those roles are building capability inside financial institutions — models for underwriting, surveillance, client service — then the demand is anchored to an industry that is not going to relocate, and the position is durable. Finance has kept its people in New York through every previous technology cycle.
If a meaningful part is implementation work that a maturing tool market eventually absorbs, the same roles are exposed to exactly the automation they are being hired to deploy. That is not speculative. The Philippine IT and business process sector revised its 2028 outlook in July to a range whose lower end sits below its current headcount, on precisely that reasoning.
Layoffs are doing half the work
To be precise, the Bay Area did not just grow more slowly; it contracted in non-AI tech employment. That contraction is half the reason the ranking flipped.
A lead taken from a rival's losses is weaker than one from your own growth; it can vanish when the other market's cycle turns. The Bay Area has been written off before, in 2001 and again in 2009, and both times the concentration of capital and founders pulled it back.
This time might be different if New York's AI hiring is truly anchored to finance and other industries that will not move. The next two surveys will answer that, not this one.
Where AI is named in a layoff and where it is the cause
The wider labour data around this needs handling carefully, because the attribution is doing more work than the measurement.
Across the year to date, AI has been cited in roughly 50,000 United States job cuts, somewhere between 17 and 26 per cent of the total depending on the month. The share of layoff announcements naming AI rose from about 7 per cent in January to around 40 per cent by May.
A share that rises fivefold in four months measures corporate framing as much as it measures cause. Naming AI is a choice that can be true, convenient, or both. Any reading of the Bay Area's contraction that treats those citations as a clean causal series is reading a narrative rather than a number.
The pattern is dispersal, and it is not only American
A dominant technology centre losing its lead is not a local story; it is a structural one about dispersal.
Talent follows capital and capital has been spreading. Together AI booked up to 10,000 Nvidia B300s at a Chennai campus. Brazil announced US$444 million of AI investment including a supercomputing project with Huawei and iFlytek. Indonesia's sovereign fund has committed up to 30 per cent of annual deployment to digital infrastructure.
Those are compute buildouts rather than talent markets, and compute has historically pulled employment toward it with a lag. The Bay Area's advantage was never only its engineers; it was that everything was in one place. That property is what is eroding.
What it means from here
For an ASEAN employer competing for the same people, a second large American market bidding for AI talent is straightforwardly bad news on price, and it removes an argument that used to work.
The pitch for a Singapore or Kuala Lumpur role was never that it paid like San Francisco. It was proximity to Asian markets, a lower cost of living against the package, and work with regional reach. That pitch survives New York's rise intact, because a New York AI job in a bank is not competing on lifestyle either.
What does change is the comparison a candidate runs. Two American markets hiring hard for the same skills sets a wider band of reference salaries, and regional employers who benchmark against a single city will find the benchmark has moved without anyone telling them.