In May 2026, Remote asked 3,250 hiring leaders one question: has AI changed where you hire? 97.7% said yes. This is what the fastest redraw of the global talent map looks like from inside the companies making it.
For most of the last decade, the hiring map looked settled. Engineering clustered in Silicon Valley, finance in London, and companies hired against those pools where they already sat — going global later, if ever, once they were big enough to afford the entities and the lawyers.
AI redrew that map in about eighteen months.
In May 2026, Remote surveyed 3,250 business and HR leaders with direct recruitment responsibility across eight markets: the UK, the US, Spain, Singapore, the Netherlands, Germany, France, and Australia. We put one question at the center of it: has AI adoption changed where you hire? 97.7% said yes. Not "we're considering it." Yes, it already has.
Staying local isn't a strategy anymore. For most of these companies, it isn't even a choice.
Traditional skill sets take decades to concentrate. A city builds a university pipeline, a few anchor employers, a supporting ecosystem, and eventually a talent pool deep enough to hire from without leaving town. AI expertise hasn't had a single decade to do any of that. It's too new and too thin, and demand is stampeding well ahead of the supply any one market can produce.
So companies stopped competing locally and started competing globally — all of them, for the same people, in the same few places, at once.
For many, it wasn't ambition. It was necessity. 72% of the same leaders told us they'd missed key business goals in the past 12 months because of a talent shortage. When the skill you need to ship your AI roadmap doesn't exist within a thousand miles, "hire local" stops being conservative and starts being the risky option.
Here's the part most coverage of the "AI talent war" gets wrong. The bottleneck isn't sourcing. Companies can already identify and reach an AI engineer almost anywhere on earth; LinkedIn and a recruiter solved that years ago. The bottleneck is everything that happens after the offer.
Picture two companies that spot the same standout engineer in Hyderabad. One can get her on payroll in three days: compliantly, under Indian employment law, with benefits and taxes handled. The other needs three months to stand up the infrastructure to employ a single person in a country it doesn't operate in. They're bidding for the same hire. Only one of them can actually close.
That's the whole game now. It's not a recruiting problem; it's an onboarding-speed problem. And it's why employment infrastructure, not sourcing, is what separates the companies that can compete for AI talent from the ones that just watch it get hired elsewhere.
Opening a company up to international talent is not a low-friction decision, so the scale of the movement is the signal. 40% of these companies have opened operations in entirely new countries specifically to reach AI talent. Another 58% are scaling their hiring globally inside markets where they already operate.
The trajectory is steeper than the snapshot. By the end of 2026, nearly half of these companies expect the majority of their new hires to sit outside their primary country of operations. International hiring is crossing over from exception to default in the space of a single planning cycle.
Wanting to hire globally and being able to are two different problems, and the second one is where most companies stall.
The instinct, when you decide to hire in a new country, is to build: register a legal entity, open a local bank account, set up payroll, retain local employment counsel, learn the filing calendar. There's a fair case for it: an entity gives you permanence and full control. But it takes months, and by the time it's live, the engineer you built it for has been at a competitor for a quarter. You've paid for infrastructure to hire a person you already lost.
An Employer of Record (EOR) inverts the trade. Instead of building infrastructure in every market, you borrow one that already exists. The EOR is the legal employer on paper; you own the hiring, the pay, and the day-to-day. Everything underneath sits with the provider: payroll across currencies and tax regimes, compliance with local employment law, benefits, statutory filings. With Remote, that infrastructure is owned outright: Remote operates its own entities in the markets it covers, so there's no third-party handoff in the chain and no mystery layer between you and the person you're hiring.
The difference is measured in days versus months, and in skipping the entity setup, local accounting, and compliance overhead you'd otherwise have to build before your first hire signs. What it buys isn't just speed. It's the freedom to make hiring decisions based on where the talent actually is, rather than where you happened to plant a flag years ago.
The shift isn't regional. It's universal. In every market we surveyed, more than nine in ten AI adopters have changed where they hire. Singapore leads at 100%. The US, the Netherlands, and Germany all clear 99%. Even France, the lowest, sits at 94%. Talent scarcity doesn't respect borders, and neither, anymore, does hiring.
Has AI adoption changed where you hire? (Yes, net) UK 96.4% · US 99.2% · Spain 97.4% · Singapore 100% · Netherlands 99.0% · Germany 99.4% · France 94.0% · Australia 97.6%
What diverges is where everyone is looking. Only 39.8% have gone as far as opening operations in new countries, and appetite for that bolder move varies sharply: US companies lead at 52.8%, France follows at 48.2%, and Germany is the outlier at just 23.6%.
The regional strategies split along lines of proximity and familiarity:
European companies hire close to home. The UK leans on Europe (53.4%), with South Asia a strong second (33.8%). The Netherlands mirrors it (56.8% Europe); Spain follows the pattern (41.8%).
US companies run three regions at once: North America (48.4%), South Asia (31%), and Latin America (30.8%), hedging across time zones rather than committing to one.
Singapore concentrates. Southeast Asia captures 59.2% of its hiring attention — the only case in the survey where a single region wins an outright majority from a single country, reflecting both proximity and the region's rise as an AI hub.
Australia, more isolated, spreads out but shows unusual strength in Oceania (36.8%), leaning on nearby pools before reaching further.
The most revealing split isn't geographic. It's by size, and it cuts against the usual assumption that the biggest companies move most aggressively abroad.
Smaller enterprises (500–749 employees) are the most likely to open operations in entirely new countries, at 43.2%. Mid-market (750–999) follows at 38.5%. Large enterprises (1,000+) trail at 35.0%. Flip to reshuffling work across an existing footprint, though, and the order reverses: 61% of large enterprises have moved roles to different countries, versus 55.8% of smaller ones.
Two different playbooks, each rational for who's running it:
Smaller and mid-market companies build. They don't have operations abroad to redeploy, so reaching new talent means taking on new markets — and the infrastructure risk that comes with them. For these companies, borrowed employment infrastructure isn't a convenience; it's the only version of the move that's affordable.
Large enterprises reshuffle. They optimize within what they already run, moving engineers from London to Singapore or teams from San Francisco to Toronto, because the footprint is already there to move within.
Global hiring does two things simultaneously. It changes how companies operate, and it rewrites the incentives for the people they want and the countries competing to keep them.
For companies: whoever can employ the hire fastest wins them. The competitive edge has moved from recruiting to execution. Can you get her on payroll in three days or three months? Can you navigate local employment law without lawyers gating every offer? Can you run her benefits across currencies without rebuilding your system? Answer no to any of these and you default, quietly, to hiring only where you already operate — locked into geographic decisions made years ago, for reasons that no longer apply.
For the people being hired: a global salary without leaving home, and global competition for the role. The old model forced relocation: join the company, move to its headquarters. That barrier is gone. An engineer in São Paulo can work for a US company without emigrating; a researcher in Bangalore can join a German firm without leaving India, at local salary, under local law. The opportunity is enormous. So is the new competition. That same engineer is no longer up against the talent in her city. She's up against everyone, everywhere the company is willing to hire.
For economies: countries can finally keep the talent they used to export. Talent has historically flowed one way, from developing economies to developed ones, draining the countries that produced it. Direct global hiring reverses the current. India keeps its AI engineers because companies employ them in India. Eastern European developers build for London firms without moving to London. The talent stays home, and so does the economic value it creates — but only where the systems exist to employ people legally and at scale. The distribution of AI capability is becoming less about where people live and more about where the infrastructure exists to employ them.
Here's what "just hire globally" hides. Every time a company adds a country, complexity doesn't add. It multiplies.
Run payroll in five countries and you don't inherit one rulebook five times over. You inherit five tax authorities, five sets of benefit regulations, five compliance calendars, and five bodies of labor law that share almost nothing. Employment law in Singapore doesn't resemble the Netherlands'. A French contract runs on rules an Australian one has never heard of. Statutory contributions, pension mandates, local benefits, filing deadlines — all of it forks at every border.
And so does the risk. Miss a filing deadline in Germany and you're fined under German rules; miss one in the UK and you're penalized under different rules entirely. Most companies don't have that expertise in-house, and the ones that try to build it end up spending on compliance the resources they meant to spend on talent.
The point of shared employment infrastructure is that it makes the tenth country cost roughly what the first did. You're not redesigning payroll or rebuilding compliance each time you expand. The system carries over, and the efficiency compounds as you grow. It lowers the cost of experimenting, too: test a market, and if the talent isn't there, scale back without eating a sunk entity. The people you hire get the same onboarding, benefits, and reliable payroll whether they're in Hyderabad or Berlin — which is what makes a distributed team feel like one team instead of a set of disconnected operations.
Global hiring infrastructure won't stay a differentiator for long. Within a few years it becomes table stakes; nearly every company will expect to employ people across borders. The question stops being can you hire globally and becomes can you do it well.
When everyone has access to the same rails, the contest moves back up a level — to execution. Who actually knows where the deepest talent pools are? Who can move fastest on the hire everyone wants? Who keeps the people they hire? That's where the next advantage lives.
Which is exactly why moving now still matters. The companies hiring internationally today are learning which markets deliver, building the operational muscle, and forming the teams — while slower competitors are still deciding whether the shift is real. By the time the infrastructure is universal, the early movers won't be starting from the same line. They'll be years down the road.
The geographic assumptions companies made a decade ago no longer hold. AI expertise redrew the talent map, and this version won't be redrawn back. The scarcity is structural and will last years; the companies that have already restructured around global hiring have invested in the infrastructure, built the distributed teams, and learned which markets work — and none of that unwinds when the market cools.
For any company betting on AI, the conclusion is uncomfortable but simple: your ability to compete on AI is now inseparable from your ability to employ the people who build it, wherever they are, before someone faster does.
