While artificial intelligence could automate much of the workplace, leading to fewer workers and less office space, it's also prompting the creation of companies that are major users of real estate.
ChatGPT maker OpenAI, Anthropic, Nvidia, Databricks and other AI companies have collectively become the largest pursuers of space across the U.S. market, helping to lower record-high levels of office availability and leading a post-pandemic recovery. That is despite other major companies, such as tech giant Amazon, choosing to cut their expansive footprints.
Against this shifting landscape, no one knows whether AI will lead to a commercial real estate boom or bust — or both. Just as the technology is acting "as a catalyst for both growth and decline" in the job market, it can have differing effects on office property, according to real estate services firm CBRE.
"AI brings augmentation of human capabilities to a new level, fostering significant productivity gains and creating new job functions," according to a report from CBRE Investment Management. "As the composition of office work changes, so too will space requirements, workplace design and tenant demand."
Seattle-based Amazon, for one, is allowing leases to expire, temporarily suspending, or "hibernating," activities at some offices, and subleasing or terminating deals for hubs that have become largely vacant, Senior Real Estate Manager Martha Schwarzkopf Doyle said at a Global Real Estate and Facilities team meeting earlier this year.
"As it relates to how AI companies are going to affect other companies in terms of their leasing, that's the million-dollar question," Louis Thibault, Avison Young's manager of market intelligence, told CoStar News.
But for now, office landlords are sharpening their focus on Class A offices — the nicest and newest properties with tenant perks you can't find online — as they say there will be winners and losers in the wake of this shift toward automation, just as there was during the pandemic.
AI industry's real estate growth
Among the country's largest cities, commercial leasing generated by AI and tech companies accounted for about 20% of the total volume last year, the most of any industry, according to data from CoStar and real estate services firm CBRE.
That has been especially pronounced in tech hubs such as the San Francisco Bay Area, where AI firms have accounted for half of the major leases signed since the start of 2026. The activity has boosted their collective footprint well beyond 5 million square feet of office space, a presence expected to surge to at least 15 million square feet over the next four years.
"In previous tech booms, most of the capital companies raised went to hiring and expanding their workforce, but the big difference this time around with the AI sector is that they're spending a large amount of funding on infrastructure," Colin Yasukochi, head of CBRE's Tech Insights Center, told CoStar News. "The AI industry is in the office five, six days a week, and many are already starting to open additional offices in cities like Seattle, Boston and London. It has so far led to a lot of economic growth."
While they may start with a small office between 3,000 square feet and 5,000 square feet, AI firms and startups have been quick to tack on additional space, often within a matter of months, said Mike McCarthy, a Transwestern broker who has worked on a number of deals with such tenants. Especially as companies close additional funding rounds and bulk up their workforces, those spatial requirements are accelerating to 20,000 square feet, 40,000 square feet or, in some recent cases, full-building deals.
There has been "a tremendous amount of new business formation and growth, much of which is coming on the back of what's happened with AI over the last few years," Kilory Realty CEO Angela Aman told analysts on the landlord's latest earnings call. "We have AI tenants we've signed deals with that are already talking about expansion and growth, and there are a lot of additional new companies in the market thinking about taking additional space."
The CEO went on to say the AI-fueled boost, while only one part of how the technology is changing real estate, is creating "an exciting dynamic" that extends far beyond the most concentrated tech hubs, as the demand has helped prop up the broader national office market.
Braced for impact
At the same time, companies outside of the AI industry are using the technology to streamline their workforces, making cuts that have office market stakeholders watching for any sign of a potential bust.
Layoffs among corporate giants such as Salesforce, Meta, Workday and Pinterest in recent months have cast a pall over the market, as companies have attributed the cuts to their growing use of AI and a broader push to automate more traditional office roles.
Payments company Block, for example, unveiled plans in February to slash its workforce by at least 40%, a move CEO Jack Dorsey attributed to technology that has improved to the point where it's possible to do more with far fewer people.
"Intelligence tools have changed what it means to build and run a company," Dorsey said in public remarks about the planned cuts. "A significantly smaller team using the tools we're building can do more and do it better."
Still, the link between AI-related efficiency gains, layoffs and potential office real estate cuts remains tenuous. Some have questioned whether AI is an excuse for some CEOs to make cuts when a business doesn't want to admit it isn't generating enough revenue to justify its current spending.
Yet some of the nation's largest landlords, including Kilroy, Cousins Properties, Vornado Realty Trust and BXP, have been watching for any sign that the link between AI and reduced workforces is beginning to solidify — and translate into smaller space requirements.
They've found that, at least for now, tenants appear willing to commit to space, just as long as it's at the highest end of the market.
Bifurcated future
A flight-to-quality shift that took off in the earlier years of the pandemic, when companies began turning to high-end space to help entice staff to give up remote work and return to the office, has become permanently embedded in the national leasing landscape. Demand for trophy and premium properties far outpaces the desire by firms for aging, older alternatives.
"What we've been articulating for a couple of years is that AI is going to create jobs and it's also going to disrupt jobs that are more back office and processing," BXP CEO Owen Thomas said at Citi's Miami Global Property CEO Conference last month. "Our portfolio is geared towards that first group of employees, and that's what we're seeing."
So far, preliminary data backs up the job-creation effect of AI. Across the country, AI-related job postings have outpaced those for traditional tech roles, a sign that the technology is becoming more integrated across a broadening range of services and industries, Avison Young's Thibault said.
Moreover, it's an early indicator of the types of positions that will be formed to accommodate the AI boom as more companies race to adopt the technology.
"At the end of the day we might see certain industries and jobs go away, but historically we've always seen those jobs get replaced by something that requires new skills," Thibault said. "Folks behind typewriters that used to fill entire floors in office buildings went away but then were replaced by folks behind computers. Don't see AI, especially looking back historically, replacing all of the workers we have in office buildings today."
Committed to space
In the wake of the pandemic, which sent office vacancy rates to record highs and valuations to near lows, some office investors pulled out of the property type to focus on more lucrative investments, such as multifamily or healthcare.
Others converted some outdated offices entirely. For the first time in years, the rate of office demolitions and conversions has overtaken new development, according to CBRE and CoStar research.
Now, AI holds the potential to further accelerate changes in the office market, creating sharper winners and losers.
Based on the leasing trajectory in BXP's own portfolio, Thomas said demand for premium office properties remains firmly on the upswing.
Average lease sizes have climbed as terms are extended, the executive said, the clearest indicator that the need for physical office space isn't at any immediate risk.
"If a company was worried about AI, why are they in 2025 signing 10-year leases with us?" Thomas said, adding that, with the deals BXP has signed so far this year, the average terms have been even longer. "These are major financial commitments, and they're signing long-term leases, so I don't think they're forecasting big impacts AI will have on their space demand."
The outcome for older, more commodity properties is expected to be far bleaker, however, with AI technology expected to eliminate the need for many back-office and support roles that typically occupy those types of buildings.
Yet the pandemic had already determined the fate of those properties, and the increased adoption of AI is simply expected to deliver one of the final blows.
That means BXP is honing its focus on "upping the portfolio quality even further and getting even bigger in the gateway markets where we operate," Thomas said. "So yes, AI will absolutely have an impact, but that's why our strategy is taking an even narrower path."