The Likes of Blackstone and Brookfield Are Using AI to Build Real Estate Tech In-House

Even as some of the biggest commercial real estate companies shoulder the challenges and massive costs of developing proprietary artificial intelligence-powered tools, this shift marks not so much an inflection point for proptech as a sign that the commercial real estate industry has never embraced technology more seriously to run its business efficiently.

General speculation that proptech startups and their venture capital backers might be in retreat spiked this spring, when a pair of multibillion-dollar deals made the industry take notice. 

On May 4, Anthropic announced a $1.5 billion joint venture with financial leaders, including Blackstone and Goldman Sachs, originally called the Claude Partner Network and renamed Ode, which will customize AI solutions and provide customer support to select clients. In a news release, the partner group said midsize firms “lack the in-house resources to build and run frontier deployments.” 

The new network didn’t single out real estate, but it exemplifies the potential to reshape proptech at enterprise scale. General Atlantic, Leonard Green, Apollo Global Management, GIC and Sequoia Capital, all financial backers of the Claude network, clearly see the possibilities. 

That same May day, OpenAI announced the $10 billion OpenAI Deployment Company, a joint venture with TPG, Brookfield, Bain Capital, Advent, SoftBank and Dragoneer focused squarely on real estate portfolio companies.

Industry experts hope in-house proptech development will handle the same core workflows independent proptech companies already tackle, such as portfolio management and underwriting.

Still, the same experts view these commitments by leading investors and landlords as the emergence of another pillar of the proptech investment market rather than a sign that VCs, startups and independent players need to worry about being eclipsed by capital and technical expertise. 

The numbers bear this out: Venture capital investment in proptech reached $16.7 billion in 2025, a 68 percent increase over the previous year, according to the Center for Real Estate Technology and Innovation (CRETI). AI-native companies captured $4.5 billion of that, growing their share of proptech VC dollars 42 percent year-over-year — nearly double their software as a service (SaaS) counterparts, which face a real threat from AI innovators. That suggests investors aren’t fleeing the category so much as rotating hard toward AI-native newcomers and away from legacy SaaS wrapper models.

Josh Panknin, director of real estate AI research and innovation at Columbia’s Fu Foundation School of Engineering and Applied Science, has surveyed the sprawling $50 billion to $60 billion proptech startup landscape and predicts the bigger players will find enterprise-scale AI development trickier than it looks.

“Real estate companies are not technology development companies, and haven’t been as successful as them in the past,” Panknin said, which leaves room for a robust startup space in a decentralized industry.

“It’s very fragmented — the data that comes in on real estate comes from very different sources, different formats, different market definitions,” he said, which keeps proptech applications from scaling across markets and problems.

That also means the proptech startup market hasn’t produced many billion-dollar unicorns. Panknin said the most upside is not in point solutions and automating workflows, but in focusing on infrastructure and foundational capabilities. 

The five largest independent proptech companies — CoStar Group, Yardi Systems, RealPage, Procore Technologies and AppFolio — focus on market data, property management and construction management, specific slices of the complex, multifaceted CRE industry. Collectively, they reported about $9.2 billion in fiscal 2025 revenue, and will likely anchor proptech for some time.

Still, the barriers that once protected proptech incumbents — technical complexity, cost and talent — are eroding. Ashkan Zandieh, managing director of CRETI Ventures at CRETI, calls the change significant but subtle.

“Large owners building AI capabilities won’t shrink the proptech sector,” Zandieh said. “We’ve seen this in banking, retail and health care, where internal enterprise software teams ultimately fostered larger vendor ecosystems. What we’re seeing in real estate is a sign of market maturity, more than a threat to proptech.”

He said he views the headline-grabbing deals as overhyped, at the very least. Blackstone’s push into AI aims to improve portfolio intelligence, since the $1.35 trillion alternative asset manager generates revenue that way, and harnessing internal data will create efficiencies. Brookfield, likewise, is seeking enterprise-wide AI solutions to cover its vast global portfolio. 

Neither poses a significant threat to proptech, Zandieh said: “They’re not in the business of selling software solutions.”

Most venture capital firms still focus on startups that address discrete corners of the diffuse real estate industry: construction, transaction management, design and the nuts-and-bolts work of running large commercial properties.

Travis Connors, co-founder and general partner at building ventures, a Boston VC firm whose portfolio includes Built Robotics, Dyn and Clearstory, said the Claude partnership and Brookfield’s tech ramp-up create “a more noisy environment” for startups. But he echoed Zandieh’s view that the industry is simply maturing and learning how to incorporate AI into its businesses.

“Real estate is still underleveraged from a technology perspective, and AI presents a new opportunity to get tech into all those processes,” Connors said. “Ten years ago, technology wasn’t the first answer for improving processes in this domain. Now it’s, ‘Is there technology I can use to improve whatever this process is?'”

He said a combination of the C-suite’s awakening and the broader hype over enterprise AI has real estate leaders “saying ‘Go find us AI solutions to everything’” — an unrealistic expectation. What works is letting AI boost efficiencies, because for many businesses that support commercial real estate, time truly is money.

“For a contractor, someone in a low-margin business, you can improve that cycle loop and time loop, and then you can improve your business,” Connors said.

Brokerage services giant JLL takes a multi-track approach, backing startups, forming strategic partnerships, making acquisitions and developing proptech internally through its venture arm, JLL Spark Global Ventures. It has invested about $450 million across roughly 55 proptech startups since launching in 2018.

Ajey Kaushal, a principal, points to its 2018 stake in Dealpath, a transaction management platform now widely used across the industry, including by competitors Cushman & Wakefield and CBRE, as proof that big companies can stay engaged in the startup space. 

More recently, Spark backed the Israeli firm Qbiq, which visualizes customized office environments faster and more cheaply than traditional architecture firms can. JLL Spark takes a flexible approach, focusing on industry pain points with startup investments that represent larger market opportunities. It keeps some investments in-house, exits others and retains stakes in independent companies its competitors are willing to pay for.

That multifaceted mandate may become a template for other big firms as they take the AI plunge. But broader concerns about the technology’s hype and Wall Street’s occasional doubts — underscored by the July 26 tech stock plunge — could also temper CRE’s embrace of AI.

“What JLL realized was that technology strategy was built around four pillars: build, buy, partner and invest,” Kaushal said.