Blackstone and Google have announced a joint venture to establish a new U.S.-based company that will offer data center capacity, operations, networking, and Google Cloud's Tensor Processing Units as a compute-as-a-service offering, giving enterprise customers an additional channel to access cloud TPUs beyond Google Cloud's existing platform.
Structure of the Deal
Blackstone is committing an initial USD 5 billion in equity capital drawn from funds it manages as part of the partnership. The new company expects to bring its first 500 megawatts of capacity online in 2027, with both parties indicating plans to scale the venture significantly beyond that threshold over time.
Google will supply hardware, including TPUs, alongside software and services, to allow the new entity to accelerate its capacity buildout in response to what both companies describe as growing demand for accelerated computing infrastructure.
The joint venture is structured so that customers will have a choice between accessing TPUs through the new company or through Google Cloud's existing service.
Blackstone describes itself as the world's largest alternative asset manager, with over USD 1.3 trillion in assets under management, and identifies its data center operations as the largest global provider of such infrastructure. The company's involvement reflects a broader push by major alternative asset managers to deploy capital into AI-related physical infrastructure.
Leadership Appointment
Blackstone has named Benjamin Treynor Sloss to serve as chief executive officer of the newly formed company. Treynor Sloss is a Google executive with more than two decades of experience building and operating Google's global infrastructure and operations.
His appointment signals an intent to draw on deep institutional knowledge of the technical and operational demands involved in running large-scale cloud compute environments.
Trusted by Leading EPCs & Manufacturers
Find the Latest Data Center Projects Around the World
Gain exclusive access to our industry-leading database of data center opportunities with detailed project timelines and stakeholder information
Request Free Trial → Learn More →
No credit card Up-to-date coverage
What TPUs Are and Why They Matter
Google's Tensor Processing Units are custom chips purpose-built for artificial intelligence workloads, specifically optimized for training and inference of advanced AI models. Google has been developing and deploying TPUs in production environments for more than a decade.
The chips currently power workloads for a range of clients the company describes as including many of the world's top AI laboratories, capital markets firms, and companies running high-performance computing applications. TPUs also underpin Google's Gemini AI models and the AI-driven products Google delivers to its billions of users globally.
The decade-long development history of TPUs was cited explicitly by Jas Khaira, Head of Blackstone N1, the unit within Blackstone that is backing the venture.
Khaira described the chips as "foundational to the AI economy" and said the joint venture represented exactly the kind of platform that Blackstone N1 was structured to support.
He emphasized that capital alone is not sufficient to build what he characterized as category-defining platforms, citing the importance of the right partner, the right deal structure, and the conviction to back singular opportunities.
Strategic Rationale from Both Sides
Jon Gray, President and Chief Operating Officer of Blackstone, framed the investment as part of what he called a generational opportunity to deploy capital at scale in AI infrastructure.
Gray said the new company had significant potential given what he characterized as unprecedented demand for compute, and pointed to the combination of Google's TPU and AI capabilities with Blackstone's strengths in energy and digital infrastructure as the basis for the partnership.
Thomas Kurian, Chief Executive Officer of Google Cloud, said the joint venture was designed to meet growing demand for TPUs, describing the chips as optimized specifically for efficiency and performance in the AI era. Kurian said the arrangement would accelerate AI transformation and provide organizations with more options to access accelerated compute capability.
The framing from both sides positions the venture as a supply-side response to infrastructure constraints in the AI compute market, with the new entity intended to operate as a distinct offering rather than a simple reselling arrangement.
Google's role as hardware and software supplier, combined with Blackstone's role as the capital provider and infrastructure operator, establishes a division of responsibilities that gives each party leverage in its respective area of strength.
Blackstone's Broader Infrastructure Position
Blackstone's participation in this venture is consistent with its stated strategy of channeling large amounts of institutional and individual investor capital into physical infrastructure tied to technology and energy.
The firm manages investment strategies spanning real estate, private equity, credit, infrastructure, life sciences, growth equity, secondaries, and hedge funds.
Its existing position as the world's largest global provider of data centers gives it operational experience relevant to standing up and running the kind of high-density compute facilities that TPU workloads require.
The new company will be U.S.-based, and the announcement did not specify the geographic locations of the data center facilities planned for the initial 500-megawatt deployment. The 2027 target for first capacity coming online implies a roughly 12-to-18-month buildout timeline from the date of announcement in May 2026.
Power Your Data Center Pipeline with Intelligence That Spans Markets Worldwide
How much opportunity is slipping past your team simply because the signals were buried in fragmented sources across markets worldwide? In the data center sector, where land acquisition, power procurement, and hyperscaler demand can shift a project's trajectory overnight, the organizations winning mandates are the ones who see what's coming before their competitors do.
The Global Project Tracking (GPT) platform by Blackridge Research was built to give developers, contractors, investors, and technology vendors a single, authoritative lens on the entire data center project lifecycle. From greenfield campuses to edge deployments, every stage of development is tracked and structured for immediate commercial use.
Whether you are targeting colocation expansions in emerging corridors or monitoring hyperscale awards across established hubs, the platform delivers the depth and timeliness your business development teams need to act with confidence.
Upcoming Projects
Tender Notices
Contract Awards
Projects Under Construction
Completed Projects
See exactly how the Global Project Tracking (GPT) platform by Blackridge Research maps to your data center growth strategy — Book a Free Demo with our team today.
Leave a Comment
We love hearing from our readers and value your feedback. If you have any questions or comments about our content, feel free to leave a comment below.
We read every comment and do our best to respond to them all.