Blackstone-Google AI cloud JV ignites investor buzz—but can it outpace nvidia’s shadow?

Blackstone and Google have launched a $5 billion joint venture to build a U.S.-based AI compute infrastructure company focused on leveraging Google's Tensor Processing Units (TPUs) for large-scale AI workloads. Treynor Sloss, a senior Google executive, has been named CEO, underscoring Google's strategic leadership role in scaling this new platform to rival established AI infrastructure providers like CoreWeave. This leadership choice signals a focused ambition to compete aggressively in the rapidly evolving AI compute market.

The joint venture marks a strategic expansion of TPU access beyond Google Cloud by creating a standalone cloud platform that offers enterprises an alternative channel to tap into Google’s custom AI accelerator chips. By decoupling TPU availability from Google Cloud’s traditional consumption model, the partnership aims to disrupt the Nvidia-dominated AI infrastructure landscape and meet surging demand for both AI training and inference workloads, particularly driven by generative AI and autonomous systems.

Blackstone’s commitment of an initial $5 billion in equity funding, combined with Google’s provision of hardware, software, and services, targets the delivery of roughly 500 megawatts of data center capacity by 2027. This ambitious scale reflects Blackstone’s strategic pivot from traditional data center REIT activities toward becoming a major player in AI compute infrastructure, with President and COO Jon Gray emphasizing that AI will catalyze one of the largest infrastructure buildouts in history.

Beyond capital and scale, the venture aims to leverage TPU technology’s unique architecture optimized for matrix-heavy workloads, potentially reshaping cost and performance dynamics for large-model AI training and inference. Google CEO Sundar Pichai highlights that this partnership not only broadens TPU access but also helps Google achieve greater economies of scale in its compute environment, enabling investments in next-generation AI infrastructure critical for maintaining technological leadership.

Google’s joint venture with Blackstone to expand TPU-driven AI infrastructure beyond Google Cloud represents a strategic extension of its vertically integrated AI platform, which combines proprietary TPUs, Frontier models like Gemini and DeepMind, and broad distribution across Search, Android, YouTube, and Workspace. This integration not only enhances Google’s control over AI cost structures but also positions it uniquely against competitors such as Nvidia and CoreWeave, who rely heavily on external GPUs, by enabling more efficient and cost-effective AI service delivery at scale.

The TPU infrastructure offers Google a significant cost and performance advantage that could be decisive in the AI race, where the ability to serve intelligence at the lowest cost across a massive user base is as critical as model quality. By leveraging its long-standing expertise in operating large-scale infrastructure, Google aims to reduce AI serving costs while simultaneously driving revenue growth through enhanced AI monetization across its vast ecosystem, shifting market perceptions from an advertising incumbent to a diversified AI platform company.

Investor sentiment towards Blackstone’s AI data center REIT is cautiously optimistic, driven by its REIT structure that appeals to income-focused investors through potential dividend growth and tax advantages. The strategy to lease newly built, mission-critical data centers to investment-grade hyperscalers under long-term triple-net leases with annual rent escalators provides a stable and escalating cash flow profile, which reassures institutional investors despite the REIT currently being asset-light and externally managed—a structure often viewed less favorably compared to internally managed REITs.

Blackstone’s $5 billion joint venture with Google and its broader push into digital infrastructure are reshaping its growth narrative by expanding fee and capital deployment catalysts, yet they concentrate execution and technology risks amid market uncertainties. Analyst forecasts vary widely, with some projecting revenues near $21.9 billion and earnings around $8.9 billion by 2029, while others caution that rapid technology shifts and macroeconomic headwinds could undermine these ambitions, highlighting the delicate balance between growth potential and operational challenges.

The financing landscape for AI infrastructure is evolving rapidly, with Blackstone and peers like Apollo structuring massive asset-backed debt deals—such as the approximately $36 billion Anthropic financing—that leverage tangible assets like custom chips and data center capacity to mitigate risk. This sophisticated capital formation involves institutional investors across insurance, pension, and sovereign wealth funds, reinforcing Blackstone’s positioning not only as a real estate owner but also as a key provider of structured capital for AI compute infrastructure.

Institutional investor activity underscores growing confidence in Blackstone’s AI infrastructure strategy, exemplified by MUFG Securities EMEA plc’s 373% stake increase and analyst upgrades like DBS Bank’s 'moderate buy' rating. However, this enthusiasm is tempered by ongoing risks in other portfolio segments, including significant office real estate losses—such as a planned sale of a Seattle office building at a 54% loss—and broader valuation concerns, which collectively suggest that while AI ventures offer promising growth avenues, overall investor sentiment remains measured.

The $5 billion AI data center joint venture with Google marks a pivotal expansion for Blackstone into the digital infrastructure realm, reinforcing its growth narrative by potentially boosting fee generation and deployment catalysts. By 2028, Blackstone projects revenues of $21.5 billion and earnings of $10.5 billion, with this AI infrastructure initiative contributing to a fair value estimate of $162.26 per share—implying a 37% upside from current levels. This strategic move underscores digital infrastructure as an increasingly central pillar in Blackstone’s long-term growth trajectory.

Despite the promising growth outlook, investor sentiment remains tempered by the heightened execution and technology risks inherent in large-scale, capital-intensive digital infrastructure projects like the Google AI data center. Analysts caution that rapid technological shifts and operational complexities could strain Blackstone’s expanding digital pipeline, potentially offsetting anticipated benefits. Moreover, macroeconomic headwinds such as trade tensions and delayed deal activity continue to pose risks to near-term realizations and distributions, indicating that the venture, while transformative, does not insulate Blackstone from broader market vulnerabilities.