📰 Key Takeaways

Blackstone is positioning itself at the center of the AI boom. Senior executive Jas Khaira said at TechCrunch Disrupt that the firm’s approach includes big loans to data centers as well as investments in frontier AI models and applications. Earlier this year, Blackstone set up an AI-focused hub in San Francisco, Blackstone N1 (BXN1), and Khaira relocated from New York to lead it. He pointed out that AI isn’t like the mobile or SaaS eras, where you could build a big company with modest capital. It takes an astonishing amount of money, and that’s exactly where Blackstone, with $1.3 trillion in assets under management, can play a key role. Khaira stressed that what sets Blackstone apart from traditional VCs is its focus on compute financing and its ability to offer flexible capital structures, from debt to equity. Today it’s the largest holder of compute outside the hyperscalers. Deals BXN1 has been involved in include Crux AI, a neocloud working with Google to promote its own TPUs, where Blackstone committed $5 billion as an investor, and Etched, a four-year-old chip startup it also invested in. Blackstone pioneered a financing model that evaluates the customer’s committed contracts while using pledged GPUs as collateral, lending huge sums to VC-backed neoclouds. The first example came in 2023 with CoreWeave: demand had exploded after ChatGPT launched, but CoreWeave had only about $30 million in revenue and limited profit. It still asked Blackstone for a $2.3 billion loan, secured by a large number of Nvidia GPUs, with top-tier customers like Microsoft. That deal later became the template for Blackstone and other credit firms extending billions of dollars in follow-on debt to CoreWeave and its peers. Khaira said the key is who the customer is. If the customer is an investment-grade company like Microsoft or Meta, the risk is priced off that customer’s credit. The original article covers more neocloud financing cases, so see the link for details.

💬 JudyAI Lab Take

Blackstone manages $1.3 trillion in assets, and it’s treating AI infrastructure as its main battlefield, going all in from data center loans to frontier model investments. This story matters because it shows AI competition has moved beyond a technology race and into the design of capital structures.

Blackstone’s Jas Khaira put it bluntly: AI isn’t like the mobile or SaaS eras, where you could scale on modest capital. It needs an astonishing amount of money. For AI builders, this reflects a design mindset: compute itself can be financialized. The 2023 CoreWeave deal is the template. Back then it had only about $30 million in revenue, yet by pledging Nvidia GPUs as collateral and adding committed contracts from customers like Microsoft, it landed a $2.3 billion loan. Khaira also explained that risk is priced off the customer’s credit. If the customer is an investment-grade company like Microsoft or Meta, the financing terms look very different. In other words, a stable contract with a big customer is itself an asset you can borrow against. And since Blackstone is now the largest holder of compute outside the hyperscalers, capital and compute are concentrating fast.

If your product depends on a lot of compute, take stock of which long-term contracts or steady customers you have, and think about whether those commitments could turn into a more favorable cost structure.

📅 Source Info

🔗 Further Reading