AivexaNewsSearch
AI news for builders and product teamsChecked every hour

Blackstone’s Jas Khaira says fears over AI-buildout debt are overblown

Collected Oct 21, 2026

Blackstone, the private equity firm with $1.3 trillion in assets under management, is positioning itself at the center of the AI buildout, and its AI chief used a TechCrunch Disrupt appearance last week to push back on fears that the debt financing that infrastructure is spiraling toward a credit crisis. Jas Khaira, who leads the firm's AI efforts and relocated from New York to San Francisco for the role, argued the situation does not resemble 2008, when shaky bank credits threatened the financial system. His reasoning rests on who ultimately backs the borrowing.

Earlier this year Blackstone launched Blackstone N1, known as BXN1, a San Francisco hub for its AI initiatives. The group deploys capital across a wide range of strategies: lending large sums to data centers, taking equity in frontier model developers and applications, and financing compute itself. Khaira framed that mix as a differentiator from traditional venture capital, which typically takes equity positions. Blackstone can structure deals anywhere from debt to equity, and it has leaned heavily on the debt side.

Khaira said Blackstone is the largest owner of compute outside the hyperscalers, counting investments in neoclouds and data centers. Neoclouds are cloud providers that rent out AI GPU infrastructure rather than building full-service platforms. BXN1 has been involved in several compute deals, including Crux AI, a neocloud launched with Google to commercialize Google's custom TPUs, to which Blackstone committed $5 billion as an investor. It also backed Etched, a four-year-old chip startup whose pace of development has drawn investor attention.

The financing model that made Blackstone central to this market started in 2023 with CoreWeave, shortly after demand for that neocloud surged following ChatGPT's launch. According to Khaira, CoreWeave had roughly $30 million in revenue and minimal profitability when it approached Blackstone for a $2.3 billion loan. Blackstone was willing to lend against the value of CoreWeave's large fleet of Nvidia GPUs, with Nvidia itself an investor, and it grew more comfortable because CoreWeave held contracts with blue-chip customers such as Microsoft. That structure, analyzing committed customer contracts and treating GPUs as pledged collateral, became the template for how Blackstone and other credit institutions extended billions in follow-on debt to CoreWeave and rival neoclouds.

Khaira described customer quality as a primary consideration. When the customer is investment-grade, such as Microsoft or Meta, the risk is priced relative to that customer. Blackstone has since expanded the playbook, including co-leading with Coatue a $10 billion credit facility for the Australian GPU provider Firmus earlier this year.

Pressure has nonetheless built in recent weeks. CoreWeave's debt has grown to about $50 billion, prompting Altimeter Capital founder Brad Gerstner to say in a CNBC interview that the collateral CoreWeave puts up, essentially GPUs, is no longer good enough and had previously helped it obtain cheap debt. Khaira attributed market jitters on the credit side to several forces: interest rates ticked up in September, hotter-than-expected inflation suggests borrowing costs may keep climbing, and a flood of AI infrastructure debt offerings is adding supply. Credit spreads have widened, meaning lenders charge higher premiums on top of already elevated base rates. Even companies with huge balance sheets and solid earnings, like Meta and Microsoft, now pay more to borrow, he said.

The repricing has reached public markets. CoreWeave shares fell more than 20% in recent days, and Firmus scrapped a planned IPO on the Australian Securities Exchange at a target valuation of roughly $30 billion. Khaira said the market simply was not there at that valuation. The broader question investors are working through, in his telling, is what the right price is for neoclouds when interest rates are moving up.

Even so, demand for compute remains intense, and AI companies' appetite continues to drive data center construction. Khaira's core argument against a systemic crisis is structural: the vast majority of data centers are built for hyperscalers, so the debt financing them is ultimately backed by the most creditworthy companies in the world. This suggests the risk sits less with speculative neocloud borrowers than with the investment-grade tenants whose contracts underpin the loans, and that the widening spreads reflect repricing rather than distress. A likely trade-off is that the same discipline cuts both ways: as spreads widen, smaller or less creditworthy operators face higher costs, which is arguably what the Firmus listing outcome and CoreWeave's stock reaction already reflect.

Beyond infrastructure, Blackstone is funding adoption. With Anthropic and Hellman & Friedman it formed a $1.5 billion joint venture called Ode with Anthropic this year. Ode supplies forward-deployed engineers who help companies integrate AI into operations, beginning with Blackstone's 280 portfolio companies. Asked whether those companies see tangible returns, Khaira said there definitely is ROI but that it is uneven, describing adoption as a sawtooth pattern that to skeptics can look like oversupply of compute before usage catches up.

That unevenness is the piece most likely to matter for product teams. Compute financing at this scale tends to determine which neoclouds can keep renting GPU capacity cheaply, and widening credit spreads can feed through to pricing and availability. In practice, developers weighing providers may see more differentiation on contract strength and credit quality, not just on hardware. For product teams inside large enterprises, services like Ode are an attempt to shorten the gap between buying AI capacity and getting returns from it. Whether the hyperscaler-backed lending thesis holds is an inference this market will test as rates and spreads move.

Why it matters: Neocloud operators, data center developers and the AI startups that depend on rented GPU capacity are directly exposed to how lenders like Blackstone price this debt. If the backers are right, capacity keeps flowing and small teams keep access to compute; if spreads keep widening, marginal providers face harder terms and fewer options, and pricing and availability could shift for developers building on them.

Read at TechCrunch · AI

Based on reporting from the original publisher. Visit the source for full context and later updates.

Publisher excerpt

Blackstone’s AI head on why the PE giant isn't panicked by market jitters.