The Deal at a Glance

Apollo Global Management and Blackstone are structuring a financing arrangement tied to Broadcom that would total approximately $35 billion, according to The Wall Street Journal (published 9 June 2026). The capital is earmarked for AI infrastructure development, and if completed at that scale, the transaction would rank among the largest private credit deals ever assembled.

The mechanics still being worked through, but the headline figure — $35 billion from two of the world's most capitalised alternative asset managers — signals where the biggest pools of non-bank capital are being directed as 2026's AI infrastructure buildout accelerates well past the pace of 2024 and 2025.

Who Is Doing What

Apollo and Blackstone occupy distinct niches within the alternative asset universe, but both have aggressively built out credit and infrastructure franchises over the past half-decade. Apollo's credit platform is among the largest non-bank lenders globally, with a particular emphasis on investment-grade private credit and asset-backed financing. Blackstone's credit and insurance arm — as well as its infrastructure fund complex — gives it parallel firepower.

Broadcom, for its part, has re-positioned itself as a critical AI infrastructure vendor following its $69 billion acquisition of VMware and its growing role supplying custom AI accelerators (XPUs) to hyperscalers. That client base — Google, Meta, and others — has made Broadcom a logical conduit for large-scale infrastructure financing, given the long-dated, contracted revenue streams that underpin the economics of data centre and networking buildouts.

The WSJ report does not specify the precise structure of the arrangement — whether it takes the form of a direct lending facility, an asset-backed securitisation, or a broader infrastructure financing platform — but the involvement of two firms with complementary credit architectures suggests a syndicated, multi-tranche construct designed to absorb capital at a scale no single lender could accommodate alone.

Why Private Credit, and Why Now

The turn to private credit for mega-scale AI infrastructure financing is not accidental. Public debt markets can absorb large investment-grade issuance, but they impose disclosure requirements, syndication timelines, and pricing volatility that bespoke private arrangements can sidestep. For an infrastructure sponsor or technology company executing complex, multi-year deployment programmes, certainty of execution often outweighs the marginal cost advantage of public bond markets.

Private credit's structural advantages are well understood by practitioners: bilateral or club-style deals allow bespoke covenant packages, flexible drawdown mechanics, and, critically, confidentiality around commercially sensitive deployment timelines. In a sector where the race to commission data centre capacity is measured in months, not fiscal years, those attributes carry real premium.

The broader context here is one of asset class evolution. Private credit AUM crossed $2 trillion globally in 2025 by most industry estimates, and the search for deployment opportunities at the top end of the ticket size has been relentless. AI infrastructure — with its long asset lives, contracted offtake from investment-grade hyperscalers, and enormous capital requirements — fits the duration and credit-quality profile that large private credit platforms are engineered to hold.

We have seen this pattern before. In the mid-2010s, as telecom carriers began sweating their tower and fibre assets through sale-leaseback structures, private credit and infrastructure capital stepped in to provide the financing layer that traditional banks, constrained by Basel III risk-weighting rules, were retreating from. The result was a generation of infrastructure deals — many of them now canonical case studies — that locked in long-duration, inflation-linked cash flows for LPs and gave operating companies off-balance-sheet flexibility. The AI infrastructure wave is drawing from the same playbook, only the asset class is GPU clusters and optical interconnects rather than steel towers.

The Scale Question

Thirty-five billion dollars is a number worth anchoring. For context, the leveraged loan market's average large syndicated deal in 2025 ranged between $5 billion and $15 billion for the largest jumbo transactions. Investment-grade bond issuance by a single corporate in a single tranche rarely breaches $10 billion. A $35 billion private credit facility — even if drawn over several years — compresses what would ordinarily be a multi-year public market financing programme into a single bilateral or small-club arrangement.

That compression carries risks for the lender consortium. Concentration is the obvious one: even for firms with hundreds of billions in AUM, a single $35 billion commitment represents a material proportion of any individual fund's capital. The deal's structure will almost certainly involve syndication to insurance companies, pension funds, and sovereign wealth vehicles — the same LP base that has fuelled private credit's expansion — rather than sitting entirely on Apollo's and Blackstone's own balance sheets.

Pricing remains unreported, but in the current rate environment — with the Federal Reserve having cut its policy rate from 2023 highs but still maintaining a positive real rate posture — large private infrastructure credits of this type typically price at spreads of 150 to 250 basis points over SOFR depending on seniority and covenant protections. A basis point is one-hundredth of a percentage point; a 200 basis point spread on $35 billion equates to roughly $700 million in annual interest income for the lender group at full deployment. That is the economic prize being contested here.

What the Broadcom Angle Tells You

Broadcom's centrality in this arrangement is analytically significant. The company's custom silicon business — where it designs XPUs to hyperscaler specification — generates long-term supply agreements that create the contracted revenue visibility lenders require to underwrite large infrastructure credits. Unlike speculative AI software ventures, Broadcom's counterparties are investment-grade or near-investment-grade technology giants with multi-year capex commitments.

That credit quality cascade — from hyperscaler commitment, through Broadcom's revenue stream, down to the infrastructure financing vehicle — is precisely the kind of structural credit enhancement that allows private lenders to deploy at scale without requiring government-backed guarantees or excessive equity subordination beneath the debt.

Implications for the Private Credit Market

If this transaction closes at or near its reported $35 billion size, it will likely recalibrate deal-size expectations across the sector. Competition among large private credit platforms for mega-ticket AI infrastructure mandates has been intensifying since late 2024, with managers including Ares, Blue Owl, and others publicly flagging AI infrastructure as a priority deployment channel.

The more consequential read, however, is what it tells you about the structural shift in corporate financing. Investment-grade companies with contractually underpinned cash flows are increasingly bypassing public bond markets for bespoke private arrangements — not because the bond market is closed to them, but because the flexibility and certainty of private credit better suits the execution tempo of AI infrastructure deployment. That shift has real implications for public credit spreads, for bank balance sheet utilisation, and for the LPs — pension funds, endowments, insurers — whose allocation decisions are driving the channel.

The details of this specific deal remain sparse as of 9 June 2026. Structure, tenor, pricing, and the identity of downstream co-lenders have not been publicly confirmed. But the direction of travel is unambiguous: the largest alternative asset managers are building the financial plumbing for the AI infrastructure era, and the ticket sizes involved are moving into territory that redefines what "large" means in private credit.