Anthropic has secured a $35 billion private credit financing package arranged by Apollo Global Management and Blackstone to fund AI chip infrastructure, according to reports published June 6, 2026. The structure will finance data-center hardware that will be leased to Anthropic to support its rapidly growing AI workloads.
This article aggregates reporting from 2 news sources. The TL;DR is AI-generated from original reporting. Race to AGI's analysis provides editorial context on implications for AGI development.
A $35 billion compute financing package is not just another debt deal; it effectively pre-purchases a multi‑year supply of frontier AI hardware for Anthropic. In a world where GPUs and advanced accelerators are the scarcest inputs to progress, locking in that capacity signals that private credit is now a primary engine of the AI arms race, not just venture capital or big‑tech balance sheets. It also mirrors earlier infrastructure financing around hyperscaler data centers and specialist clouds like CoreWeave, but at a materially larger scale. ([cls.cn](https://www.cls.cn/detail/2392501))
Strategically, this deal gives Anthropic a clearer line of sight on the compute needed to keep pace with (or briefly outrun) OpenAI, Google, and Meta at the frontier. Instead of raising equivalent sums in dilutive equity, Anthropic can lease capacity via a special‑purpose vehicle while it pursues an IPO and large commercial contracts. That moves alternative asset managers—Apollo, Blackstone and peers—into a pivotal role in determining who gets the chips required for the next generation of models.
The competitive implication is stark: frontier research will increasingly be gated by access to structured, long‑dated financing for hardware. Labs that can convince private credit to underwrite tens of billions in AI infrastructure will enjoy a structural advantage in scaling experiments, safety work, and product launches—regardless of their current profitability.