Broadcom and Blackstone Lead $60B Debt Package to Fund AI Chip and Data Center Buildout for Anthropic
TREE NEWS reports: Broadcom is spearheading a $60 billion debt financing to help artificial intelligence companies including Anthropic purchase chips and build data centers, in one of the largest debt transactions ever assembled for the AI chip sector. The deal has been in the works for several weeks and has not yet been formally announced.
Broadcom is working with Wall Street banks to arrange the package. Roughly $42 billion is expected to be issued as Class A senior secured bonds, syndicated by banks and distributed to investors. A further $18 billion takes the form of Class B subordinated debt, led by Blackstone, which is anchoring the tranche with $9 billion of its own fund capital and raising the remainder from external syndicate investors.
Two-Tier Structure Splits Risk Between Banks and Private Credit
The senior/subordinated split is the defining feature of the transaction. The senior secured tranche sits at the top of the capital structure and will appeal to insurance companies, pension funds and bond managers hunting for yield in investment-grade-plus credit. The subordinated layer, anchored by Blackstone, absorbs first losses and carries a higher coupon — a structure that shifts meaningful risk onto private credit providers rather than bank balance sheets.
Blackstone’s $9 billion anchor commitment signals that large private capital pools are now willing to underwrite AI infrastructure at scale, effectively acting as a shadow banking system for the buildout. The remaining $9 billion of the subordinated tranche will be marketed to external investors, testing how deep demand runs beyond the anchor commitment.
Why This Matters for Markets
- Credit markets: A $60 billion print would be among the largest AI-related debt raises on record and could tighten spreads if demand is strong, but a weak reception would be an early warning that AI infrastructure financing is nearing saturation. Watch the pricing of the subordinated tranche in particular — it is the cleanest read on investor appetite for AI credit risk.
- Equities: Broadcom’s role as arranger, not just chip supplier, deepens its exposure to the AI cycle. Success reinforces the thesis that AI capex is being financed rather than funded from operating cash flow, which supports semiconductor demand but raises leverage concerns across the ecosystem. Nvidia’s earlier move to mobilize over $500 billion with six major financial institutions, including Blackstone, shows this is now an industry-wide financing model.
- Rates and bonds: Heavy AI-related issuance adds supply to corporate credit markets. If the pipeline of similar deals keeps growing, it could pressure long-dated corporate spreads and compete with Treasuries for institutional allocations.
- Crypto: No direct crypto exposure, but AI-infrastructure debt competes for the same risk capital that has flowed into digital assets. A successful deal could pull marginal capital toward credit and away from speculative assets; a failed one would likely be read as risk-off across high-beta markets.
- Commodities: Data center construction is power-hungry. Financing of this scale supports demand for electricity, copper, cooling equipment and, indirectly, natural gas and uranium as power sources.
- Currencies: Dollar-denominated issuance supports global demand for USD funding, a mild tailwind for the dollar against funding currencies.
Key Takeaways for Investors
- The AI trade is increasingly a credit trade. The marginal dollar funding AI expansion now comes from bond and private credit investors, not equity or retained earnings.
- Blackstone’s anchor role confirms private credit’s arrival as a primary financier of the AI buildout — a structural shift with implications for bank lending, spreads and systemic risk monitoring.
- Public backlash against data center construction is rising, and this deal is a live test of whether investors will keep funding AI expansion amid that scrutiny.
- Watch the subordinated tranche pricing. It is the single best gauge of how much risk the market is willing to take on AI infrastructure right now.
