Goldman Sachs’ high-yield AI datacenter basket is now trading at a spread of 353 basis points, wider than where it sat back in June 2022. Broader AI-related bond issuance expected across 2025 and 2026 carries an option-adjusted spread of 381 basis points, a meaningful premium over the wider high-yield market.
The numbers behind the nervousness
Goldman launched its high-yield AI issuer basket in July 2026, comprising 18 equal-weighted US high-yield names. At inception, the basket carried an average spread of 319 bps and a yield of 7.45%. For context, the broader high-yield market was sitting at 267 bps and a yield of 7.3% at the time.
Goldman’s AI leadership basket, a separate instrument tracking investment-grade AI-adjacent credits, has seen its spread widen from 74 bps to nearly 150 bps over the past 12 months.
The basket includes names like CoreWeave alongside various joint venture debt structures. Of the 23 recent datacenter joint-deal ventures Goldman tracks, 17 are now trading wider than their origination yields. New-issue concessions on large deals have widened by up to 20 bps, as demand for longer-dated tranches has cooled.
Half a trillion dollars of AI debt
Goldman estimates that close to $500 billion of AI-related debt will be issued in 2026 alone, representing approximately 18% of total US investment-grade supply.
Goldman has responded to this dynamic by creating tradable instruments, custom baskets and swaps, that let investors express views on AI credit without needing to buy individual bonds.
What the spread widening actually means
The distinction between the datacenter basket at 353 bps and the broader 2025-2026 AI issuance OAS of 381 bps is worth noting. The wider spread on forward issuance suggests that the market expects conditions to get tighter, not looser, as more supply hits.
For the companies issuing this debt, wider spreads translate directly into higher borrowing costs. A datacenter operator that could have financed a billion-dollar facility at 7.45% a few months ago is now looking at meaningfully higher rates. At the scale of the AI buildout, even 30-40 bps of additional spread across hundreds of billions in issuance represents billions of dollars in incremental interest expense over the life of these bonds.
Hyperscalers like Microsoft, Google, and Amazon can finance datacenter construction off their investment-grade balance sheets at far tighter spreads. Independent operators and joint ventures, the names populating Goldman’s HY basket, don’t have that luxury.