Goldman Sachs Asset Management is currently underweight on bonds issued by major AI borrowers, or "hyperscalers," as a significant increase in debt issuance is expected in the market. Lindsay Rosner, head of multi-sector fixed income investing at Goldman Sachs Asset Management, noted in a recent interview that they anticipate a substantial volume of new hyperscaler issuance. This strategic underweight reflects concerns about heavy supply potentially repricing the market, rather than a pessimistic view on AI itself. Rosner emphasized that the team believes in the "story of AI" but is mindful of market dynamics.
Big Tech companies, including Amazon, Meta Platforms, and Alphabet, have been major players in the high-grade corporate bond market this year, collectively borrowing hundreds of billions of dollars to fund their artificial intelligence initiatives. This trend has pushed AI-related issuer credit spreads to around 115 basis points, compared to approximately 78 basis points for the broader investment-grade market. Investors are showing increased caution regarding bonds from AI-related companies, preferring industrial and financial issuers, despite the strong credit profiles of these tech giants.
The volume of projected issuance is substantial, with Goldman Sachs estimating hyperscaler gross debt issuance to reach a record $420 billion in 2027, marking a 60% increase from 2026 forecasts. JPMorgan Asset Management also projects significant issuance, with $17 billion in 2024, $109 billion in 2025, and $194 billion in the first half of 2026 alone. This rapid increase in borrowing is causing concern among investors, who are now demanding higher compensation for the growth, capital expenditure intensity, and increasing off-balance-sheet exposures associated with these companies' AI buildouts. The ratio of bids to offerings for new hyperscaler bonds has also declined, indicating a softening in investor demand relative to the rising supply.
