Goldman Sachs: Investors worried about rising AI debt, as performance lags
Investor anxiety regarding the explosion of artificial intelligence-related debt is manifesting differently across investment-grade and high-yield markets, according to new research from Goldman Sachs . While big tech firms and emerging AI players have flooded the debt markets this year to fund a massive global data center build-out, the reception has cooled. AI-linked bonds are currently underperforming the broader credit market, and Goldman analysts note that the nature of investor worry depends heavily on the quality of the credit.
In a research note released this week, the bank stated that concerns within the investment-grade (IG) market appear to be issuer-specific. In contrast, anxiety in the high-yield (junk bond) market is viewed as a broader, sector-wide concern.
"Untested" Execution Risks
The underperformance comes as investors weigh the risks of construction delays and unproven revenue models against the massive capital expenditures required for AI infrastructure .
Al Cattermole, fixed income portfolio manager at Mirabaud, told Reuters that as of November 25, his team had chosen not to invest in recent AI-linked investment-grade or junk bonds. Mirabaud, which manages approximately $37.4 billion (30 billion Swiss francs), cited opaque contracts and execution risks.
"Until we see data centers being delivered on time and on budget and providing the computing power that they are intended to—and there still being the demand for it—it is untested," Cattermole said. "And because it's untested, that's why I think you need to be compensated like an equity... not debt."
Selecting the "Haves and Have Nots"
Despite the jitters, some managers see the volatility as a chance to buy into high-quality names.
Christopher Kramer, portfolio manager on the investment-grade credit team at Neuberger Berman, noted that the structural shift to AI-linked debt creates opportunities for selective investors. He described the market as dividing into "haves and have nots."
"Doing the credit work on the 'on balance' sheet versus 'off balance' sheet continues to be paramount," Kramer said. "Security selection is going to be the name of the game and we're excited just from the standpoint that the market's changing."
Kramer did not comment on whether Neuberger, which manages $558 billion in assets, had participated in recent top-tier AI bond issuances.
The market skepticism coincides with warnings from regulators. The Bank of England issued a warning this week stating that the growing role of debt in the AI infrastructure boom could heighten financial stability risks if valuations were to correct.
While Goldman Sachs maintains that the overall credit market looks fundamentally healthy, the widening gap between AI-related debt and the broader market suggests that for debt investors, the AI hype cycle is facing a reality check.
In a research note released this week, the bank stated that concerns within the investment-grade (IG) market appear to be issuer-specific. In contrast, anxiety in the high-yield (junk bond) market is viewed as a broader, sector-wide concern.
"Untested" Execution Risks
The underperformance comes as investors weigh the risks of construction delays and unproven revenue models against the massive capital expenditures required for AI infrastructure .
Al Cattermole, fixed income portfolio manager at Mirabaud, told Reuters that as of November 25, his team had chosen not to invest in recent AI-linked investment-grade or junk bonds. Mirabaud, which manages approximately $37.4 billion (30 billion Swiss francs), cited opaque contracts and execution risks.
"Until we see data centers being delivered on time and on budget and providing the computing power that they are intended to—and there still being the demand for it—it is untested," Cattermole said. "And because it's untested, that's why I think you need to be compensated like an equity... not debt."
Selecting the "Haves and Have Nots"
Despite the jitters, some managers see the volatility as a chance to buy into high-quality names.
Christopher Kramer, portfolio manager on the investment-grade credit team at Neuberger Berman, noted that the structural shift to AI-linked debt creates opportunities for selective investors. He described the market as dividing into "haves and have nots."
"Doing the credit work on the 'on balance' sheet versus 'off balance' sheet continues to be paramount," Kramer said. "Security selection is going to be the name of the game and we're excited just from the standpoint that the market's changing."
Kramer did not comment on whether Neuberger, which manages $558 billion in assets, had participated in recent top-tier AI bond issuances.
The market skepticism coincides with warnings from regulators. The Bank of England issued a warning this week stating that the growing role of debt in the AI infrastructure boom could heighten financial stability risks if valuations were to correct.
While Goldman Sachs maintains that the overall credit market looks fundamentally healthy, the widening gap between AI-related debt and the broader market suggests that for debt investors, the AI hype cycle is facing a reality check.
Next Story