NextFin News - The insatiable appetite for artificial intelligence infrastructure is beginning to warp the fundamental risk assessments of the corporate bond market, according to Robert Cohen, director of global developed credit at DoubleLine Capital. Speaking at the Bloomberg Global Credit Forum in New York on Wednesday, Cohen warned that the credit markets are on the precipice of an "AI bubble" as companies rush to secure massive debt financing for technology whose ultimate profitability remains opaque.
Cohen, who joined DoubleLine in 2012 and oversees the firm’s corporate credit team, has long maintained a reputation for rigorous fundamental analysis, often leaning toward a cautious, value-oriented stance in developed credit markets. His latest assessment suggests that while the "credit party" is currently in full swing, the lack of visibility into returns on capital for AI-related debt issuance is creating a dangerous disconnect between market pricing and underlying corporate health.
The surge in borrowing comes at a time when corporate credit supply has actually shrunk since the pandemic, creating a supply-demand imbalance that has kept spreads deceptively tight. Cohen noted that many issuers are not breaking out AI-specific profitability in their financial statements, making it nearly impossible for creditors to determine if the billions being deployed into data centers and hardware will generate the cash flow necessary to service the debt. This lack of transparency is particularly acute among companies that are pivoting their entire business models toward AI integration without a proven track record of monetization.
This skeptical view is not yet the consensus on Wall Street, where many sell-side analysts continue to project that AI will drive a multi-year productivity boom capable of supporting higher leverage ratios. Major investment banks have largely maintained "overweight" ratings on the technology and communications sectors, citing the transformative potential of generative AI as a structural tailwind that justifies current valuations. Cohen’s warning represents a minority, more defensive position that prioritizes balance sheet integrity over speculative growth narratives.
Cracks are already appearing in related segments of the market. Cohen pointed to the syndicated bank loan market as a "window" into the broader credit landscape, noting that software-related loans are currently trading roughly 10 points below the broader index. Furthermore, loans rated CCC—the lowest tier of junk debt—have declined 8% this year, suggesting that investors are beginning to penalize the most leveraged players even as headline investment-grade spreads remain narrow.
The risk of a bubble is further complicated by the rise of private credit, a market Cohen described as having "growing cracks." With roughly 90% of the private credit market rated B3 or lower, the sector is highly sensitive to any slowdown in the AI-driven capital expenditure cycle. If the anticipated returns on AI investments fail to materialize within the next 18 to 24 months, the credit market could face a sharp repricing as the "insatiable" demand for debt suddenly evaporates, leaving highly leveraged issuers exposed to a liquidity crunch.
Explore more exclusive insights at nextfin.ai.
