Analysis:AI borrowers face tough sell in risky corners of US credit market

AI borrowers face increasingly challenging financing conditions in the riskier segments of the US credit market, according to recent analysis. The artificial intelligence boom has spurred $88 billion in AI-related borrowing by low-rated firms this year, predominantly from US issuers, as reported by Goldman Sachs. However, this represents a significant increase from the $20 billion issued in leveraged finance (mainly through junk bonds and loans) during the first 11 months of 2025.

Investors are scrutinizing these less-established borrowers more closely, questioning their revenue projections, collateral value, and debt support capabilities. Higher-rated AI issuers have been aggressively borrowing, while investors demand higher compensation due to the uncertainty surrounding AI borrowers' cash flow and the potential for significant losses if projections fail to materialize.

Risk appetite for AI-related credit remains subdued, despite rising issuance, according to market participants. Leveraged finance buyers have focused on higher-quality borrowers, such as double-B-rated companies with predictable revenue streams backed by long-term contracts, tangible assets, and established customer bases. Data centers, often meeting these criteria, have been driving demand in the high-yield market.

However, AI companies are less likely to push their credit ratings lower due to higher borrowing costs. Near-investment-grade issuers like SoftBank Group pay yields ranging from 8.625% to 9.75% on various debt maturities, which are typically paid by lower-rated companies. Credit investors demand proof that revenues will eventually support the debt being raised, leading to increased skepticism and cautious investment decisions.