KBRA Releases Research – Private Credit: Framing AI and Software Risk
5 Feb 2026 | New York
KBRA releases research presenting data and observations to help frame the potential risks artificial intelligence (AI) may pose to the direct lending landscape, in the context of recent market volatility.
The analysis focuses on underlying middle market (MM) borrowers whose loans are held in MM collateralized loan obligations, business development companies (BDC), rated feeder notes into direct lending funds, and recurring revenue loan asset-backed securities. Unlike research that relies on “as reported” sector classifications and debt totals, this report leverages KBRA’s access to years of financial statements, lender investment memoranda, and credit agreements for more than 2,400 sponsor-backed MM companies in its assessment portfolio. This proprietary dataset enables KBRA to determine industry classifications, financial performance, and debt levels with greater precision.
Even with this level of precision, the software sector remains highly heterogeneous, making broad, sector-wide stress assumptions inappropriate. Instead, credit analysis should focus on each company’s specific end markets, competitive position, and defensibility.
Key Takeaways
- Software is the second-largest sector in KBRA’s assessment portfolio, representing 17% of borrowers by count and 22% ($224 billion) of total debt exposure. Despite this concentration, the sector is entering the AI transition from a position of relative credit strength.
- KBRA believes its sector analysis provides a more consistent view of private credit exposures. Unlike other publicly reported sources—which rely on manager-defined sector labels and reflect only a portion of total loan exposure, KBRA assigns sector classification using a consistent framework with direct visibility into companies’ business models. This enables a more precise comparison across the private credit landscape.
- The software sector is likely the most exposed to AI risks; however, exposure extends beyond technology companies. Companies in other industries, notably Health Care Services and Technology, Commercial and Professional Services, and Media are also facing AI-related disruption.
- AI-enabled alternatives may pressure pricing, retention, and margins, regardless of whether the borrower is formally classified as a software company, making a product’s competitive moat crucial. Obligors with proprietary data and deep integration with enterprise workflows tend to exhibit greater stickiness, which can support defensibility.
- Even well-positioned investors may face exit or refinancing risk if market uncertainty around AI delays realizations or depresses valuations, creating incremental pressure on returns. For lenders, valuation-dependent sponsor support still appears present. We are closely monitoring the situation as liquidity injections act as the primary deterrent to defaults and losses for lenders.
- Amid recent market volatility and heightened commentary, attention has drifted from opportunities AI presents for borrowers—particularly in productivity and margin improvements, which we believe remain in the early innings. Effective AI deployment among direct lending obligors has already delivered tangible margin and revenue benefits from automation. These gains suggest companies that combine AI capabilities with proprietary data and embedded customer relationships may enhance scalability and profitability—although it remains too early to assess the durability for each company.
Click here to view the report.