AI-driven software disruption threatens Golub Capital BDC's private‑credit portfolio

AI-driven software shift pressures Golub’s private-credit bets

Golub Capital BDC faces rising risk as the latest wave of generative AI tools threatens the cash flows of enterprise software companies that make up a sizable slice of business development companies’ lending activity. The unveiling of advanced models by Anthropic is accelerating concerns that incumbents’ software offerings—long a primary revenue source for many middle‑market borrowers—could be supplanted or commoditized, eroding the predictable cash flows that private lenders rely on for repayment and covenant cushion.

The threat is acute for BDCs and other private-credit managers that favour unitranche and other illiquid loan structures, which can be slower to reprice or enforce protections when borrower performance deteriorates. Market watchers note that since 2020 enterprise software has been a preferred private‑credit sector, and portfolios concentrated in that area may face greater stress if AI adoption outpaces borrowers’ ability to adapt. Industry observers warn that loan extensions, liquidity shortfalls and fragile covenant stacks that predate the current AI shock could amplify losses in an adverse scenario.

For Golub Capital BDC, which focuses on middle‑market direct lending where unitranche and first‑lien structures are common, the development prompts immediate underwriting and portfolio‑management questions. The firm is likely reassessing sector exposures, revisiting covenant protections and conducting stress tests that factor in faster revenue erosion from AI substitution. Credit officers and risk teams across BDCs are expected to scrutinize valuation assumptions, renegotiation risk and syndication prospects for deals where software revenue is central to debt service.

Scale and default scenarios

Private credit is large and concentrated: roughly a $3 trillion market in which software accounts for about 17% of U.S. BDC investments by deal count, second only to commercial services, according to PitchBook. UBS models show that in an aggressive disruption scenario U.S. private‑credit default rates could climb to about 13%, materially above current leveraged‑loan and high‑yield baseline estimates, underscoring the systemic nature of the risk.

Anthropic’s move accelerates substitution risk

Analysts say Anthropic’s models are designed to perform complex professional tasks that many incumbent vendors charge for, raising the prospect that adoption could outpace borrowers’ ability to adapt revenue models. That dynamic puts pressure on valuations, covenant protections and the liquidity of unitranche and other illiquid loans that underwrite leveraged buyouts financed by private credit. Experts caution the sector may now face a fresh stressor on already fragile footing.