The private credit industry is entering a stress test it has not previously encountered. The 2008 and 2015 credit stress events that shaped public high-yield risk management were cyclical—driven by economic contraction and commodity price collapses that played out over known timeframes. The AI-displacement scenario now pressuring private credit software portfolios is structural, with an uncertain timeline and asymmetric impact across sub-categories that fund disclosures do not clearly separate.
The Setup: Seven Years of Capital Redeployment
The exposure was assembled through a deliberate and documented capital strategy. Over the seven years ending in 2025, large PE firms acquired life-insurance and annuity businesses and channeled the resulting policyholder reserves into proprietary private credit funds. Those funds, operating with limited disclosure and infrequent marks, deployed capital into PE-owned portfolio companies. The concentration in mid-market application software companies during 2022–2024 reflected the thesis that SaaS subscription revenue was among the most credit-friendly collateral available to private lenders.
Eileen Appelbaum of the Center for Economic and Policy Research documented this chain in an April 2026 analysis focused on the structural risks it creates: the disclosure gap between the insurance entity and the underlying loan portfolio, the conflict of interest in PE firms setting marks on loans to their own portfolio companies, and the insulation the structure provides from normal market feedback mechanisms.
AI as a Novel Credit Variable
The AI-disruption thesis for software is not uniformly distributed across the category. Infrastructure software—the code that runs databases, security systems, and cloud infrastructure—is not a near-term AI-substitution target. The enterprise market is not replacing its database software with LLM-generated alternatives. Vertical SaaS with deep workflow integration and regulatory dependencies—industry-specific tools in healthcare, legal, or financial compliance—has switching costs that insulate it from rapid AI displacement.
The category at the center of the concern is horizontal application software: productivity suites, CRM platforms, project management tools, document automation. These are products where AI functionality is already technically competitive and where enterprise buyers’ cost calculations are shifting. A borrower in this category that took on six-times EBITDA leverage in 2023 may face very different revenue dynamics in 2026 and 2027 than the underwriting modeled.
Gates, Discounts, and What Comes Next
Two perpetual private credit vehicles imposed quarterly outflow caps in March 2026. A third followed in April. None of the three disclosed material credit impairment. Secondary buyers of interests in these funds have established discounts above stated NAVs—pricing the AI-displacement scenario at a probability the funds themselves have not reflected in their marks. Each gate announcement widened the secondary discount and triggered additional redemption activity from LPs watching the queue grow.
The Structural Advantage That Has Not Been Tested
Private credit managers have, across multiple communications, emphasized the structural features that distinguish their asset class from public bond markets in a stress scenario: directly negotiated covenants with tighter terms, private workout processes that avoid the forced-sale dynamics of distressed-bond markets, and the ability to restructure quietly without public market signaling. All of these are genuine characteristics of the direct lending structure.
They have not been tested in a scenario where the stress is sector-specific, AI-driven, and potentially persistent rather than cyclical and recoverable. A software borrower that faces structural revenue decline driven by AI substitution does not recover when economic conditions improve, the way a cyclical manufacturing borrower does. The workout process for that kind of borrower is categorically different—closer to a business model transformation than a covenant cure.
What the next two quarters provide: NAV prints that will show whether fund managers begin moving marks to reflect AI-displacement risk before LP pressure forces them to, and LP letters that will indicate whether the asset class is moving toward sub-category AI-risk disclosure. Both answers will define how this stress cycle proceeds—and how much credibility the private credit structure retains with the institutional LP base that funded its expansion.
Source: Private Credit Fund Redemptions Climb Sharply, Some Caps Now in Place