Limited partners are reaching the same conclusion through different routes: the AI-displacement risk in private credit software portfolios cannot be priced from the outside, so the rational move is to reduce exposure while liquidity remains available. Redemption requests at major funds have climbed through the fourth quarter of 2025 and the first quarter of 2026, and several large managers have responded by capping quarterly outflows.
Insurance Reserves, PE Acquisitions, Software Loans
The structure that created this exposure took about seven years to assemble. Private equity firms acquired life-insurance and annuity businesses, giving them access to policyholder reserves. Those reserves were deployed into proprietary private credit funds with limited mark-to-market discipline and minimal asset-level disclosure. The credit funds lent heavily to mid-market software companies owned by PE sponsors, particularly between 2022 and 2024 when enterprise software valuations and leverage multiples were both running high.
Eileen Appelbaum of the Center for Economic and Policy Research traced this architecture in a detailed analysis published in April 2026. Her focus was on the insulation of the structure from normal market feedback—the combination of thin disclosure, infrequent marks, and policyholder capital creates a system that is slow to register deteriorating credit conditions.
The Disclosure Problem
Fund letters from major private credit managers categorize exposure at the sector level. Software is a line item. The distribution within software—infrastructure versus horizontal application, AI-substitutable versus workflow-locked—is not disclosed. An LP trying to estimate what a 25% revenue decline at horizontal SaaS borrowers does to portfolio NAV is working without the necessary inputs.
That opacity is not a new feature of the asset class. It became a crisis trigger when the risk category shifted. Prior to 2025, the general concern about private credit was cyclical: would borrowers stay current through a recession? The AI-displacement risk is different in character—it is a structural question about whether specific revenue models remain viable, and it affects different borrowers very differently depending on what they sell and to whom.
The March–April Gate Sequence
Two perpetual private credit vehicles imposed quarterly outflow caps in March 2026. A third followed in early April. The gates were announced without any corresponding disclosure of credit losses—the funds are managing against the possibility of future deterioration, not confirmed current losses. Secondary buyers of interests in these funds have assigned discounts that exceed what the NAVs reflect, effectively pricing in mark-downs the funds have not yet taken.
Portfolio Differentiation
The funds that entered this period with the highest exposure to horizontal application software face the most direct LP scrutiny. These portfolios lent to companies whose value proposition depends on software adoption patterns that AI is beginning to replicate or circumvent. Funds anchored in infrastructure software—database management, security, cloud plumbing—or vertical SaaS with regulatory and workflow lock-in are better positioned. Asset-backed books with physical collateral face a categorically different risk profile.
The structural argument made by fund managers—that private credit covenants are tighter, workouts are more controllable, and the forced-sale dynamic of public markets does not apply—is accurate as a description of the mechanics. It is not, however, a test result. The test will come if a sufficient number of software borrowers encounter revenue stress that triggers covenant violations, and the workout process has to perform under that load simultaneously across multiple portfolios.
NAV prints through the second and third quarters of 2026 will supply the first externally observable signal. LP letters that begin disclosing AI-displacement-risk metrics by portfolio segment will indicate that the informal pressure has converted into formal disclosure demand—a shift that, historically, follows rather than precedes the acute phase of a credit concern.
Source: Private Credit Fund Redemptions Climb Sharply, Some Caps Now in Place