Private credit managers spent a decade arguing that their asset class was insulated from the volatility of public markets. Opaque NAVs, infrequent marks, private workout mechanics—all of it was framed as a feature. In the first half of 2026, the same opacity is generating the redemption pressure they were designed to avoid.

The Underwriting Error That Is Becoming Visible

Between 2022 and 2024, private credit funds lent heavily to mid-market enterprise software companies owned by PE sponsors. Leverage multiples ran at six to eight times EBITDA in many cases. The underwriting thesis was that subscription-based software revenue was durable, predictable, and therefore serviceable at high leverage. That thesis made sense in a world where the software market grew at historical rates.

The AI disruption question is whether horizontal application software—the category most represented in these portfolios—can sustain those revenue assumptions through 2027 and 2028. Generative AI tools are beginning to replicate functions that mid-market software companies charge monthly per seat. Enterprise buyers do not need to switch vendors en masse to cause meaningful revenue pressure. A 15% seat reduction across a large enterprise client base is enough to push leveraged borrowers toward covenant stress.

Insurance Capital at the Foundation

Eileen Appelbaum of the Center for Economic and Policy Research traced the capital structure that amplifies this risk in an April 2026 analysis. PE firms acquired life-insurance and annuity businesses over the prior seven years, creating a flow of policyholder reserves into proprietary credit funds. The combination of insurance-scale capital, limited disclosure requirements, and high-leverage software lending created a structure that concentrates risk without surfacing it clearly to the ultimate capital providers—policyholders and pension funds that allocated through the insurance intermediary.

The current LP redemption wave sits one level above the insurance capital: institutional investors and family offices that allocated directly to the private credit funds. But the structural point holds: the information flow is poor at every level of the stack.

March and April Gate Announcements

Two perpetual private credit vehicles capped quarterly withdrawals in March 2026. A third followed in early April. None disclosed material credit losses alongside the gate announcements. The market interpretation—secondary buyers assigning discounts above stated NAV—reflects the view that marks have not yet moved to where they will eventually settle if software borrower revenues decline.

The Differentiated Portfolio

Risk concentration within the asset class follows the 2022–2024 vintage lending patterns. Portfolios that funded horizontal application software at high multiples carry the most immediate exposure. Portfolios that concentrated in infrastructure software, vertical SaaS with deep workflow integration, or asset-backed lending outside of software face a categorically lower AI-displacement risk. The aggregate private credit statistics do not separate these portfolios, which is part of why LP-level analysis is difficult.

Fund managers defending the asset class point to the structural differences that private direct lending provides relative to public high-yield: negotiated covenants, private workouts, no forced-sale mechanics. These are genuine advantages in a stress scenario. They are advantages that have not been tested against a broad, sector-specific credit event in the current configuration of the market—PE ownership plus insurance capital plus high-leverage software lending is a combination without a historical analogue.

The next two quarters of NAV prints from the largest perpetual vehicles will be the first externally observable signal. LP letters that begin disclosing AI-displacement metrics will signal that the LP community has demanded enough specificity to change what managers put in writing. Both signals are quarters away. Until then, the redemption queue is the most current data available.

Source: Private Credit Fund Redemptions Climb Sharply, Some Caps Now in Place