Private credit funds face elevated risks in 2026 from two distinct but overlapping AI-related channels

Private credit funds face significant 2026 risks from two AI-related channels: disruption to software borrowers and demand shortfalls in AI infrastructure. These are compounded by liquidity mismatches, sector concentration, and ties to insurers and pensions. Non-bank direct lending, via vehicles like BDCs, has heavily expanded into both areas, heightening vulnerability.

Two Primary AI-Related Risk Channels

Private credit faces two primary AI-related risk channels. First, substantial exposure to software and SaaS loans, estimated at 20-30% of many BDC portfolios and over $500 billion in outstanding loans by end-2025, is threatened by generative AI’s potential to automate or displace traditional software models. Loans originated before widespread AI awareness often didn’t price this disruption risk, while credit spreads have narrowed, reducing loss buffers.

Rising payment-in-kind interest signals stress, and default rates have increased (peaking near 5.8% in early 2026), with projections for software portfolios rising further as maturity walls approach in 2027-2028. Shared borrower pools across large BDCs amplify potential correlated losses.

Second, private credit is a major capital provider for AI data centers, SPVs, and neocloud providers, accounting for over a third of deals in 2025. Risks include utilization shortfalls if AI monetization lags, power supply constraints or construction delays, rapid GPU/ hardware obsolescence (2-4 year economic lives vs longer loan terms), high concentration on few tenants (e.g., Meta, OpenAI), and renewal risk where loan maturities exceed customer contracts.

The Financial Stability Board has warned that a sharp correction in AI asset valuations could produce “sizeable” credit losses for private credit investors.

Liquidity and Redemption Pressures

Semi-liquid private credit funds faced high redemption pressure in 2026, with firms like Blue Owl and Blackstone’s BCRED seeing large requests or gates triggered. These funds hold illiquid loans while promising periodic redemptions, but cash buffers and repayments proved insufficient for sustained outflows, leading to asset sales, borrowing, or delays. Slowing inflows during rising outflows created feedback loops, marking the first major stress test for this sector.

Structural and Interconnected Risks

Private credit’s growing role in AI financing introduces structural and interconnected risks. Heavy sector concentration in tech, software, and AI infrastructure, combined with overlapping exposures across funds, heightens correlated risk. Valuation challenges arise from illiquid loans and complex special purpose vehicle (SPV) structures, which may lag economic reality. Links to insurers and pensions amplify concerns: private credit held by life insurers (including PE-affiliated platforms) and pension funds means impairments could pressure reserves or capital.

Some analyses note potential policyholder withdrawal risks and the role of state guaranty funds, which socialize losses differently than bank deposit insurance, with Michael Burry flagging this as a possible contagion path from an AI downturn. Bank linkages create indirect transmission channels through funding lines to private credit funds and overlapping borrowers. Collateral and recovery risks are significant: GPU-backed or data-center loans face secondary-market flooding if multiple projects stress simultaneously, potentially lowering recoveries.

Regulatory bodies, BIS, FSB, ECB, and Bank of England have all highlighted private credit’s expanding role in AI financing, emphasizing the need for monitoring both idiosyncratic and systemic risks. Losses are expected to first impact private credit funds and their limited partners (pensions, endowments), rather than directly affecting regulated banks, thus limiting classic banking-crisis dynamics. However, this still poses meaningful investor and confidence risks, as impairments could cascade through interconnected financial systems.

Private credit is undergoing its first major stress test due to AI-driven disruption to legacy software portfolios and rapid AI infrastructure lending expansion. Mitigating factors include lower leverage than pre-2008 banks, secured lending with tighter documentation, and strong counterparties in infrastructure deals. However, manager skill varies widely, with disciplined managers likely faring better. Digital infrastructure can still offer attractive returns if utilization holds. Overall, liquidity pressures in semi-liquid vehicles, rising defaults, payment-in-kind usage, concentration risks, and warnings from the FSB and BIS indicate material vulnerabilities. An AI demand shortfall or accelerated software disruption could cause significant losses concentrated in private credit, with secondary effects on insurers, pensions, and market confidence. Key indicators to monitor include redemption queues, gating activity, software-sector default and recovery rates, PIK loan shares, GPU secondary pricing, data-center utilization metrics, fund leverage, cash buffers, and insurer disclosure. Differentiation by manager and portfolio composition is critical, as broad private-credit exposure carries elevated AI-related risks.