The Drawdown

March 23, 2026 · Duration Team

← Back to Insights

The Topline

Macro

The SaaS model is being stress-tested in real time: AI agents are eliminating the human seats that monthly recurring revenue is built on, forcing companies to reprice or rebundle — and the venture debt market, which has underwritten billions against SaaS ARR, is watching closely.

Structure

Two major lending platform moves advanced in February — P10's $250M acquisition of Stellus Capital and the continuing Horizon/Monroe merger — continued consolidation in private credit.

Risk

Institutional capital is positioning for credit stress — Oaktree raised a record $2.4B first close on its Special Situations Fund IV in February, per Bloomberg, with a target of approximately $5B; simultaneously, Blue Owl completed a $1.4B secondary sale of private credit assets under redemption pressure.

Duration Insights

The venture debt market has spent the better part of a decade underwriting against SaaS ARR as the primary indicator of creditworthiness. The logic was defensible: recurring revenue from diversified enterprise customers was sticky, predictable, and relatively insulated from macro shocks. That logic is now being tested in a way it hasn't been before.

AI agents are not a future risk. They are actively eliminating the human seat licenses that SaaS ARR is built on. A company that counts 10,000 per-seat licenses as its revenue base is already having conversations with its largest customers about renegotiating structure. The outcome — whether repricing, rebundling around outcomes, or contract renegotiation — compresses ARR in the near term even for businesses with strong underlying utility.

For venture lenders, the underwriting question has shifted. Historical ARR growth as a proxy for business quality is insufficient. What matters now is how defensible the revenue is against AI-driven seat reduction, and whether the borrower has a credible path to a new pricing model that maintains or improves their revenue capture.

Two Underwriting Pillars Worth Revisiting

  • Harvest vs. run-off analysis: For SaaS lenders, separating the portion of ARR that is at structural risk from the portion that is durable has become a first-order underwriting input, not a scenario exercise.
  • LTV analysis on ARR: Loan-to-value calculations anchored to ARR need stress scenarios that include AI-driven churn. A business carrying $5M in ARR against a $3M loan looks different if 30% of that ARR is seat-based and the top three customers are running AI pilots.

The opportunity sits on the other side of that risk. Companies that are genuinely AI-native — building their own AI leverage into cost structure and product development — are reaching milestones at a fraction of the burn that the prior generation required. The venture debt structures that work for those businesses are different: smaller checks, faster paths to profitability milestones, and potentially shorter duration. Lenders willing to adapt their credit box to the new operating economics of AI-native businesses will find a less competitive market with borrowers who are structurally more capital-efficient.

The macro signal worth tracking: Oaktree's $2.4B first close on a Special Situations fund is not noise. Institutional allocators of that caliber do not raise at that scale without conviction that credit stress is coming. Blue Owl's $1.4B secondary sale at 99.7% of par tells a different story — that existing positions remain high-quality, but that redemption pressure from LPs is creating forced liquidity events. Both data points can be simultaneously true, and both are relevant to how lenders think about the next 18 months.

Recent Debt Financings

HealthTech & Biotech
HardTech & CleanTech
PropTech & E-Commerce
EdTech & FinTech
SaaS & CyberSecurity

Platform & Capital Moves

Consolidation
Fundraising & Capital Formation
Market Expansion & Launches

Recommended Reading

Supercruise Capital

AI: The Ultimate Bootstrapping Multiplier

AI-first companies are reaching significant ARR per employee at early stages — changing the capital efficiency assumptions that venture debt structures are built around.

Euclid Ventures

The Fundability Trap

Non-consensus seed investments produce 50x+ outcomes at nearly 4x the rate of highly-funded counterparts. A case for reconsidering what strong early traction actually signals.