AI Can Tell Us Something About Credit Market Weakness

What It Covers

Dan Wertman of Noetica AI explains how AI analyzes credit deal terms to reveal market anxiety, tracking structural protections that signal creditors preparing for distress amid private credit blow-ups and complex AI infrastructure financing.

Key Questions Answered

• •Anti-distress protections surge: Anti-SIRTA lien subordination terms jumped from 61% to 84% of deals in Q3 2024, the highest quarterly increase ever recorded, indicating lenders prioritize recovery position over preventing liability management exercises when bankruptcy occurs.
• •Off-balance sheet financing risk: FirstBrands disclosed $11 billion in total obligations versus $5-6 billion reported debt through receivables financing facilities, enabled by deep private credit markets. This structure mirrors concerns about AI infrastructure deals with 90% leverage ratios on immature assets.
• •EBITDA add-back flexibility peaks: Cost savings add-backs now appear in 64% of deals at highest levels recorded, with 51% allowing add-backs above 20% of EBITDA. Borrowers gain economic flexibility to weather uncertainty while lenders secure structural protections in simultaneous fortification.
• •AI semantic analysis advantage: Language models attribute semantic meaning to deal terms phrased thousands of different ways across documents, tracking precedent from billions of terms mapped to deal characteristics. Traditional control-F searches cannot detect innovation in clause structure or long-range dependencies across sections.