Skip to content
§ 06 — Insights

Field notes on the post-bubble AI workout.

Analysis of the reconciliation gap between committed AI capital and the record required to retain, restructure, impair, terminate, or sell it. Written for boards, audit committees, CFOs, restructuring advisors, and investors carrying distressed AI positions.

§ INDEX

Articles

  1. § 06.aThe Slopification Crash: How Rushed Deployment and Usage-Based Pricing Broke the Numerator and the DenominatorMIT says 95 percent of generative AI initiatives returned nothing. Deloitte says only 10 percent of organizations saw significant returns on agentic AI. The crash was not a model failure. It was a governance failure, compounded by consumption pricing, over a portfolio no one kept records on.July 28, 2026 · 9 min
  2. § 06.bImpair, Terminate, or Restructure? A Framework for Distressed AI InvestmentsOnce the AI record has been reconstructed, six workout dispositions are available. The choice is dictated by what the evidence can actually support — not by the sponsor's preferred outcome.July 21, 2026 · 6 min
  3. § 06.cWhen the Model Changes and No One Documents It: The Provenance ProblemAI systems in production are not the systems that were approved. Vendors deprecate models, swap defaults, and revise prompts on their own schedule. Without a provenance record, the asset cannot be valued or unwound.July 17, 2026 · 7 min
  4. § 06.dShadow AI Spend: The Cloud, License, and Labor Costs No One ReconciledApproved AI initiatives are only the visible fraction of enterprise AI expenditure. The unallocated portion — cloud, seat licenses, embedded features, and human rework — often exceeds it.July 14, 2026 · 6 min
  5. § 06.eAnatomy of a Failed Enterprise AI Rollout: Five Recurring Failure ModesKlarna, McDonald's, Air Canada, Zillow, and IBM Watson Health each produced high-visibility AI reversals. The individual stories differ. The evidentiary gap they share is the same.July 10, 2026 · 8 min
  6. § 06.fThe $30–40 Billion ROI Gap: Why Enterprise AI Spend Isn't Producing ReturnsRoughly $30–40 billion has been committed to enterprise generative AI. Independent research finds that 95 percent of the pilots show no measurable impact on the P&L. The gap is not a technology problem. It is a reconciliation problem.July 8, 2026 · 7 min
Browse by tag
Case Studies · 2Governance & Provenance · 4ROI & Capital Loss · 4