enterprise-AI
8 articles
- The Conviction-Weighted Restructuring of Portfolio Exposure
The 5% hurdle-rate regime sorts capital by asset duration, financing structure, and pricing power. Conviction-weighted portfolio restructuring captures structural rents — nuclear uprate rent, energy delivery premium, defense industrial base — while hedging the mechanical repricing risk of concentrated technology valuations. Calibrated probabilities for a 10–15% correction, a systemic bear market, and multiple persistence are illustrated.
- PepsiCo’s Industrial AI as a Working-Capital Program
PepsiCo’s industrial AI program creates value through working-capital efficiency, service levels, scheduling, and capex relief. The durable economics concentrate in the network planning-and-execution layer.
- The Mechanical Repricing of Concentrated Tech Valuations
The U.S. equity market is in a mechanical repricing cycle. The convergence of the 2027 earnings-growth cliff, the discount-rate regime shift, and the AI capex circularity will compress concentrated tech valuations by 10–15% within the November 2026 – May 2027 window. The 20%+ tail is gated by the AI capex → revenue conversion.
- B2B AI Adoption Metrics Measure Experimentation, Not Revenue
B2B AI surveys measure task-level experimentation rather than scaled enterprise workflows. Revenue depends on paid contracts, production deployment, usage depth, and renewal, so adoption-based TAM models can overstate the sustainable market by up to 2.2×.
- The Agentic Extinction of Middle Management: Why AI Agents Are Replacing the Coordination Layer
The 2026 restructuring wave is hitting middle management, not just entry-level roles. The reason is structural: coordination work is repetitive, rule-governed, and embedded in software systems, making it a natural target for agentic AI.
- The AI Empirical Grounding Bottleneck
Open-weight foundation models have collapsed the intelligence generation layer to near-zero marginal cost. The structural bottleneck is migrating to the verification layer — where search engines, financial data aggregators, and content publishers are converting their historical indexing moats into paid API tollbooths. This is not a single monopoly rent; it is a fragmented patchwork of metered, unevenly-priced access gates whose integration complexity is the real constraint.
- The AI Compliance Tax: Why Tail Risk Scales Faster Than Governance Buildout
AI deployment creates a real compliance layer — but the base-rate cost and the tail risk are wildly mismatched in maturity. Governance spend is manageable; uninsured liability is where the real economics live.
- The AI Implementation Tax: Where Enterprise AI Dollars Actually Go
Enterprises spent $684 billion on AI in 2025, and $547 billion produced no measurable result. The money is going to the wrong places — idle compute, tool sprawl, consulting markup — while actual return drivers get starved.