---
id: "contrarian-moat-workflow-not-tech"
type: "contrarian-insight"
source_timestamps: ["§ 5. The AI-driven flywheel."]
tags: ["competitive-strategy", "moats", "contrarian-insight"]
related: ["concept-ai-driven-flywheel", "entity-org-anterior"]
challenges: "The conventional view that proprietary AI models or algorithms are the primary source of competitive advantage for AI startups."
enrichment_verdict: "Supported as a strategic pattern; but the moat is strong only in messy, idiosyncratic domains and weaker where open standards / agent portability / regulation force interoperability."
sources: ["futures"]
sourceVaultSlug: "hbr-seg-futures"
originDay: 2
articleStem: "hbr-new-24-agentic-ai-supercharges-startups"
sourceUrl: "https://hbr.org/2026/07/how-agentic-ai-supercharges-startups-and-threatens-incumbents"
sourceTitle: "How Agentic AI Supercharges Startups and Threatens Incumbents"
---
# AI Moats Are Built on Workflow Expertise, Not AI Models

**Contrarian insight.** Conventional thinking assumes the competitive advantage in AI lies in having the best *foundational model*. The authors argue that customized AI interfaces are no longer scarce — the true competitive **moat** lies in the deep operational and workflow expertise gained by solving intractable manual tasks for enterprises, which creates high switching costs.

This is the strategic reading of the [[concept-ai-driven-flywheel]]; [[entity-org-anterior]]'s prior-authorization workflow (turning 600-page faxes into structured data) is the exemplar.

**What it challenges:** the assumption that proprietary AI models or algorithms are the primary source of competitive advantage for AI startups.

**Enrichment note.** Strategy discussions (from AI-focused VCs and researchers) agree that as foundation models commoditize, moats shift to distribution, data, and workflow integration. *Verdict: Supported as a pattern.* **Counter-perspective:** workflow moats may be weaker where standards and open APIs let competitors plug into the same systems, where customers demand agent portability to avoid lock-in, or where foundation-model providers push horizontal agent platforms. The moat is strongest in messy, idiosyncratic domains (like prior authorization) and weakest where processes are standardizable and regulation forces interoperability.


## Related across articles
- [[framework-moat-evolution]]
- [[action-secure-proprietary-data]]
- [[claim-moat-vulnerability]]