---
id: "concept-first-line-defense-shift"
type: "concept"
source_timestamps: ["§ Integrating the Framework"]
source_url: "https://hbr.org/2026/05/what-are-your-companys-ai-nightmares"
source_title: "What Are Your Company's AI Nightmares?"
tags: ["organizational-design", "risk-boards", "bottlenecks"]
related: ["concept-enc-teams", "action-repurpose-risk-boards"]
definition: "The transition of primary AI risk management from centralized risk boards to decentralized, project-level cross-functional teams."
sources: ["governance"]
sourceVaultSlug: "hbr-seg-governance"
originDay: 7
articleStem: "hbr-cl-82-ai-nightmares"
sourceUrl: "https://hbr.org/2026/05/what-are-your-companys-ai-nightmares"
sourceTitle: "What Are Your Company’s AI Nightmares?"
---
# Shifting the First Line of Defense

A structural change in *where* AI risk is first caught inside an enterprise.

**Standard model:** A centralized AI risk board is the **first line of defense** for all high-risk AI use cases. As AI adoption scales, this board inevitably becomes an **innovation bottleneck**, leaving teams waiting for reviews and direction (part of the critique in [[concept-standard-rai-approach]]).

**Under the ENC:** The first line of defense shifts to decentralized, project-level [[concept-enc-teams]]. The centralized risk board is **repurposed to handle only exceptions** — cases where an ENC team identifies a nightmare it cannot mitigate itself, or where the stakes are exceptionally high. Existing enterprise policies are re-cast as **guardrails** for the ENC teams rather than bureaucratic hurdles.

The concrete action is [[action-repurpose-risk-boards]].

**Enrichment note:** This distributed, "first-line ownership in product/operational teams with central committees handling escalations" pattern aligns with broader trends in AI governance. The precise "first line vs. exception handling" articulation goes somewhat beyond what Blackman spells out in public materials — treat it as a faithful extrapolation of his method. A standing counter-perspective: central boards remain indispensable for systemic risk, cross-jurisdictional decisions, and consistency across decentralized teams, so shifting the first line *entirely* to ENC teams risks fragmentation if not tightly coordinated.


## Related across articles
- [[framework-autonomous-scrum]]
- [[concept-modular-leadership-systems]]