ACP Compared to the NTSB and Analogous High-Reliability Institutions

1. What the NTSB is (structurally, not rhetorically)

The National Transportation Safety Board (NTSB) is not a regulator, operator, or enforcer.

Its defining characteristics:

• Post-incident authority
• Investigatory independence
• No direct coercive power
• High epistemic legitimacy
• Public-facing, auditable findings
• Norm-setting through analysis, not command

Critically:

• The NTSB cannot force action.
• Yet its findings shape action system-wide.

This is the key structural insight.

2. Why the NTSB works (institutionally)

The NTSB succeeds because it solves three problems simultaneously:

a. Authority without operation

• It does not build planes.
• It does not fly planes.
• It does not regulate airlines.

This separation:

• prevents conflict of interest,
• preserves credibility,
• protects dissent.

b. Blame-minimizing truth production

• Focus is on causal chains, not punishment.
• Individuals are not scapegoated.
• Organizational and systemic factors are foregrounded.

This encourages:

• disclosure,
• cooperation,
• learning.

c. Structural memory

• Each investigation feeds a growing body of precedent.
• Patterns are tracked across incidents.
• Recommendations compound over time.

The NTSB is a memory institution for failure.

3. Where ACP is directly analogous to the NTSB

ACP converges strongly with the NTSB model, even though it operates pre-incident as well as post-incident.

a. Independence from builders and operators

Like the NTSB, ACP:

• must not be embedded in the systems it governs,
• must not be owned by optimization incentives,
• must not be subordinated to speed or deployment pressure.

ACP’s refusal power parallels the NTSB’s investigatory independence.

b. Focus on failure, not success

NTSB:

• Studies crashes, near-misses, anomalies.

ACP:

• Studies governance failures,
• misdelegations of authority,
• illegitimate accelerations,
• breakdowns of contestability.

Both are failure-first institutions.

c. Systemic causation over individual blame

NTSB findings emphasize:

• training,
• incentives,
• procedures,
• organizational culture.

ACP similarly:

• avoids moralizing individual error,
• treats mistakes as predictable under structure,
• targets institutional design.

4. Where ACP goes beyond the NTSB (and must)

The analogy is strong but incomplete.

a. Pre-incident intervention

NTSB:

• Investigates after harm occurs.

ACP:

• Intervenes before harm when:
  • authority is misallocated,
  • claims exceed evidence,
  • systems are insulated from challenge.

This is the major extension.

ACP is closer to:

• an NTSB + flight-readiness veto,
  but without becoming a regulator.

b. Epistemic scope

NTSB focuses on:

• mechanical,
• human-factors,
• procedural failures.

ACP must also govern:

• epistemic claims,
• future-oriented assertions,
• model-based reasoning,
• speculative justification.

This makes ACP closer to a meta-NTSB for knowledge-bearing systems.

c. Contestability as a first-class object

The NTSB assumes:

• aviation has mature norms of safety and reporting.

ACP cannot assume this.

Therefore ACP must explicitly enforce:

• dissent channels,
• refusal legitimacy,
• protection against retaliation,
• clarity about who may challenge whom.

ACP governs the conditions under which NTSB-like truth production is even possible.

5. Comparison table: ACP vs. NTSB vs. Regulators

6. Other NTSB-like institutions ACP resembles

ACP also converges with:

• Air Accident Investigation Branch (UK)
• BEA (France)
• Nuclear safety boards
• Medical morbidity & mortality conferences
• High-reliability organization review boards

Common thread:

They exist to learn from failure without being captured by the incentives that produced it.

ACP generalizes this logic to:

• AI,
• knowledge systems,
• decision infrastructure,
• institutional governance.

7. The critical difference: refusal vs. recommendation

This is the sharpest distinction.

• NTSB:
  • “Here is what happened. You should change this.”
• ACP:
  • “You may not proceed until this governance failure is resolved.”

That makes ACP more confrontational.

It also makes ACP harder to institutionalize.

8. The central risk ACP shares with the NTSB

The NTSB only works because:

• its findings are respected,
• its independence is protected,
• its budget and authority are insulated from industry retaliation.

ACP faces the same existential risk:

The moment its findings consistently inconvenience power, it will be pressured, diluted, or bypassed.

This is not hypothetical.
It is historically predictable.

9. Final placement

ACP can be understood as:

An NTSB-like institution for epistemic and governance failure — extended upstream, hardened against speculation, and equipped with refusal rather than enforcement.