---
title: "Six Board Concerns"
date: 2026-08-23
description: A model that holds Board AI conversations open across six distinct concerns, preventing their collapse into a single question of risk.
author: Mario Thomas
canonical: https://mariothomas.com/remake/library/six-board-concerns/
---

## What it is {#what-it-is}

Most Board AI conversations collapse into a single concern — usually risk. The model exists to prevent that collapse: six concerns, held in view together, because they are interconnected — tightening risk management in a way that suffocates experimentation simply moves the exposure to Safeguarding Innovation; chasing stakeholder confidence without strategic alignment buys applause for a portfolio that serves no strategy.

Interconnected by design: addressing one while ignoring the others creates new vulnerabilities. Governed as a system, never a checklist.

- **Strategic Alignment** 
AI effort serving the strategy the Board has actually set — not a parallel portfolio of enthusiasms.

- **Ethical & Legal Responsibility** 
The duties that do not delegate: lawful, explainable, defensible use of systems the organisation deploys.

- **Financial & Operational Impact** 
What AI does to the numbers and the operations — investment, return, and the operational dependencies being created.

- **Risk Management** 
The risk picture as it actually is: model behaviour, data exposure, third parties, and the failure modes of autonomy.

- **Stakeholder Confidence** 
What customers, regulators, investors, and employees believe about your AI — and whether the evidence supports it.

- **Safeguarding Innovation** 
Protecting the organisation's capacity to experiment and adapt — governance that never becomes the reason nothing new ships.

## How to read it {#how-to-read}

As a system, never a checklist. The reading is the *balance*: which concerns are absorbing the Board's attention, which are unattended, and what the current governance posture is silently trading between them. A Board that can name where each of the six stands — and where the tension between them currently sits — is governing; a Board working down a list is auditing.

What the model makes visible is the single-concern failure mode. Risk is the concern Boards reach for because it has familiar machinery — registers, appetites, committees. The other five have no such machinery unless the Board builds it, so they go ungoverned precisely while feeling governed. The six-concern reading shows the whole agenda at once, and the neglected concerns show up as the empty columns.

## A worked reading {#worked-reading}

A typical first reading: Risk Management well-attended (a standing register, a committee), Financial Impact partially attended (investment tracked, operational dependency not), Strategic Alignment asserted but unevidenced, Ethical & Legal delegated to a policy document, Stakeholder Confidence unmeasured, Safeguarding Innovation absent from the agenda entirely. The model does not score this; it makes the shape of the attention visible — and the shape is the finding.

Within [Remake](/remake/), the Six Board Concerns are declared a **Model** — the Board-altitude lens held open through the whole ADAPT motion: one system answered continuously, so the Board's whole agenda stays in view while the operators work stage by stage.