Board Reporting and Metrics
Build an AI scorecard that covers value, adoption, risk, and capability; prepare for the questions boards ask about AI; and report honestly when results are mixed.
The Board’s Job and Yours
Boards are responsible for overseeing strategy and risk. For AI, that means asking whether management has a credible strategy, whether it’s delivering, and whether the risks are understood and controlled. Directors increasingly face questions from investors and regulators about AI oversight.
The CAIO’s job is to make that oversight possible: clear, consistent, honest information, presented the same way each time so trends are visible.
An AI Scorecard
A one-page scorecard with four sections, reported quarterly:
1. Value
- Realized value from production use cases, measured against baselines, with ranges where estimated.
- Pipeline value: what’s expected from use cases in pilot, with confidence levels.
- Total AI spend, and spend against plan.
2. Adoption
- Depth of use in priority roles (not just logins).
- Number of workflows redesigned to include AI.
- Literacy program coverage, by role layer.
3. Risk
- AI inventory size, and count by risk tier.
- High-risk systems reviewed, and any overdue reviews.
- Incidents: count, severity, and status.
- Regulatory changes affecting the organization, and readiness.
4. Capability
- Platform maturity milestones.
- Key hires and skills gaps.
- Vendor concentration and lock-in exposure.
Use the same structure every quarter. Show trends, not just snapshots, and use traffic lights sparingly and honestly.
Common mistake: presenting a highlight reel of successful pilots and impressive demos.
Boards that only hear good news either lose trust when problems surface or, worse, don’t find out until a crisis. Report what’s stalled, what you stopped and why, incidents, and where value fell short. Credibility built in a mixed quarter is what lets you ask for investment in a good one.
Questions Boards Ask
Prepare clear answers to the questions directors commonly raise:
- Strategy: How does AI support our business strategy? What are competitors doing, and are we ahead, level, or behind?
- Value: What have we spent, and what have we got? How do we know?
- Risk: What AI do we use that could harm customers, employees, or the company? How do we know what’s running?
- Regulation: Which laws apply, and are we ready for upcoming deadlines?
- People: What does AI mean for our workforce? What have we told employees?
- Security: How are we protecting data in AI tools, and against AI-enabled attacks?
- Dependencies: How dependent are we on a few AI vendors? What happens if one changes terms or fails?
- Oversight: Who’s accountable for AI, and how does the board stay informed?
Board AI Literacy
Directors need enough understanding to oversee AI, not to build it. Offer the board:
- A short education session each year on what AI can and can’t do, using your organization’s own examples.
- Hands-on time with the tools employees use.
- Clarity on which committee oversees AI risk (often audit or risk) and how often it hears from you.
- A risk appetite statement for AI that the board approves: what kinds of AI use the organization will and won’t pursue.
A board that has agreed an AI risk appetite makes hard decisions easier, because the CAIO can point to it.
Telling the Story
Numbers alone don’t land. Pair the scorecard with:
- One case study per quarter: a use case’s before and after, told through the people involved.
- One lesson learned: something that didn’t work and what changed as a result.
- One decision needed: a clear ask of the board, if there is one.
- Fill in the four sections of the scorecard for your organization, using real or plausible figures.
- Mark anything that’s estimated rather than measured.
- Choose one case study, one lesson learned, and one decision you’d bring to the board.
- Write a two-sentence draft of an AI risk appetite statement.
1. Why should the AI scorecard use the same structure every quarter?
Consistency lets directors see whether things are improving or getting worse.
2. What's the risk of presenting only good news about AI to the board?
Honest reporting builds the credibility needed for future investment.
3. What does a board-approved AI risk appetite statement help with?
An agreed risk appetite makes trade-offs explicit and easier to decide.
Up Next
The final lesson covers AI incidents, vendor management, and your 12-month roadmap.