Every six months, a Deep Narrative Intelligence programme reaches a milestone: the executive committee synthesis. This is not a data dump of every measurement cycle, nor a technical report on model behaviour. It is a strategic narrative that connects the AI narrative layer to business outcomes, presented in a format that a busy executive committee can absorb in 20 minutes and act on. Here is how to structure it, what to include, and what to leave out.
What the synthesis is — and is not
The synthesis is a strategic document, not an operational report. It does not list every error found and corrected; it identifies the three to five narrative themes that mattered over the period, the direction of each, and the business implication. It does not present model-by-model data; it presents the aggregate AI Image Score trend with annotations on what drove major movements. It does not recount the correction cycle in detail; it reports the correction persistence rate and flags any regressions that the committee should be aware of. The synthesis is the document that helps the executive committee understand the AI narrative layer as a strategic asset — or liability — that they are now governing.
The three-part structure
The synthesis has three parts. Part 1: The state of the narrative — the current AI Image Score, the six-month trend, and the top three narrative themes (e.g., "the market perceives us as lagging on AI adoption", "our safety record is accurately represented", "a stale acquisition rumour persists in two models"). Part 2: What we did and what happened — the corrections executed, the persistence rate, the regressions encountered, and the provenance insights gained. Part 3: What we recommend — the three decisions the committee should make, each with a clear recommendation, a rationale, and a resource implication. The structure is deliberately simple because the committee's time is finite, and a document that tries to say everything ends up saying nothing.
Don't know what AI says about you today?
A SetRecord audit queries supported AI models about your organisation and measures the gap with verifiable facts — within five business days.
Connecting to business outcomes
The synthesis earns its place on the executive committee agenda only if it connects to business outcomes. This means translating narrative findings into the language the committee already speaks. If the AI Image Score on "financial reliability" is low, the connection is to investor confidence and cost of capital. If a competitor is consistently framed as the market leader in model answers, the connection is to pipeline conversion and RFP win rates. If a phantom controversy persists, the connection is to talent acquisition and partnership conversations. Every narrative finding in the synthesis is paired with a business implication, because an AI narrative finding without a business connection is a technical curiosity, not a strategic input. SetRecord structures its six-month syntheses around these business connections, because the committee does not govern what it does not understand in its own terms.
The three decision points
The synthesis must end with three decision points — no more, no less. Three is the number a committee can genuinely deliberate in a single session. Each decision point has a recommendation (what we think you should do), a rationale (why), and a resource implication (what it costs in time, money, or people). Examples: "Approve the expansion of the interrogation panel to cover the new product line (rationale: the model layer does not yet reflect the launch; resource: 2 additional weeks of programme time)", "Approve a targeted correction campaign on the stale acquisition narrative (rationale: persistence rate on prior corrections in this topic is below threshold; resource: co-development with corporate communications for 4 weeks)", "Note and endorse the current monitoring cadence without changes (rationale: the programme is stable and performing; resource: none)". The committee may accept, modify, or reject each recommendation, but the synthesis has done its job: it has presented clear choices, not vague concerns.
How to deliver it
The synthesis is delivered as a 20-minute presentation with a 5-minute Q&A, supported by a 6-page written document. The presentation follows the three-part structure, with one slide per narrative theme, one slide for the correction summary, and one slide for the decision points. The written document provides the supporting detail for anyone who wants to go deeper. The tone is strategic, not technical: the committee hears about narrative themes and business implications, not about model architectures and training cycles. The goal is for every committee member to leave the room understanding one thing: the AI narrative layer is now a managed asset, it has a measurable state, that state has a trend, and the trend is responsive to deliberate intervention. That understanding — not any single data point — is the deliverable.