The Seven Enterprise AI Decisions That Belong in the Boardroom

What is changing

AI is moving from isolated tools into decisions, workflows and transactions. As that happens, choices that once looked technical can change enterprise investment, delegated authority, workforce design, critical dependencies and the level of machine autonomy the organization is prepared to accept.

The governance question is therefore narrower than “Should the board oversee AI?” The useful question is which enterprise choices require board direction, limits, visibility or designated approval. Routine implementation remains with management. Material changes follow the governance route defined by enterprise consequence and documented delegation.

Why it matters

The risk is accumulation. AI investment can spread across separate initiatives, authority can expand in small steps, and dependencies can concentrate across models, data, platforms and providers. Each choice may sit inside an existing approval limit while the combined enterprise consequence becomes material.

External evidence illustrates why operating governance deserves attention. ISS STOXX reported that 8% of 3,048 US companies in its January 2026 disclosure dataset disclosed board-level AI oversight. EY separately reported that 47% of 202 senior AI decision-makers at large US public companies surveyed in May and June 2026 had bypassed governance processes for urgent deployments while 98% reported formal policies. These are different measures and populations. They do not establish governance effectiveness, but they reinforce the need to test how governance works under operating pressure.

6xD perspective

Through D2, Digital Cognitive Organization, the organization is part of the AI architecture. Decision rights, process ownership, human review, information quality, skills, escalation and accountability shape what AI-enabled operations can responsibly do.

RP01 provides the master Enterprise AI Governance System: five governing functions explain how the board governs, eight responsibility domains define what must remain visible, and one board-management boundary preserves management ownership of implementation. The seven decisions below are the boardroom application lens. They translate that architecture into recurring choices directors can recognize and act on.

The seven board decisions

How to use the seven decisions

Start with the enterprise consequence, then identify which of the seven decisions is changing. A routine change can remain within management delegation. A material change may require board visibility and challenge. A reserved matter or threshold breach should move to the designated approval route before investment, exposure or AI authority expands.

Readiness is the cross-cutting gate. Before the enterprise scales a material capability, management should be able to show the intended outcome, accountable owner, delegated authority, material limits, current evidence, critical dependencies, intervention arrangements and the condition that would trigger reassessment.

What the board should require

For each material AI item, the board pack should make the decision visible before the technology detail. It should state the enterprise outcome, which of the seven board decisions is changing, the management recommendation, the authority already delegated, the evidence supporting the recommendation, any material dependency, and the event that would trigger escalation or reassessment. This keeps the discussion focused on the enterprise boundary being set while leaving implementation choices with management.

Practical board test

At the next Board or Technology/Digital Committee discussion, pick one material AI initiative and ask three questions: Which of the seven decisions does it change? Which governance route applies? What evidence shows the organization is ready for the proposed level of investment, authority or autonomy?

If those answers are unclear, the board does not need more AI inventory. It needs a clearer enterprise decision before the boundary moves.

ASSESS GOVERNANCE READINESS

Assess where your board still lacks clear direction, decision rights or evidence before material AI authority or investment expands →