What is named accountability in AI governance?
Named accountability means every AI governance decision — screening, classification, approval, override, or re-screening — has a specific person responsible. Not a team, not a department, but a named individual whose decision can be traced in the governance record.
This is a structural requirement for audit-ready governance, because decisions that cannot be attributed to a responsible person cannot be reconstructed or defended at audit.
Key points
- Attribution requires named responsibility at each governance step: who screened, who reviewed, who approved, who overrode.
- Accountability cannot be delegated informally. The governance record must show the specific person, not a general role reference.
- Named accountability extends to non-decisions. If a system was not screened, the absence of a responsible reviewer is itself an audit finding.
- Accountability and authority must align. The person approving a governance decision must have the organizational authority to do so.
Why it matters
AI governance without named accountability creates a structural gap that cannot be closed retroactively. When an auditor asks who approved a specific AI system's deployment, the answer must be a name, a timestamp, and a documented decision basis — not a reference to a committee or a policy document. Named accountability transforms governance from organisational intent into verifiable execution.
How EAB approaches this
EAB enforces named accountability through its role-based governance structure. Every action — system registration, screening completion, supervisor approval, override — is logged with the specific user, timestamp, and decision context. The audit trail preserves attribution across the full governance lifecycle. The Executive Cockpit shows accountability status at the portfolio level.