AIRiskAware
AI Governance Glossary
Governance Practice

What Is Algorithmic Accountability?

Algorithmic Accountability is the principle that the organisations deploying automated systems remain answerable for those systems' decisions and impacts.

Definition

Algorithmic Accountabilitythe principle that the organisations deploying automated systems remain answerable for those systems' decisions and impacts.

Algorithmic accountability holds that responsibility cannot be outsourced to a model: a human organisation must be able to explain, justify, and remedy the outcomes its automated systems produce. In practice it is operationalised through documentation, impact assessments, audit trails, clear internal ownership, and avenues for affected people to contest decisions. It is the connective tissue between abstract AI principles and the concrete governance controls — record-keeping, oversight, redress — that regulators expect to see.

Source: OECD AI Principles; NIST AI RMF (GOVERN function)

Plain-language explanation

Algorithmic accountability holds that responsibility cannot be outsourced to a model: a human organisation must be able to explain, justify, and remedy the outcomes its automated systems produce. In practice it is operationalised through documentation, impact assessments, audit trails, clear internal ownership, and avenues for affected people to contest decisions. It is the connective tissue between abstract AI principles and the concrete governance controls — record-keeping, oversight, redress — that regulators expect to see.

Primary source: OECD AI Principles; NIST AI RMF (GOVERN function)

See where you stand on AI governance

Take the free 7-question maturity assessment and get a personalised action plan.

Free assessment — 3 minutes →