Concept dictionary
Algorithmic Accountability
Algorithmic accountability is the principle and practice that organizations remain answerable for the design, use, effects, and remediation of decisions made or supported by algorithmic systems.
- Review state
- Needs review
- Freshness
- Current
- Version
- 2
Also called
- Algorithmic Accountability
Remember it this wayAutomation can move the decision, not erase accountability.
In plain language
Responsibility does not disappear because software recommended or made a decision. People and institutions must be able to explain authority, examine evidence, correct harm, and change the system.
Why it matters
Automated systems can spread decisions at scale while making ownership harder to see. Accountability preserves a path for review, challenge, correction, and consequences.
Example
A company using an automated hiring screen maintains a named owner, audit evidence, applicant appeal process, and authority to suspend the system.
Common misconception
Accountability is not satisfied by saying a vendor or model made the decision. Deploying organizations retain responsibilities that depend on law and context.
Related terms
Sources
- NIST AI RMF — Accountable and Transparentofficial guidance connecting accountability with transparency across AI design, data, training, deployment, and decision-making.
- NIST SP 1270, Towards a Standard for Identifying and Managing Bias in Artificial Intelligenceofficial guidance on governance, assurance, documentation, and practices for identifying and managing harmful bias.
Learn more
- Use [[human-ownership|Human Ownership]] to make accountability operational.