A fundamental rights impact assessment is a structured process for evaluating, before and during deployment, how a system, policy, or AI application may affect people’s fundamental rights such as dignity, non-discrimination, privacy, and freedom of expression. It documents the context of use, identifies affected groups and potential harms, assesses risks, and defines mitigation and oversight measures. It has become a governance instrument for high-risk AI, complementing data protection impact assessments and broader risk assessment.
- A fundamental rights impact assessment is an Impact Assessment focused on effects on Human Rights.
- It is a governance instrument within AI Governance.
- It bridges to the Data Protection Impact Assessment while widening scope beyond privacy.
- It requires Risk Assessment and Stakeholder Engagement.
Overview
- The assessment examines how a system or policy could affect rights such as dignity, equality, privacy, and expression.
- It is performed before deployment and revisited as context changes, documenting affected groups and potential harms.
- It defines mitigations, oversight arrangements, and monitoring to manage residual risk.
- It has become a recognised governance step for high-risk AI systems.
Key aspects
- Context analysis of where, how, and on whom a system will be used.
- Identification of affected groups and rights at stake.
- Risk evaluation and definition of proportionate mitigations.
- Oversight, transparency, and ongoing monitoring obligations.
Applications
- Pre-deployment review of high-risk public-sector AI.
- Complementing data protection impact assessments for AI systems.
- Procurement and due-diligence gating for sensitive deployments.
- Documenting accountability for regulators and affected communities.