Last updated:
A mandatory evaluation under the EU AI Act for certain deployers to determine how the use of a high-risk AI system might impact civil liberties and human rights.
Required under EU AI Act Article 27 for public bodies and recommended for private entities, a FRIA involves identifying the specific groups of people likely to be affected by an AI system. The assessment must describe the intended use, the time period of operation, and the specific measures taken to mitigate risks to rights like privacy, non-discrimination, and freedom of expression.
Real world example:
A municipal police department conducts a FRIA before deploying an AI-powered crowd-monitoring system to ensure the technology does not lead to biased surveillance of specific ethnic neighborhoods.
