What is required for high-risk AI solutions under the EU AI Act?

• A. A Fundamental Rights Impact Assessment (FRIA).
• B. A simple profitability analysis.
• C. A data-minimization check only.
• D. A basic code review.

Answer: (A)

High-risk AI solutions under the EU AI Act must undergo a fundamental rights impact assessment to identify and mitigate potential effects on rights such as privacy, non-discrimination, and freedom of expression. This FRIA is part of a broader risk management approach built into the lifecycle of the system, guiding design choices, data governance, controls, and how the system is monitored and demonstrated for compliance. It ensures that risks to fundamental rights are considered early and continually addressed.

Other options miss this essential focus on rights and governance. A profitability analysis isn’t about legal or ethical compliance. A data-minimization check alone only covers one aspect of data handling and doesn’t evaluate broader rights impacts or systemic risks. A basic code review doesn’t capture the overarching risk management, accountability, and ongoing monitoring obligations that the act requires for high-risk AI.