The four categories of risk classification in the EU AI Act?

Prepare for the ISACA Advanced in AI Security Management (AAISM) Test. Study with in-depth multiple choice questions, each offering insightful hints and detailed explanations. Equip yourself with expert knowledge and get exam-ready!

Multiple Choice

The four categories of risk classification in the EU AI Act?

Explanation:
The EU AI Act uses a risk-based framework to regulate AI systems, grouping them into four levels to determine what controls apply. The four categories are Prohibited, High risk, Limited risk, and Minimal risk. Prohibited covers uses deemed to pose unacceptable risk and are banned outright. High risk includes systems in areas like critical sectors and decision-making that affect people’s rights, which require robust governance, documentation, data management, conformity assessments, and human oversight. Limited risk includes applications with some potential for harm that come with specific transparency obligations, such as informing users they’re interacting with AI. Minimal risk covers most everyday AI uses and carries little to no additional regulatory burden beyond general consumer safety and ethics considerations. This set of categories matches the actual EU framework, whereas other term sets don’t reflect the Act’s risk-based tiers.

The EU AI Act uses a risk-based framework to regulate AI systems, grouping them into four levels to determine what controls apply. The four categories are Prohibited, High risk, Limited risk, and Minimal risk. Prohibited covers uses deemed to pose unacceptable risk and are banned outright. High risk includes systems in areas like critical sectors and decision-making that affect people’s rights, which require robust governance, documentation, data management, conformity assessments, and human oversight. Limited risk includes applications with some potential for harm that come with specific transparency obligations, such as informing users they’re interacting with AI. Minimal risk covers most everyday AI uses and carries little to no additional regulatory burden beyond general consumer safety and ethics considerations.

This set of categories matches the actual EU framework, whereas other term sets don’t reflect the Act’s risk-based tiers.

More Multiple Choice Questions
Subscribe

Get the latest from Passetra

You can unsubscribe at any time. Read our privacy policy