Human oversight: Under the EU AI Act, which requirement does this design measure most directly implement?
Enabling a person to monitor, intervene in, or stop the system implements the EU AI Act's human-oversight requirement.
The question
A deployer of a high-risk AI system must ensure a person can intervene in or halt the system's operation when needed. Under the EU AI Act, which requirement does this design measure most directly implement?
Preparing for AIGP? Take the free 5-min readiness quiz →
- Human oversight, ensuring a natural person can effectively monitor, intervene in, or stop the high-risk system's operation. ✓Correct: enabling a person to monitor and intervene is the human-oversight requirement for high-risk systems.
- Data minimization, ensuring the system collects and retains only the least personal data needed to reach its stated output goals.Plausible but wrong: data minimization concerns data volume, not the ability to intervene in operation.
- Copyright compliance, ensuring the training-data summary discloses the protected works used to build the model.Plausible but wrong: copyright/training-data duties are unrelated to operational intervention.
- Cross-border transfer control, ensuring personal data leaves the Union only under an approved legal mechanism.Wrong: transfer controls govern data movement, not human intervention in the system.
The trap
Mixing up human oversight (operational control) with data-protection duties like minimization or transfers. How to remember it
Enabling a person to monitor, intervene in, or stop the system implements the EU AI Act's human-oversight requirement.
How many of these would you get right?
One of 499 AIGP questions on Certsqill. Take a free five-minute check and see your score per domain — not one number, but which section to open tonight.
Test your AIGP readiness — freeMore Understanding How Laws, Standards and Frameworks Apply to AI questions
- Disparate-impact liability can arise from a facially: Under nondiscrimination law, why can this still create →
- GOVERN, the cross-cutting function establishing a risk: Which core function is cross-cutting and informs the →
- A valid Article 9 condition: Because fingerprints used to identify people are special-category data, what must →
- All 125 Understanding How Laws, Standards and Frameworks Apply to AI questions →
Part of the Certsqill AIGP question bank · Understanding How Laws, Standards and Frameworks Apply to AI ·
Every answer, right and wrong, comes with its own explanation.
