What is a high-risk AI system?
The AI Act classifies as high-risk those AI systems used in domains where errors harm fundamental rights: hiring, education, credit, justice, essential public services, biometrics, critical infrastructure. These systems will be subject to strict requirements on risk management, data quality, human oversight and transparency.
Regulation EU 2024/1689 classifies as high-risk AI systems used in areas where an error or malfunction can seriously harm health, safety or fundamental rights. These systems are not prohibited, but are subject to strict requirements: risk management, data quality and governance, technical documentation, transparency, human oversight, robustness and security. This is the core of the regulation, most of which applies from August 2, 2026.
Areas concerned
The regulation identifies high-risk uses, particularly in recruitment and worker management, education and training, access to essential services such as credit, law enforcement, justice, migration and asylum, biometrics, and security of critical infrastructure. The common point is the possible impact on fundamental rights.
Requirements imposed
Risk management system throughout the lifecycle. Quality training, validation and test data, managed to limit biases. Technical documentation and event logging. Transparency and information for deployers. Effective human oversight, enabling understanding, monitoring and, if necessary, interruption of the system. Appropriate level of accuracy, robustness and cybersecurity.
Why this level of requirement
In these areas, a biased or erroneous automated decision can deprive a person of a job, credit, a training place, or affect their freedoms. The regulation therefore imposes guarantees proportionate to the seriousness of the possible consequences.
What this brings you
Faced with a high-risk system that concerns you, you benefit from guarantees: quality, human oversight, transparency. These requirements reinforce your existing rights, particularly those from the GDPR on automated decisions.
Frequently Asked Questions
Is a high-risk system prohibited?
No, it is authorized but subject to strict requirements.
Is CV sorting high-risk?
AI systems used in recruitment are among the high-risk uses targeted by the regulation.
When do these requirements apply?
Most apply from August 2, 2026, with a deadline in 2027 for certain systems integrated into already regulated products.
ICIA Resource
ICIA offers MentivisOS ICIA, a free lifelong platform to learn AI with a personalized learning path.
Discover MentivisOS ICIA