Article 26High-risk
Obligations of deployers of high-risk AI systems
Deployer
Deadline: August 2, 2026
Article 26 defines the specific obligations of deployers of high-risk AI systems. Deployers must ensure they use the system in accordance with instructions for use, ensure human oversight and inform affected persons.
Key points
1Use in accordance with instructions provided by the provider
2Implementation of human oversight measures by competent persons
3Ensuring input data is relevant and representative
4Monitoring operation and reporting serious incidents
5Retention of automatically generated logs for at least six months
Related definitions
Related articles
Check your compliance with Article 26
Our free diagnostic identifies the obligations applicable to your AI system and guides you to the necessary documentation.