RequirementsArticle 15(4)–(5)
Cybersécurité
L’exigence selon laquelle les systèmes d’IA à haut risque sont résilients face aux tentatives de tiers visant à modifier leur usage, leur comportement ou leurs performances de manière susceptible d’entraîner des risques pour la santé, la sécurité ou les droits fondamentaux, y compris la protection contre la contamination des données, les exemples contradictoires et les attaques par évitement du modèle.
Termes associés
- Risk Management SystemA continuous iterative process that must be established, implemented, documented, and maintained by providers of high-risk AI systems throughout the…
- Data GovernancePractices and policies applicable to training, validation, and testing data sets used for high-risk AI systems, covering the design choices, data…
- Human OversightMeasures built into high-risk AI systems enabling natural persons to understand, monitor, and — where necessary — override or shut down the system. Must…
- AccuracyThe requirement that high-risk AI systems achieve an appropriate level of accuracy in relation to their intended purpose, as specified in the technical…
- RobustnessThe ability of a high-risk AI system to maintain its level of performance under adverse conditions — including technical limitations, adversarial inputs,…
- Transparency RequirementsObligations under Article 13 requiring that high-risk AI systems are designed to ensure that their operation is sufficiently transparent to enable…
Contrôles de conformité associés
- DG-009Statistical Testing and ValidationProviders must apply appropriate statistical methods to validate that training data meets quality thresholds and that the resulting model generalises appropriately to the intended deployment distribution.
- TD-006Accuracy Metrics and Performance DocumentationProviders must document the accuracy, robustness, and cybersecurity metrics for which the high-risk AI system was designed and the relevant testing methodologies, including the context in which these metrics were measured.
- RK-006Performance and Operational Metrics LoggingProviders must log performance and operational metrics relevant to verifying that the AI system operates within the parameters established in the technical documentation, including accuracy and response time indicators.
- TR-003Performance Metrics Disclosure to DeployersThe instructions for use must include the performance metrics for which the system has been designed and the level of accuracy that the provider committed to achieving, enabling deployers to assess whether the system meets their operational requirements.
Passez à la vitesse supérieure quand la conformité doit être
Le questionnaire gratuit fournit des signaux préliminaires. L'Évaluation Complète transforme les données réelles de vos systèmes en un dossier de décision gouverné : extraction, séparation des composants, preuves, surcouches nationales et dossier prêt à être audité.
Démarrer l'aperçu gratuit des risques