Supervision Ethique
At TOP, ethical supervision is one of our major concerns, and we constantly ensure its proper implementation.
TOP ensures the quality of its model's training data by constantly monitoring its performance and accuracy.
The platform actively identifies biases, such as discriminatory variables, and corrects them through debiasing metrics, ensuring maximum fairness and objectivity.
In pursuit of transparency and traceability, TOP explicitly defines its service and the context of using Machine Learning tools in its models, providing a clear understanding of their application.
TOP takes responsibility for the results of its algorithm, especially in the case of negative consequences, demonstrating its commitment to the reliability and security of decisions made.
To strengthen trust, TOP anticipates CE standard certification by Afnor for its application, demonstrating compliance with European standards for quality and safety.
To ensure complete auditability, TOP makes its entire system auditable by the SIA, ensuring total transparency on its processes and decisions.
TOP integrates an ethical supervision view of its algorithm accessible via the administrator account, providing complete visibility into the ethical principles guiding its use.
