Detecting Insecure Authentication Flows at Design Time with Graph Machine Learning

• Unité de recherche: INNOSUISSE
• Numéro de projet: 136.969 INNO-ICT
• Titre du projet anglais: Detecting Insecure Authentication Flows at Design Time with Graph Machine Learning

Mots-clé (Anglais)

ICT, IT security and cryptography

Résumé des résultats (Abstract) (Anglais)

Modern IAM systems let non-experts design login flows visually, but small mistakes can open serious security gaps. This feasibility study explores using graph machine learning to analyse these flows as they are built and to highlight insecure configurations before they ever reach production.