Leonhard Kestel

Brief description of my doctoral project

Motivated by the growing influence of AI systems in high-stakes domains such as hiring, healthcare, and public administration, my work addresses a central problem of contemporary AI: ethical guidelines are typically implemented implicitly by a small group of technical experts and private actors. Trustworthy AI, however, requires transparent and controllable mechanisms for aligning model behavior with societal values. Therefore, I develop different neurosymbolic architectures that integrate neural learning with symbolic reasoning over formal constraints. The proposed approaches introduce an explicit interface between neural latent spaces and declarative ethical specifications. Thereby, they enable interpretable, modular, and adaptable alignment mechanisms for learning algorithms.