In this service, our experts check an existing Explainable AI (XAI) implementation and/or give advice on how to implement or improve explainability for your system.
Explainability of AI systems is often demanded, for example, for the traceability of error cases, auditing or in laws (e.g., the EU AI Act). However, implementing corresponding explainability methods is challenging, as many methods exist, but their validation currently poses an open problem. In this service, our experts evaluate whether your implementation follows current best practices, standardization and legal interpretation, to aid you in getting your system ready for deployment.
Report with assessment results and, where relevant, with suggested XAI methods. Optionally the implementation of those methods.
joint workshops, if wanted implementation of xAI-methods
Open for anyone with a (nearly) finished AI application