« Forschungslandschaft: Projekte

Explainable AI is dedicated to solving the so-called Black Box Problem in AI: the problem that arises because the computing systems developed for artificial intelligence are increasingly opaque. Although such systems may be capable of solving many complex problems in AI, it is often unclear why they do what they do or how they work. Explainable AI is dedicated to solving the Black Box Problem by rendering such opaque systems transparent. But how should Explainable AI explain? What are the explanatory norms that should be satisfied by current XAI methods?