My project concerns the role of explanations in ensuring that artificial intelligence is applied in an ethically responsible way. Specifically, I will examine the advantages and disadvantages of what is called “Explainable Artificial Intelligence”, that is, different technologies for generating explanations of decisions made using artificial intelligence. Such explanations are important in order to enable people to understand and thereby think critically about the artificially intelligent systems that are increasingly impacting our everyday lives. But they also risk creating a false sense of understanding, which can be exploited to mislead or even manipulate. To resolve this dilemma, the project will therefore develop both philosophical theories and practical guidelines for how we can distinguish between ethically beneficial and ethically pernicious explanations.

Rune Nyrup — Independent Research Fund (dff.dk)