Trustworthy Human-Centric AI

The European Union has taken the stance that AI should be trustworthy and developed in a human-centric way with the goal of improving individual and societal well-being. This talk will present the European approach to trustworthy human-centric AI and some research challenges related to it. To be trustworthy an AI-system should be lawful, ethical and robust, as defined by the European Commission High-Level Expert Group on AI. To operationalize these is a major challenge and will require new research. The second part of the talk gives an overview of the state-of-the-art and potential future solutions to these challenges.