Operationalizing AI Ethics

The strong increase in the development and deployment of Artificial Intelligence (AI) systems has led to growing concerns about the ethical aspects of these systems. As a result, a proliferation of high-level ethical principles for AI has emerged. But, as it turns out, these principles are very difficult to bring into practice: principles may place conflicting demands on AI design or are hard to formalize in a univocal way. We will discuss these hard questions of operationalizing AI ethics via thought-provoking questions and dilemmas for the audience to consider. And we will look at the preliminary results from the Moral AI-Gency project: an initiative supported by the SIDN Fonds, Erasmus University Rotterdam and de Volksbank to provide research driven answers on how we can balance AI’s unique potential for problem solving with the serious ethical risks of discrimination, lack of transparency and the diffusion of responsibility.