Design Principles for Human-Centred Artificial Intelligence

Advancements within artificial intelligence (AI) enable organisations to reformulate strategies forexploiting data in order to refine their business models, make better decisions and maintain acompetitive advantage. We recognise the technical advantages of AI. However, our view is that thetechnical perspective as a base for decision-making is necessary but insufficient. Several studies inhuman science report that essential human knowledge and competencies that affect decision making arenot represented in AI systems. Based on this observation, we have developed design principles fordeveloping decision-support systems (DSS) that combine human intelligence (HI) with AI. The designprinciples are: design for amplified decision-making, design for unbiased decision-making and designfor human and AI learning. The design principles constitute the scientific contribution to the emergentfield of Human-Centred AI. The contribution to practice consists of a DSS (a digital prototype) thatsupports the combination of HI and AI.Finansiär: Handelsrådet