Publications

Studies show that interactions with an AI system fosters trust in human users towards AI. An often overlooked element of such interaction dynamics is the (sense of) urgency when the human user is prompted by an AI agent, e.g., for advice or guidance. In this paper, we show that although the presence of urgency in human-AI interactions does not affect the trust in AI, it may be detrimental to the human user's self-confidence and self-efficacy. In the long run, the loss of confidence may lead to performance loss, suboptimal decisions, human errors, and ultimately, unsustainable AI systems. Our evidence comes from an experiment with 30 human participants. Our results indicate that users may feel more confident in their work when they are eased into the human-AI setup rather than exposed to it without preparation. We elaborate on the implications of this finding for software engineers and decision-makers.

Twin transition is the method of parallel digital and sustainability transitions in a mutually supporting way or, in common terms, “greening of and by IT and data.” Twin transition reacts to the growing problem of unsustainable digitalization, particularly in the ecological sense. Ignoring this problem will eventually limit the digital adeptness of society and the problem-solving capacity of humankind. Information systems engineering must find ways to support twin transition journeys through its substantial body of knowledge, methods, and techniques. To this end, we systematically survey the academic and gray literature on twin transition, clarify key concepts, and derive leads for researchers and practitioners to steer their innovation efforts.

This paper introduces a novel approach to reduce inconsistencies in pairwise comparison matrices using a genetic algorithm inspired by the process of natural selection. The method applies a distance-based inconsistency index as the fitness function within an evolutionary process. Through experiments, we show that our genetic algorithm greatly reduces inconsistencies over generations. Smaller matrices quickly become consistent in just a few generations, showing the method’s effectiveness and efficiency for different scenarios and use cases. Larger matrices, however, require more generations to reach the acceptable consistency level. The algorithm works well for different matrix sizes, adapting effectively to various challenging situations.

As digitalization accelerates at a rapid pace, software-intensive systems—including AI—elevate stress, anxiety, and self-doubt, raising concerns about the long-term sustainability of digital systems. This thesis argues that sustainable digitalization requires attention to social and individual dimensions to preserve human values and well-being over time. It presents evidence from a multivocal literature review and an experiment with 30 participants, then recommends research directions for the software engineering community.