Our research associates, Andreas Schauer and Philipp Hartl, presented the following papers:
Hartl, P. & Manzke, L. (2026). Information Sensitivity Contamination In Sequential Data Donation Requests.
https://aisel.aisnet.org/ecis2026/security/security/1/ (external link, opens in a new window)
Using a pre-registered experiment involving actual Duolingo data requests, this paper investigates how the sequence and presentation of requests influence the decision to donate data. The results show that sequential request strategies significantly reduce the willingness to make low-effort donations—driven by a "contamination effect" where exposure to more demanding requests increases the perceived sensitivity of subsequent requests, as well as by the increased cognitive load associated with simultaneous presentation. Furthermore, the study reveals that "warm-glow" motivation—but not a general altruistic attitude—positively predicts donation willingness, whereas general privacy concerns reduce it. The study thus questions the applicability of classical prosocial theories to privacy-sensitive digital contexts.
This research project is funded by the Bavarian Research Institute for Digital Transformation (bidt), an institute of the Bavarian Academy of Sciences and Humanities.
Förster, M.; Hagn, M.; Hambauer, N.; Jaki, Paula. K. V.; Obermeier, A. A.; Schauer, A.; and Schiller, A. (2026). Understanding Uncertainties In Explainable AI: A Structured Literature Review and Research Agenda.
AIS Electronic Library (AISeL) - ECIS 2026 Proceedings: Understanding Uncertainties In Explainable A… (external link, opens in a new window)
Based on a structured literature review, this paper examines the sources of uncertainty in XAI-supported decision-making and how these uncertainties influence human-AI collaboration in high-risk sectors such as healthcare. The results show that data, model, XAI method, and human uncertainties have predominantly been considered in isolation to date, resulting in a fragmented understanding of their interplay. Building on this, the authors develop an integrated approach for incorporating multiple sources of uncertainty into uncertainty-aware explanations and derive a research agenda for more trustworthy, uncertainty-aware XAI systems.
Further information about the conference:
https://ecis2026.it/ (external link, opens in a new window)
