67471 research outputs found
Sort by
Bridging Design Science Research and Formal Design Theories: Leveraging C-K Theory for Impactful Research
Get PDFThis paper proposes a novel integration of formal design theories—specifically C-K theory—into Design Science Research (DSR) to better address complex and ill-defined problems. Traditional DSR methods often assume well-framed problem spaces, limiting their generative potential in tackling grand societal challenges. By embedding C-K theory into DSR, we outline a methodology that enhances generativity through iterative co-expansion of concept and knowledge spaces. An empirical case in the healthcare sector illustrates this approach, showing how C-K theory fostered the discovery of anomalies, expanded the knowledge base, and led to new artefacts and business model innovations. Our findings demonstrate the relevance of formal design theories in supporting both knowledge exploration and generative artefact development. We conclude that such integration enables more generative, reflexive, and impactful research processes
The Effect of Type of Explanation on Algorithm Appreciation: The Role of Risk Perceptions in Healthcare Decision-Making
Get PDFThis study examines the impact of various types of Artificial Intelligence (AI) explanations—local, counterfactual, and global—on individuals' algorithm appreciation in healthcare decision-making. Using a scenario-based experiment involving 611 participants, we take a risk perspective to examine how eXplainable (XAI) system credibility (risk probability) and perceived condition severity (risk severity) mediate the relationship between explanation type and algorithm appreciation. We also explore how decision-makers’ risk-taking propensity (risk perception) moderates these relationships. Participants assessed diabetes risk predictions for a hypothetical relative based on explanations generated by an XAI system. Findings reveal that explanation type significantly influences algorithm appreciation through the perceived severity of the condition, but not through the credibility of the XAI system. Importantly, the effects of explanation type vary with participants' risk-taking propensity. Hence, this research highlights the need for personalized XAI strategies to maximize algorithm appreciation in high-risk healthcare decision-making contexts involving non-expert decision-makers
Link in Bio, But Where? Strategic Platform Adoption by Influencers in the Creator Economy
Get PDFSocial media influencers have set the stage for the emergence of the creator economy, turning content creation into a viable source of income for both themselves and collaborating brands. However, limited attention has been given to how influencers can strategically grow their engagement. Our study addresses this gap by examining how adopting new platforms helps influencers grow different forms of engagement depending on their type. The analysis builds on the Bass diffusion model and its multi-market and multi-product extensions. Using a unique proprietary dataset with longitudinal engagement data from Instagram and TikTok, we show that new platform adoption leads to significant increases in engagement, with expert influencers benefiting primarily from interactive engagement and entertainer influencers from appreciative engagement. The effect, however, is not uniform and could vary by influencer type, adoption order, and mixing behavior, underscoring the importance of tailored multi-platform growth strategies to optimize engagement outcomes
Assessing the Impact of Algorithmic Quantity Regulations on Sharing Platforms: Evidence from Airbnb in Paris
Get PDFIn recent years, several automated caps, or algorithmic quantity regulations (AQRs), have been deployed to police supply conditions in sharing economy platforms. AQRs constitute a paradigm shift in platform regulation, as they enable exhaustive, and low-cost enforcement, thus comprehensively influencing interactions both within and outside the focal platform. However, their actual impact is not known, and has not been studied so far. In this work, we employ a series of difference-in-differences analyses to provide causal evidence on the impact of AQR. We find that the quality of platform offerings was negatively affected after the introduction of an algorithmic quantity regulation - marked by 6% decline in ratings. Additionally, we find that the AQR affected certain platform participants disproportionately. Providers without organic and designated trust building signals, i.e., inexperienced hosts and non-superhosts, bore the cost of the AQR, ending up worse off than their counterparts
Introduction to the Minitrack on Interactive Visual Analytics for AI, Knowledge Integration, and Decision Intelligence
No full textIntroduction to the Minitrack on Advances in Digital Technologies and Vulnerabilities: New Openings in Trust, Power, and Complacency
No full textNeeds Translation: Advocating for Marginalized Communities in Tech Design
Get PDFTo address the problem of disjointed approaches to AI and equity, our proposed project develops new methods and research frameworks to interrogate how AI technologies and applications may be used for social good. We have developed a framework called “Needs Translation” that integrates and foregrounds the complex needs of the various stakeholders in the technoculture ecosystem. Though these needs are often portrayed as being at odds with each other, we contend that each need, needs translation for the other groups, so that the rights and needs of the most marginalized and/or most impacted communities are represented in the industrial decisions and design of emerging artificial intelligence, algorithmic technologies, and technology implementation. In our talk, we explore the Needs Translation framework in action by presenting a case study of work conducted by our team with a leading dating company. We examine and refine the Needs Translation framework with an industry partner
Generative AI Competencies in Demand: A Job Market Analysis of Knowledge, Skills, and Abilities
Get PDFThe emergence of generative AI (GenAI) technologies and their rapid organizational adoption are transforming workforce requirements, yet little is known about the evolving competencies needed for GenAI-related roles. It is unclear which knowledge, skills, and abilities (KSAs) are needed to work in this emerging field. To address this knowledge gap, we processed 22,352 job postings from Stepstone in the field of GenAI. Through a text mining approach combining BERTopic modeling, clustering, and expert assessment, we identified 32 distinct KSA topics and classified them into business, technical, and analytical domains. Our findings show that GenAI roles require a diverse set of skills: new GenAI-specific competencies combined with established expertise in technical domains and business functions. We contribute to research by empirically demonstrating which competencies are demanded to work successfully in the GenAI field. Understanding these competencies can guide organizations in workforce planning and support professionals in targeted skill development
Insiders Take Longer, Retail Hits Harder: Organizational Predictors of Data Breach Outcomes
No full textThis study examines the factors that influence the scale and timeliness of data breaches in an era of escalating digital threats, where the average breach cost reached $4.88 million in 2024. Analyzing 19,255 incidents from the Privacy Rights Clearinghouse using regression models, the findings reveal that insider-led breaches result in a reporting lag of approximately 79 days longer than external breaches, although affecting fewer records. Additionally, organizations in the healthcare, government, and education sectors experience breaches that are over 100% larger in scale than those in other sectors. The analysis also demonstrates that data sensitivity has a significant impact on breach severity and disclosure dynamics. These insights offer valuable contributions to crisis communication theories, inform policymakers in developing nuanced data breach laws, and equip organizational leaders to tailor incident response strategies effectively
Rationales Derived from Seeking the Stars: A Global XAI Approach for Star Rating Estimations of Reasoning LLMs Based on Textual Online Consumer Reviews
Get PDFOnline consumer reviews are a key element of ecommerce platforms. To process online consumer reviews, large language models (LLMs) are very popular. For instance, they can be used to detect reviews with inconsistencies between the sentiment in the review text and star rating. This is important as such reviews harm consumer trust on the platform. However, the internal workings of LLMs are nontransparent, thus there is a strong demand for explainable AI (XAI) approaches in e-commerce. Recently, reasoning LLMs have been proposed which offer exciting opportunities here. In this study, we present a global XAI decision tree for star rating estimations of reasoning LLMs based on the review texts by deriving features from their thinking process. Our proposed approach yields higher fidelity compared to alternatives and provides comprehensible insights on star rating estimations of reasoning LLMs. In that way, this study supports ecommerce platforms to develop more trustworthy purchasing experiences