Practices for Responsible AI: Findings from Interviews with Experts

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Akbari Ghatar, Pouria; Pappas, Ilias; Vassilakopoulou, Polyxeni

Abstract: Along with advancements in Artificial Intelligence (AI) and organizations' abil ... Expand

Abstract: Along with advancements in Artificial Intelligence (AI) and organizations' ability to use AI systems in their operations and business, several ethical and moral issues emerge. In response to managing ethical AI issues, principles for Responsible AI (RAI) have been developed. However, beyond the availability of these principles, more insights are needed on how to operationalize RAI principles. In this paper, we provide empirical findings about actual RAI practices derived from interviews with experts in America and Europe. Drawing from these empirical insights, we make three contributions to RAI literature. First, we provide a synthesis of current practices for providing fair, explainable, and inclusive AI. Second, we provide an overview of roles and structures for governing RAI in organizations. Finally, we show how these practices can be integrated into different activities within AI workflows. Collapse

Topics: artificial intelligence competitive advantage decision making LinkedIn video conference

Methods: qualitative interview literature study personal interview synthesis

Can Fact-Checking Influence User Beliefs About Misinformation Claims: An Examination of Contingent Effects

2023 | Management Information Systems Quarterly | Citations: 0

Authors: Bhattacherjee, Anol

Abstract: Prior research has suggested that corrective fact-checking has inconsistent eff ... Expand

Abstract: Prior research has suggested that corrective fact-checking has inconsistent effects on beliefs about online misinformation claims. This study attempts to explain this inconsistency using three contingent factorsclaim-source credibility, fact-checker credibility, and attitude strength-which respectively relate to three key parties in the fact-checking process: the source of a misleading claim, the fact-checker, and the user evaluating the fact-check. I hypothesize the interplay between these factors, which is tested using two online experiments on COVID-19-related misinformation with over 900 participants. Multilevel analysis of pretest-posttest, repeated measures data supports the hypothesized moderating effects and offers additional insights about how these effects vary between earlier versus later phases of misinformation cycles. The paper concludes with a discussion of contributions to research and practice. Collapse

Topics: misinformation pandemic BuzzFeed covid 19 user attitude

Methods: experiment regression analysis method longitudinal research statistical power analysis theoretical contribution

Managing the Inner Workings of Collective Intelligence Approaches for Wicked Problems – An Assessment Model and Evaluation

2023 | Communications of the Association for Information Systems | Citations: 0

Authors: Gimpel, Henner; Laubacher, Robert; Parak, Dominik; Schoch, Manfred; Wöhl, Moritz

Abstract: Please cite this article as: Gimpel, H., Laubacher, R., Parak, D., Schoch, M., ... Expand

Abstract: Please cite this article as: Gimpel, H., Laubacher, R., Parak, D., Schoch, M., & Wöhl, M. (in press). Managing the inner workings of collective intelligence approaches for wicked problems -An assessment model and evaluation. Abstract:Ill-defined and complex problems that affect multiple stakeholders with potentially conflicting perspectives are often referred to as wicked problems. The utilization of collective intelligence (CI) via web-based platforms is a promising approach for addressing such wicked problems. The management of these information systems would benefit from evidence-based decision support regarding facilitation and improvement efforts. However, to date, there is no suitable model to guide such efforts. Existing approaches address specific applications or cover certain assessment areas but do not provide a holistic perspective. Meanwhile, research offers substantial insights into best practices for addressing wicked problems and running CI applications. This paper develops an assessment model comprising five central success dimensions for information systems that address wicked problems. Their subdimensions and associated measurement metrics allow for evidence-driven facilitation and improvement of CI applications for wicked problems. Apart from the model's capability to improve future runs and processes, it also offers the potential to provide immediate insights for facilitation efforts during runtime. The model was validated with a platform dealing with the assessment of risks presented by global climate change. This evaluation generated strong evidence for the model's applicability and usefulness.Henner Gimpel is the Chair of Digital Management at the University of Hohenheim, is a member of the Research Center for Information Management (FIM), and is the academic director of the Digital Leadership Academy. His research focuses on the analysis and design of digitalization, specifically on socio-technical information systems where people, information, and digital technologies interact. His goals are to develop and use information systems that are both human-friendly and economically effective, while also analyzing individual acceptance and use of digital technologies. He often focuses on individuals in various roles, such as consumers, customers, patients, employees, and/or users who interact with information systems.Robert Laubacher examines how technological advances over the past century-most recently innovations in communications and computing-have transformed work, organizational practices, and social patterns. His work on MIT Center for Collective Intelligence's (CCI) CoLab platforms seeks to harness the collective intelligence of large numbers of people, from all around the world, to address big, complex problems. He came to MIT in the late 1990s to join the Initiative on Inventing the Organizations of the 21st Century (21C), which examined how information technology was transforming business organizations and how those changes were reshaping employment arrangements. He subsequently worked on project that examined the mechanisms by which information technology generates organizational benefits and how the structure of social networks affects the performance of work groups. Dominik Parak. Collapse

Topics: problem solving Sustainable Development Goal electronic mail computer supported cooperative work database system

Methods: qualitative interview design theory design artifact literature sample descriptive statistic

Law, Economics, and Privacy: Implications of Government Policies on Website and Third-Party Information Sharing

2023 | Information Systems Research | Citations: 0

Authors: Gopal, Ram D.; Hidaji, Hooman; Kutlu, Sule Nur; Patterson, Raymond A.; Yaraghi, Niam

Abstract: Widespread abuse of internet users’ privacy online has prompted user advocacy gr ... Expand

Abstract: Widespread abuse of internet users’ privacy online has prompted user advocacy groups to implore governments to intervene and protect consumer rights. To study such interventions’ effects, we examine data-protection policies that policy makers and governments can enforce on websites, including consent-based user information sharing and subsidizing competing websites. We use a stylized analytical model to examine such policies’ impact on the decisions and outcomes of websites, users, and third-parties. Interestingly, we find that even though a consent-based policy may improve user surplus, in the absence of market entry and exit (static market), it has the unintended consequence of increasing the number of third-parties and, thus, sharing of user information. We also determine that both consent-based and website subsidization policies may reduce competition by driving websites out of the market—to the detriment of user surplus and social welfare. To validate our analytical model’s findings, we empirically investigated the impact of a consent-based policy on third-parties in a natural experiment of the California Consumer Privacy Act. These findings raise significant implications for policy making surrounding online privacy. History: Giri Kumar Tayi, Senior Editor; Nachiketa Sahoo, Associate Editor. Funding: Financial support from the Social Sciences and Humanities Research Council of Canada is gratefully acknowledged. R. D. Gopal gratefully acknowledges support by the Gillmore Centre for Financial Technology at the Warwick Business School. Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2022.1178. Collapse

Topics: website information privacy concern data protection privacy legal context

Methods: mathematical model natural experiment experimental group longitudinal research time series analysis

Building a deep learning-based QA system from a CQA dataset

2023 | Decision Support Systems | Citations: 0

Authors: Jin, Sol; Lian, Xu; Jung, Hanearl; Park, Jinsoo; Suh, Jihae

Abstract: A man-made machine-reading comprehension (MRC) dataset is necessary to train the ... Expand

Abstract: A man-made machine-reading comprehension (MRC) dataset is necessary to train the answer extraction part of existing Question Answering (QA) systems. However, a high-quality and well-structured dataset with questionparagraph-answer pairs is not usually found in the real world. Furthermore, updating or building an MRC dataset is a challenging and costly affair. To address these shortcomings, we propose a QA system that uses a large-scale English Community Question Answering (CQA) dataset (i.e., Stack Exchange) composed of 3,081,834 questionanswer pairs. The QA system adopts a classifier-retriever-summarizer structure design. The question classifier and the answer retriever part are based on a Bidirectional Encoder Representations from Transformers (BERT) Natural Language Processing (NLP) model by Google, and the summarizer part introduces a deep learning-based Text-to-Text Transfer Transformer (T5) model to summarize the long answers. We instantiated the proposed QA system with 140 topics from the CQA dataset (including topics such as biology, law, politics, etc.) and conducted human and automatic evaluations. Our system presented encouraging results, considering that it provides highquality answers to the questions in the test set and satisfied the requirements to develop a QA system without MRC datasets. Our results show the potential of building automatic and high-performance QA systems without being limited by man-made datasets, a significant step forward in the research of open-domain or specificdomain QA systems. Collapse

Topics: privacy decision making online community IT productivity Google+

Methods: BERT chatGPT computational algorithm experiment natural language processing

e-Justice for Socioeconomic Development

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Roztocki, Narcyz; Strzelczyk, Wojciech; Weistroffer, Heinz Roland

Abstract: e-Justice, also called cyberjustice, refers to the incorporation of information ... Expand

Abstract: e-Justice, also called cyberjustice, refers to the incorporation of information and communication technologies (ICT) into the justice system. This includes offering court services via electronic means as well as other digital services for dispute resolution purposes. The objective of this study is to explore the potential contribution of e-justice as one of the components of e-democracy to socioeconomic development. Based on the analysis of the examples of e-justice initiatives, an initial, conceptual framework that links socioeconomic development with e-justice is constructed. Collapse

Topics: negotiation support system

Methods: case study

#MyIBDHistory on Twitter: Identifying Disease Characteristics Using Personal Tweets

2023 | HICSS | Citations: 0

Authors: Stemmer, Maya; Ravid, Gilad; Parmet, Yisrael

Abstract: Inflammatory bowel disease (IBD) is usually classified into Crohn's disease (CD) ... Expand

Abstract: Inflammatory bowel disease (IBD) is usually classified into Crohn's disease (CD) or ulcerative colitis (UC). Inconclusive cases are diagnosed with IBD unclassified (IBD-U). In 2018, IBD patients shared their disease history on Twitter and signed their tweets with #MyIBDHistory. In this research, we analyzed those tweets and built a logistic regression classifier that predicts patients' IBD type. We constructed tabular classification features and assessed their importance using the regression coefficients and association rules. We identified key features that distinguished CD from UC and used the classifier to predict the disease type of IBD-U patients. Our results correlated with IBD-related research. The two most prominent features that tilted the classification towards CD were suffering from fistulas or nutrient deficiencies. We identified gender differences in disease perspective prior to diagnosis. The research shows that the personal information shared by patients on Twitter can enhance existing medical knowledge regarding their disease. Collapse

Topics: Twitter database system logistics management privacy application programming interface

Methods: logistic regression lasso regression association rule learning regression analysis method descriptive statistic

To Skimp or Step in: Does Timely Service Recovery via Social Media Win Customers’ Hearts?

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Xu, Pei; Lu, Lin; Wang, Yen-Yao; Zhao, Sidi; Han, Sumin; Rao, Shashank

Abstract: Due to the public nature of the interactions and the ability to foster quicker ... Expand

Abstract: Due to the public nature of the interactions and the ability to foster quicker responses than traditional media, social media has become a powerful channel for both travelers to post their trip-related questions and issues, and airlines to build their brand images. Based on the Signaling theory and Expectations-Confirmation theory, we propose that airline service recovery via Twitter will boost profitability, and such influence is more vital for network airlines than focused airlines and airlines with higher corporate social responsibility (CSR). Collapse

Topics: Twitter social media customer satisfaction Python corporate social responsibility

Methods: computational algorithm sentiment analysis

Theories: signaling theory expectation confirmation theory

Category Evolution Under Conditions of Stigma: The Segregation of Abortion Provision into Specialist Clinics in the United States

2022 | Organization Science | Citations: 0

Authors: Augustine, Grace L.; Piazza, Alessandro

Abstract: Organizational involvement in stigmatized practices, that is, practices that att ... Expand

Abstract: Organizational involvement in stigmatized practices, that is, practices that attract substantial societal condemnation, is often challenging, inasmuch as it requires the successful management of stakeholder disapproval. In this regard, existing work on organizational stigma has highlighted the advantages of situating stigmatized practices within large, generalist organizations, because doing so allows for stigma dilution—that is, organizations can reduce stakeholder disapproval by increasing their relative engagement in uncontested practices, thereby straddling multiple categories in the eyes of audiences. This line of argument, however, runs counter to the empirical observation that stigmatized practices often remain overwhelmingly concentrated within smaller, specialist organizations, even though these are often not optimally positioned to cope with stigma. In this paper, therefore, we undertake an in-depth historical analysis of a revelatory case—abortion provision in the United States following the landmark Roe v. Wade U.S. Supreme Court decision—to build theory of how stigmatized categories can come to be populated predominantly by specialists. Building on primary and secondary archival materials, we identify three mechanisms that shaped category evolution and resulted in the de facto segregation of abortion into specialist organizations: the founding of freestanding facilities by values-driven providers, the exit of generalist organizations from the category, and the involuntary specialization of remaining providers, as customers no longer frequented them for other services and they soon became labeled simply as “abortion clinics.” We conclude by discussing the implications of our findings for the stigma literature and the generalizability of our theorizing to other settings. Collapse

Topics: strategic management

Methods: archival research case study theory development qualitative interview triangulation

Characteristics of Great Workplaces: A Text Mining Approach

2022 | Americas Conference on Information Systems | Citations: 0

Authors: Bayrak, Tuncay

Abstract: Our aim in this study is to examine Fortune magazine's "100 Best Places to Work ... Expand

Abstract: Our aim in this study is to examine Fortune magazine's "100 Best Places to Work for" in America in 2021.To identify characteristics and attributes describing the best places to work for, we performed text analytics on the comments and statements made by their employees about their working environments, social and physical conditions, scheduling flexibility, and how they were treated during the pandemic. Collapse

Topics: pandemic

Methods: natural language processing machine learning literature study