AI, Organizing, and Management

Permanent URI for this collectionhttps://hdl.handle.net/10125/107548

Browse

Now showing 1 - 10 of 10

Streamlining the Operation of AI Systems: Examining MLOps Maturity at an Automotive Firm
(2024-01-03) Weber, Michael; Schniertshauer, Johannes; Przybilla, Leonard; Hein, Andreas; Weking, Jörg; Krcmar, Helmut
Developing and operating AI systems based on machine learning (ML) has unique challenges that render traditional practices inappropriate (e.g., managing data drift). To that end, MLOps emerged as a novel paradigm for managers and teams to develop and operate such ML systems successfully. Organizations currently employ different maturity levels for MLOps, whereas higher maturity typically corresponds to more automated, streamlined, and reliable workflows. However, we have limited insight into factors influencing MLOps maturity in ML projects. Therefore, we conducted a case study on MLOps maturity in three ML projects at an automotive firm. We identified several contextual factors that facilitate or inhibit MLOps maturity, such as the ML model’s complexity, the quality of new data, and the appropriateness of available MLOps tools. Our study contributes to research on managing and organizing AI by providing factors that explain the different adoption of MLOps in practice.
Validating Generalizations about AI and Its Uses
(2024-01-03) Alter, Steven
Many discussions of the tremendously important topic of AI are undermined by vague definitions of what AI is and by generalizations that are too distant from reality to be useful when pursuing or evaluating real world applications. Inconsistent and exaggerated definitions of AI published during 2019-2022 illustrate the need for an approach to validating current generalizations about AI and its uses and not just repeating decades-old predictions and images based on science fiction movies. Next, this paper presents a series of evaluation issues for generalizations and a series of ideas that are useful for describing AI applications as uses of algorithms based on techniques associated with AI. It incorporates those ideas into summaries of six diverse applications of algorithms associated with AI. A concluding section presents suggestions for realistic descriptions of AI applications and for generalizations about AI.
Yes, but …: Unraveling Paradoxes in Implementing Artificial Intelligence
(2024-01-03) Finze, Nikola; Zimmermann, Sina; Weeger, Andy; Wagner, Heinz-Theo
Implementing artificial intelligence (AI) applications in firms promises great potential but poses complex challenges. Especially incumbent firms often struggle to use the full potential of AI, because of paradoxes that arise in the context of implementing AI solutions, such as concerns regarding data privacy but simultaneously sharing personal data excessively. To analyze what paradoxes are caused by the challenge to implement AI in incumbent firms, we draw on the literature on technological paradoxes and followed a qualitative research approach using semi-structured interviews in eight companies on the path to AI implementation. Our results unravel that various mismatches between strategic imperatives and tactical paradigms emerge from three AI paradoxes: the privacy paradox, the potential paradox, and the integration paradox. Our results contribute to the information systems literature on AI and technological paradoxes by providing novel empirical insights on AI paradoxes and practical implications to address these paradoxes in incumbent firms.
AI Project and Deployment Risk: Articulation and Legitimization
(2024-01-03) Lahiri, Sucheta; Saltz, Jeffrey
This study explores how practitioners identify and manage AI project-related risks to reduce AI project failures. Specifically, through a qualitative research study involving 16 data science practitioners, this study presents insights into how practitioners articulate and mitigate the risk of AI project failure. A thematic analysis of this study identified six key themes (Ethical risk, BlackBox Models, Data Privacy, Data Storage, Financial Risks, and Success criteria). Further analysis explored drivers for identifying and mitigating these risks. Specifically, it was found that agency (consumer and institutional-driven) and social/cultural capital (such as management hierarchy and domain knowledge) legitimized specific AI project risks and were key drivers in ensuring risks were identified and mitigated. Results from this research suggest that future research should explore different social and cultural perspectives since these perspectives can impact the articulation of risk and how these risks can be ultimately managed within an AI project context.
Algorithmic Accountability: What Does it Mean for AI Developers and How Does it Affect AI Development Projects
(2024-01-03) Bartsch, Sebastian; Milani, Verena; Adam, Martin; Benlian, Alexander
Algorithmic accountability obligates developers to justify themselves for their artificial intelligence (AI)-based systems. Despite this positive effect, there is still insufficient information system (IS) research on how developers’ perceived algorithmic accountability can be increased and how it affects AI development projects. Within our qualitative interview study, we asked 25 developers about algorithmic accountability during their AI development projects. We observe that developers’ perceived algorithmic accountability depends on organizational factors (i.e., quality management, working method, company structure, and the facets of AI) and personal factors (i.e., understanding of AI-based systems and algorithmic accountability), leading to more scrutinized AI-based systems. Overall, this study contributes to IS development (ISD) research by providing transparency on how developers’ perceived algorithmic accountability is affected and how it affects AI development projects. These findings are also relevant for practitioners, as we suggest how they can shape their work environment to promote the positive effects of algorithmic accountability.
The Return on Investment in AI Ethics: A Holistic Framework
(2024-01-03) Bevilacqua, Marialena; Berente, Nicholas; Domin, Heather; Goehring, Brian; Rossi, Francesca
We propose a Holistic Return on Ethics (HROE) framework for understanding the return on organizational investments in artificial intelligence (AI) ethics efforts. This framework is useful for organizations that wish to quantify the return for their investment decisions. The framework identifies the direct economic returns of such investments, the indirect paths to return through intangibles associated with organizational reputation, and real options associated with capabilities. The holistic framework ultimately provides organizations with the competency to employ and justify AI ethics investments.
Transition to Human-AI Work: Shifts in Routines' Dynamics and the Implications for Roles in Knowledge-Intensive Work
(2024-01-03) Ruissalo, Joona
As solutions based on artificial intelligence grow pervasive in knowledge-work organizations, such cognitive technology is being applied both to automate and to augment work heretofore carried out predominantly by humans. This has profound socio-technical implications for the work practices in that new means of conducting associated routines are changing the knowledge workers’ involvement, transforming interaction between the human agents and information systems within the socio-technical system. Taking a processual approach to exploring how such deep transformation extends to the core of the knowledge workers’ work roles, an interpretive study examined the role transformation unfolding over two years at a financial-accounting company that was developing and implementing an artificial-intelligence system for its services. This processual study offers empirically grounded contributions in outlining how digital transformation changes routines and work roles in knowledge-intensive work.
AI Implementation and Capability Development in Manufacturing: An Action Research Case
(2024-01-03) Eklof, Jon; Snis, Ulrika; Hamelryck, Thomas; Grima, Alexander; Rønning, Ola
This action research article presents a case study of a global manufacturing company deploying artificial intelligence (AI) to develop capabilities and enhance decision-making. This study explores considerations and trade-offs involved in introducing AI into daily operations, leading up to the decision to develop AI capabilities in-house or outsource them. The case study offers in-depth technical descriptions of model selection, dataset creation, model adoption, model training and evaluation while addressing organizational obstacles and decision-making processes. The study’s findings highlight the importance of collaboration between technical experts, business leaders, and end-users, as well as the interaction and collaboration between AI systems and human employees in the workplace. The article contributes a practical perspective on AI implementation in manufacturing, emphasizing the need to balance in-house capability development with external acquisition. Although the case study company managed to create an in-house model, factors such as implementation, debugging, data requirements, training time, and performance led to outsourcing the capabilities. However, making this informed decision required capabilities and insights that were acquired through practical work. Consequently, although in-house development can be challenging, it can also enhance organizational capabilities and provide the necessary knowledge to make informed decisions about future development or outsourcing.
Organizational Challenges in Adoption and Implementation of Artificial Intelligence
(2024-01-03) Raftopoulos, Marigo
Our investigation into the organisational challenges in the adoption and implementation of artificial intelligence reveals complex dynamics in the interplay between strategic decision-making, implementation, enablement and performance outcomes. We undertook a study of the views of international industry experts on the enablers and barriers of adopting and implementing AI and found five thematic clusters of issues affecting project success and value creation. We contribute to theory development and a conceptual model on navigating organisational adoption and implementation of AI technology.
Introduction to the Minitrack on AI, Organizing, and Management
(2024-01-03) Nickerson, Jeff; Lindberg, Aron; Seidel, Stefan; Saltz, Jeffrey

Browse

Recent Submissions