Project

Back to overview

Paradox Theory in the Digital Age: Human Actors, Intelligent Machines, and the Emerging Tensions They Create

Applicant Raisch Sebastian
Number 185164
Funding scheme Project funding (Div. I-III)
Research institution Section des Hautes Études Commerciales HEC-FSES Université de Genève
Institution of higher education University of Geneva - GE
Main discipline Science of management
Start/End 01.11.2020 - 31.10.2024
Approved amount 775'417.00
Show all

Keywords (3)

organization theory; digitalization; paradox

Lay Summary (German)

Lead
Dieses Forschungsproject befasst sich mit der digitalen Transformation und der zunehmenden Zusammenarbeit von Menschen und intelligenten Maschinen in Organisationen. Wir untersuchen die durch diese Zusammenarbeit entstehenden Spannungsfelder, Ansätze von Unternehmen mit diesen Spannungsfeldern umzugehen, sowie die positiven und negativen Konsequenzen dieser Aktivitäten für Akteure, Organisationen und die Gesellschaft.
Lay summary

Die digitale Transformation ist ein zentraler Treiber des Wandels im 21. Jahrhundert. Digitale Technologien verändern nahezu alle Bereiche des sozialen Lebens und wirtschaftlicher Aktivitäten. Die massive Verfügbarkeit von Daten, das exponentielle Wachstum der Rechenleistung und neue Techniken der Künstlichen Intelligenz verändern organisatorische Prozesse der Entscheidungsfindung, der Innovation und des Wettbewerbs. Experten bezeichnen diese Entwicklungen als eine "digitale Revolution" mit Auswirkungen auf die Weltwirtschaft wie die industrielle Revolution. 

Trotz dieser Entwicklungen befindet sich die Managementforschung zur digitalen Revolution in den Kinderschuhen. In diesem Projekt verwenden wir eine Paradox-Perspektive um die sich aus der digitalen Revolution ergebenden Spannungsfelder, Ansätze von Unternehmen zum Management dieser Spannungsfelder, sowie positive und negative Konsequenzen dieser Aktivitäten für Akteure, Organisationen und die Gesellschaft zu analysieren. Unser Fokus liegt dabei auf der zunehmenden Zusammenarbeit von Mensch und intelligenten Maschinen in Unternehmen, sowie die Herausforderungen die sich daraus ergeben.  

Das Projekt umfasst drei Studien die das Phänomen der Mensch-Maschine-Kollaboration untersuchen. Die erste Studie ist konzeptioneller Natur und zeigt auf, dass unser derzeitiger Fokus auf Menschen als einzige Akteure in Unternehmen den zunehmenden Einfluss von Intelligenten Maschinen, sowie die Konsequenzen der Zusammenarbeit von Menschen mit diesen Maschinen übersieht. Wir entwickeln deshalb eine relationale Perspektive, die erfasst wie die Zusammenarbeit von Mensch und Maschine zu Verhaltensweisen führt, die deutlich von den typischen menschlichen Verhaltensweisen abweichen. Die zweite Studie baut auf diese Perspektive auf und verwendet Twitter-Daten, um die Dynamik der Mensch-Maschine-Kollaboration im Kontext sozialer Medien zu analysieren. Wir untersuchen das Phänomen der "Filterblasen" und zeigen auf wie Algorithmen menschliche Kollektive zu zunehmend extremen Positionen motivieren. Die dritte Studie verwendet umfassende Daten einer Feldstudie von Produktentwicklungsinitiativen, um die Zusammenarbeit von Mensch und Maschine in der Entwicklung von radikalen Innovationen zu untersuchen. Etablierte Unternehmen haben oft Schwierigkeiten ihre angestammte Domäne zu verlassen und disruptive Innovationen durchzuführen. Unsere Studie untersucht inwieweit der Einsatz intelligenter Maschinen es diesen Unternehmen erlaubt, ihre Pfadabhängigkeit zu überwinden.

Die drei Studien entwickeln eine neue Perspektive auf die Zusammenarbeit von Mensch und Maschine in Unternehmen. Wir zeigen die Determinanten auf, die positive oder negative Ergebnisse dieser Zusammenarbeit determinieren. Das Projekt entwickelt Erkenntnisse für die Managementforschung und -praxis und bietet eine Grundlage für die zukünftige Diskussion der digitalen Transformation, eine der zentralsten Entwicklungen unserer Zeit.

Direct link to Lay Summary Last update: 24.01.2020

Lay Summary (English)

Lead
The project deals with the digital transformation and the increasing collaboration between humans and machines in organizations that it affords. We explore the tensions that such advanced human-machine collaborations cause, analyze possible organizational approaches to address these tensions, and clarify their positive and negative outcomes for actors, organizations, and society at large.
Lay summary

The digital transformation will be the most important change agent in the 21st century. Digital technologies are transforming virtually all areas of social life and business activity. Enabled by the massive proliferation of big data and growth in computer processing power, organizations increasingly use artificial intelligence (AI) to extract value from this vast amount of data, which allows them to get better insights, take faster and more accurate decisions, and create competitive advantages. Observers describe these developments as a “management revolution” affecting the global economy similarly to the industrial revolution.

However, management scholars have only recently begun to acknowledge AI’s importance, and there are still hardly any management studies providing up-to-date insight on this topic. In this project, we take a paradox perspective on the digital transformation, which focuses on the resulting tensions that actors, companies, and society face and possible approaches to manage them. Our particular focus is on the emerging human-machine interaction in organizations and the challenges that it creates. 

This project comprises three papers that address the phenomenon of human-machine interactions. The first study is a conceptual study, which argues that our current analytical focus on humans as the sole actors in organizations has to be extended to include the growing importance of intelligent machines and humans' interactions with these machines. We propose a more relational perspective, which considers the emergent actions that can arise from the advanced human-machine interaction in organizations. The second paper builds on this perspective and uses Twitter data to explore human-machine dynamics in the context of social media use. We specifically look at the phenomenon of "filter bubbles" to describe how algorithm-driven action can lead human collectives to escalate towards extreme positions. The third study uses rich data from a field study of product development initiatives to explore how humans and machine interact in the design of groundbreaking products. Established organizations tend to focus too much on their current directions, often ignoring the potential for disruptive innovations and exploring entirely new opportunities. Our study analyzes how the use of intelligent machines in the product search process helps human designers overcome such path dependencies. 

Together, these studies intend to present a novel perspective on human-computer collaboration in the management domain, providing insights into the contingencies that determine beneficial or detrimental performance outcomes from such collaboration. The project will develop implications for management theory and practice, providing a stepping stone for continued discussion on what is likely one of the most profound developments of our time, namely the digital transformation and intelligent machine's increasingly central role in society.

Direct link to Lay Summary Last update: 24.01.2020

Responsible applicant and co-applicants

Employees

Abstract

Paradox research has grown exponentially in recent years (Schad, Lewis, Raisch, & Smith, 2016). By focusing on the sources, nature, management, and dynamics of tensions, paradox theory aims at exploring how organizations experience and deal with the tensions they face (Smith & Lewis, 2011). This proposal explores paradox theory’s current limitations in the context of digital transformation. Digitalization is characterized by the availability of large data sets (“big data”), algorithms that semi-autonomously process this data (“artificial intelligence”), and the increasingly close collaboration of humans with intelligent machines (Daugherty & Wilson, 2018). These developments render several of the paradox theory’s limitations salient: First, paradox theory takes individual cognitive biases as an explanation for organizational actions (Lewis, 2000) and is therefore limited to human actors in organizations. Consequently, paradox theory fails to account for the emergent human-machine interaction. Second, paradox deals with a set of well-established tensions (Cunha & Putnam, 2017). Yet, digitalization creates entirely new tensions, often with wider societal implications, that remain unexplored to date. Third, paradox research focuses on how organizations manage tensions, taking a closed-systems view (Schad & Bansal, 2018). Yet, many tensions in the digital age arise from distributed and open systems that reach beyond such boundaries and therefore cannot be addressed inside the organization.We address these research opportunities in three projects. Project 1 starts with the paradox literature’s assertion that it is the managers’ responsibility to deal with tensions. Managers perceive tensions, define approaches to address them, and monitor their outcomes. Yet, in the digital age, this human-centric approach fails to explain managers’ close interaction with intelligent machines (Orlikowski & Scott, 2008). Drawing on the philosophy of information (Floridi, 2008), we argue that tensions arising from that interaction demand us to revise paradox theory’s assumptions regarding its ontology, epistemology, and axiology. In a conceptual paper, we develop a new perspective on managing tensions that is relational, emergent, and normative.Project 2 explores the dynamics of tensions, which are often characterized by escalating cycles. Prior explanations of these escalating cycles are based on individual-level cognitive biases and tend to treat all individuals within an organization the same (Fairhurst et al., 2016). Yet, individuals often display a preference for one side of a tension. External influences such as social media can reinforce their orientations further. The reasons are that algorithms create ‘filter bubbles’ that reinforce the current orientation and draw attention to extreme positions (Schad & Bansal, 2018). Drawing on the attention-based view (Ocasio, 1997), we address how escalation processes unfold through collective action in the context of social media. Using Twitter data, we plan to conduct a longitudinal quantitative study and explore the data through sentiment analysis. We contribute to the paradox literature by complementing the existing atomistic perspective with a social perspective on escalation.Project 3 challenges core assumptions in the literature on learning paradoxes. These studies emphasize organizations’ difficulties to balance exploration and exploitation (Farjoun, 2010). Organizations display a bias with a tendency to overexploit and underexplore. Yet, the human biases explaining these difficulties may be overcome in the digital age when humans and machines learn collaboratively. Drawing on computer science (Fails & Olsen, 2003), the purpose of this study is to explore how humans and machines can work together synergistically when addressing exploration-exploitation tensions. We plan to conduct a longitudinal, inductive field study in the context of product development projects. We contribute to the paradox literature by developing a perspective of ‘augmented’ learning that describes the interplay between humans and machines.
-