On the current state of combining human and artificial intelligence for strategic organizational decision making

Abstract

Strategic organizational decision making in today’s complex world is a dynamic process characterized by uncertainty. Therefore, diverse groups of responsible employees deal with the large amount and variety of information, which must be acquired and interpreted correctly to deduce adequate alternatives. The technological potential of artificial intelligence (AI) is expected to offer further support, although research in this regard is still developing. However, as the technology is designed to have capabilities beyond those of traditional machines, the effects on the division of tasks and the definition of roles established in the current human–machine relationship are discussed with increasing awareness. Based on a systematic literature review, combined with content analysis, this article provides an overview of the possibilities that current research identifies for integrating AI into organizational decision making under uncertainty. The findings are summarized in a conceptual model that first explains how humans can use AI for decision making under uncertainty and then identifies the challenges, pre-conditions, and consequences that must be considered. While research on organizational structures, the choice of AI application, and the possibilities of knowledge management is extensive, a clear recommendation for ethical frameworks, despite being defined as a crucial foundation, is missing. In addition, AI, other than traditional machines, can amplify problems inherent in the decision-making process rather than help to reduce them. As a result, the human responsibility increases, while the capabilities needed to use the technology differ from other machines, thus making education necessary. These findings make the study valuable for both researchers and practitioners.

Artificial Intelligence and Business Strategy towards Digital Transformation: A Research Agenda

“…Trunk et al (2020) [46] implemented a literature review to analyze the connection between AI and decision-making under dynamic environments in businesses. Authors searched for peer-reviewed papers and conducted a content analysis as a means to offer a summary of the possibilities of existing studies defining for correlating AI and business decision-making in dynamic environments.…”

“…In the initial period of AI, the most important difficulty was to execute activities that are easily rectified by a human individual, yet difficult to formally explain with regard to mathematical principles [46]. The challenge in describing this form of activity by specifying the rules showed that AI tools required the capacity to derive regularities from data and to gain their own knowledge [24,25,50].…”

“…Human reliance on the learning process reduces from the AI layer to inner layers. Traditional machine learning algorithms are a form of AI where a person has to hand-design features that the algorithm can use to present a mapping from features by eliciting patterns and gaining their respective knowledge [18,46].…”

DECAS: a modern data-driven decision theory for big data and analytics

“…Techniques must be used to assure the legitimacy, comprehensibility, sincerity, explainability, and accuracy of what they convey, and nothing should be hidden. This further requires that all involved parties understand the tasks, responsibilities, and obligations, and high level of transparency is provided (Trunk et al, 2020). Such a collaboration, if properly researched, may bring us closer to the optimised decision so inherently discussed in classical decision theory, yet so far unattainable due satisficing and the limitations of individual rationalities.…”

“…Furthermore, the flourishing coexistence of artificial intelligence (AI) supported systems with human decision makers in organisations has further ignited an interest in synergistically augmenting their intelligence and capabilities, leading to more 'intelligent' data analysis, and consequently supporting and enhancing decision making (P. Kotsiantis et al, 2006). This collaboration has led to different dimensions of intelligence and a range of applications from less to more complex, with the aim of best combining human and machine capabilities for better data-driven decision making, depending on the environment and the type of decision (Trunk et al, 2020).…”