AI and Big Data in Industrial-Organizational Psychology

Introduction

Today, artificial intelligence (AI) is widely regarded as an important innovation that will fundamentally change the manner in which humans perform their day-to-day workplace activities, with special implications for a number of service and industrial sectors. Characterized by some observers as an existential threat and by others as a boon to humankind's efforts to facilitate decision-making and improve the effectiveness of a wide array of automated applications, AI is currently having a major impact on industrial-organizational (I-O) psychology in some unexpected ways. Although definitions vary, some indication of the pervasiveness of AI in modern organizational settings can be discerned from the following:

Taken together, it is clear that AI has already had a major impact on organizations of all sizes and types, and its potential for even greater contributions—and challenges—is limited only by computer processing speeds and human imagination. The purpose of this study was to provide a systematic review of the relevant literature together with a qualitative meta-analysis of recent studies concerning how artificial intelligence and big data can lead to new theoretical insights in I-O psychology, spanning a wide range of content domains such as attitudes, well-being, withdrawal, stress and burnout, teams, socialization, social networks, work/family balance, and diversity and inclusion.

Although many meta-analyses are strictly quantitative in design, a growing number of social science researchers have recognized the benefits of using a qualitative research design to develop timely and informed answers to complex research questions (Neuman, 2008). Given the relative recentness of many AI innovations and their increasing application to a wider array of workplace settings, a qualitative meta-analysis using a systematic review of relevant peer-reviewed and scholarly literature published within the last 24 months in the English language—with a special focus on other meta-analyses—was regarded as the optimal strategy for determining the current and anticipated impact of AI on I-O psychology. Keyword search terms included "I-O psychology," "big data," "artificial intelligence," and "machine learning," as well as various combinations thereof. The key findings that emerged from this systematic review and analysis were organized thematically and are presented below.

Artificial Intelligence and Big Data

Artificial intelligence is an umbrella term that subsumes an entire new generation of technologies capable of interacting with the external environment in ways that increasingly simulate human intelligence (Glickson & Woolley, 2020). Therefore, the successful implementation of AI into organizational computer networks is inextricably tied to human trust in AI technologies. AI technologies are characterized as being situated at the foundation of what has been termed the "fourth industrial revolution" by some authorities, with the basic transfer of individual agency and autonomy from human beings to computer-assisted technology representing the core of the process (Glickson & Woolley, 2020). Not surprisingly, this transitional process also invokes significant changes in the understanding of the relationship between humans and technology that will require further examination in order to optimize outcomes (Glickson & Woolley, 2020).

Noting that research to date has produced mixed results, a study by Lee et al. (2020) examined the potential effects of AI as well as the "big data" generated by social media platforms on human-technology interactions. One of the more noteworthy findings from that study was that "there is growing recognition that AI models that use publicly available big data can identify geographical differences in entrepreneurial personality or culture" (p. 2).

The cost-effectiveness of this approach over conventional marketing research and analysis is clear, but there were other important issues related to the Lee et al. (2020) study that warrant noting. The findings were regarded as especially trustworthy given that the study's results were based on a machine learning model that examined 1.5 billion tweets made by 5.25 million Twitter users to provide an estimation of the Big Five personality traits (extraversion, agreeableness, openness, conscientiousness, and neuroticism) and to create an entrepreneurial profile for 1,772 counties in the United States. According to Lee et al. (2020), "The Twitter-based personality estimates show substantial relationships to county-level entrepreneurship activity which represented 20% (entrepreneurial personality profile) and 32% (Big Five traits) of the variance in local entrepreneurship" (p. 3).

It is also noteworthy that these findings held true despite the researchers controlling for various other factors believed to affect entrepreneurship levels. Based on these findings, Lee et al. (2020) concluded that while additional research is clearly needed, the results indicate that AI applied to big data available through social media platforms can provide marketing researchers with timely and valuable insights into personality and culture that would not be otherwise available using conventional survey-based personality tests.

The ability to facilitate knowledge sharing, information exchange, and work collaboration has fueled growing interest in big data generated by enterprise social media (ESM) (Wu et al., 2021). These technologies have been widely embraced by business leaders to improve job performance; however, with the deepening of empirical research and practice, ESM usage has also been found to yield various negative outcomes, such as information overload, privacy invasion, turnover intention, and work-life conflicts. Over time, these negative outcomes may be the source of diminished job performance and employee morale (Wu et al., 2021). The results of this study indicate that there is a significant positive correlation between ESM usage and job performance as mediated by several factors, including job level (front-line or manager), gender-related differences, and differences in acceptance levels for new technologies attributable to national context (i.e., developed versus developing economies). These findings have a number of important implications for I-O psychological practitioners, as well as providing the foundation for additional research in this area.

Although a growing number of enterprises report using AI and machine learning algorithms to solve a wide range of organizational problems based on their ability to formulate accurate predictions by analyzing big data, there remains an urgent need to further investigate the impact of AI in various industrial settings to determine its effect on human-technology interactions. To address this gap, Pizzi et al. (2020) first define AI as "a new generation of technologies capable of interacting with the environment" (p. 1). In addition, Pizzi and colleagues report that AI achieves these outcomes by:

While the interactional characteristics of AI have been shown to facilitate machine learning, these technologies also cause corresponding changes in behavior resulting from external environmental stimuli in ways that resemble and even mimic the way humans learn. In sum, these findings suggest that AI will continue to increase in its capabilities, limited only by computer processing speeds—which have roughly maintained pace with Moore's Law—and the humans who design these systems. These findings also underscore the need for continuing research regarding the effects of these trends on the human beings who use them (Pizzi et al., 2020).

Impact of AI-Enabled Technologies on the Workplace

Singh et al. (2019) cite the proliferation of AI-enabled technologies as already having a profound impact on the workplace, with a virtually limitless future in store. To gain fresh insights into the precise effects of these trends, the purpose of the Singh et al. (2019) study was to explicate the influence of AI on computer-augmented sales technologies in order to identify emerging opportunities and corresponding threats to: (a) the sales profession in terms of its contribution to creating value for customers, organizations, and society; and (b) sales professionals, in terms of both employees in organizations and individuals seeking growth, fulfillment, and status in the functions they serve and roles they live (p. 2).

The key findings that emerged from this study included the increasing potential for AI to automate sales processes from start to finish, including customer acquisition and retention strategies in ways never before possible. In addition, Singh et al. (2019) conclude that the extent to which companies succeed in leveraging their AI resources to automate the retail sales process will likely be the extent to which they are able to develop and sustain a competitive advantage in an increasingly globalized marketplace.

The findings by Singh et al. (2019) served as a useful background for a subsequent study by Schepers and van der Borgh (2020). That study was an ambitious meta-analysis of 105 articles from 35 different countries, which found that customer expectations are the strongest antecedents to both in-role and extra-role behavior by frontline customer service employees, but that these expectations differ significantly across cultures. In support of this assertion, the authors cite several examples of frontline employees "going the extra mile" for customers—such as financial service experts assisting their clients in finding celebration venues and an airline ticket agent hurrying to place a hold on an airplane to accommodate late-arriving passengers. Although rare in the real-world workplace, these types of customer-directed extra-role behaviors can be facilitated through automated technologies while still having the bottom-line effect of promoting customer goodwill and loyalty (Schepers & van der Borgh, 2020).

Because AI-enabled technologies do not require coffee breaks, vacations, or sick days, it is little wonder that their use has become increasingly commonplace in many manufacturing settings. These same benefits can also be achieved in various service settings whose roles were limited to humans until the recent past (Feng et al., 2019). In response to these trends, a growing number of service industries are providing employees with self-service technologies (SSTs) in order to reduce costs and increase efficiency.

Not surprisingly, the humans who survive this replacement process may become modern-day Luddites who resist the use of SSTs in ways that sabotage their effectiveness and limit the ability of companies to achieve the full range of benefits these technologies can otherwise provide. Therefore, this potentially powerful constraint on human interactions with AI-enabled technologies must be taken into account during any SST implementation in customer service settings. An especially significant finding from the Feng et al. (2019) study was that negative reactions to SSTs are a universal human response to perceived threats. These responses vary from person to person but typically include negative emotions and perceptions toward SSTs as well as resistance to their further adoption in the workplace (Feng et al., 2019).

In sharp contrast to this potential Luddite-like reaction, a study by Mfanafuthi et al. (2019) cited the lengthy historical record confirming that automation will continue to replace human labor wherever possible and feasible, most especially for jobs that involve repetitive tasks. Although additional research is needed, the studies to date indicate that AI and robotics are already having a profound effect on these types of traditional occupations, and it is reasonable to posit that these trends will accelerate well into the foreseeable future as the fourth industrial revolution continues to redefine human jobs.

The uncertainty surrounding the future of human labor has intensified in recent years as a result of the proliferation of AI and robotic technologies. The concept is being applied in a number of settings under various umbrella terms such as computerization, automation, and technological advancement. While the specific impact of AI and robotics remains uncertain, their effect has been most notably felt in the service, production, office, and administration sectors (Mfanafuthi et al., 2019).

Certainly, other sectors are also already benefiting from AI-enabled technologies. The number of organizations implementing AI grew 270% since 2017 and tripled since early 2020 (Mikalef & Gupta, 2021). A study by Mikalef and Gupta (2021), however, found that the proliferation of these automated strategies has been limited due to problems associated with their implementation and the need for fundamental organizational restructuring to achieve optimal outcomes. Therefore, organizations must also invest in complementary resources in order to realize the maximum return on their AI investments and reap the benefits of increasingly accessible big data resources. Because every organization is unique, the complementary resources needed will also differ, but properly implemented and administered, AI has the potential to increase organizational creativity and performance (Mikalef & Gupta, 2021).

A study by Reis et al. (2019) also investigated which sectors AI would likely impact most in the coming years. Based on their findings, these researchers conclude that the transport sector will likely experience the most significant increases in the use of AI-enabled technologies. This prediction is based on current trends toward the commercialization and proliferation of autonomous vehicles as well as the corresponding uptake by the general public. These trends, combined with the potential of AI and big data analyses, may help identify opportunities to improve worker acceptance of these technologies in the transport sector and others (Reis et al., 2019). In addition, rather than using wholesale, across-the-board downsizing approaches to accommodate the proliferation of AI-enabled technologies, Reis et al. (2019) recommend that employees should be retrained for the new types of jobs that will emerge following the further automation of existing processes.

References

Braganza, A. et al. (2021). Productive employment and decent work: The impact of AI adoption on psychological contracts, job engagement and employee trust. Journal of Business Research. [in press]

Chassignol, M. et al. (2018). Artificial intelligence trends in education, a narrative overview. Procedia Computer Science, 136, 16–24.

Damianidou, D. et al. (2019). Technology use to support employment-related outcomes for people with intellectual and developmental disability: An updated meta-analysis. International Journal of Developmental Disabilities, 5(4), 220–290.

Feinzig, S. L. (2020). A perspective on bringing AI to HR: Opportunities, risks, and a recommended path forward. Workforce Solutions Review, 2nd Quarter, 21–27.

Feng, W. et al. (2019). Understanding forced adoption of self-service technology: The impacts of users' psychological reactance. Behavior & Information Technology, 38(8), 820–832.

Glickson, E. & Woolley, A. W. (2020). Human trust in artificial intelligence: Review of empirical evidence. Academy of Management Annals, 14(2), 627–667.

Kaltenegger, H. C. et al. (2020). Association of working conditions including digital technology use and systemic inflammation among employees: Study protocol for a systematic review. Systematic Reviews, 9(221), 1–11.

Koo, B. (2021). Examining the impact of artificial intelligence on hotel employees through job insecurity perspectives. International Journal of Hospitality Management. [in press]

Lee, A. et al. (2021). Determining causal relationships in leadership research using machine learning: The powerful synergy of experiments and data science. The Leadership Quarterly. [in press]

Lee, M. O. et al. (2020). Big data methods, social media, and the psychology of entrepreneurial regions: Capturing cross-county personality traits and their impact on entrepreneurship in the USA. Small Business Economics, 55, 567–588.

Mfanafuthi, M. et al. (2019). Analysis of the impact of artificial intelligence and robotics on human labor. Gender & Behavior, 13877–13891.

Mikalef, P. & Gupta, M. (2021). Artificial intelligence capability: Conceptualization, measurement calibration, and empirical study on its impact on organizational creativity and firm performance. Information & Management, 58, 103–134.

Neuman, W. L. (2008). Social research methods: Qualitative and quantitative approaches. Allyn & Bacon.

Pizzi, G. et al. (2020). Artificial intelligence and the new forms of interaction: Who has the control when interacting with a chatbot? Journal of Business Research. [in press]

Reis, J. et al. (2019, June). Impacts of artificial intelligence on public administration: A systematic literature review. 2019 14th Iberian Conference on Information Systems and Technologies (CISTI). June 19–22, 2019, Coimbra, Portugal.

Shaohua, L. & Shorey, S. (2021). Psychosocial interventions on psychological outcomes of parents with perinatal loss: A systematic review and meta-analysis. International Journal of Nursing Studies, 117, 1–12.

Singh, J. et al. (2019). Sales profession and professionals in the age of digitization and artificial intelligence technologies: Concepts, priorities, and questions. Journal of Personal Selling & Sales Management, 39(1–2), 2–22.

Sun, S. et al. (2020). The impact of cultural values on the acceptance of hotel technology adoption from the perspective of hotel employees. Journal of Hospitality and Tourism Management, 44, 63–69.

Trang, S. & Brendel, B. (2019). A meta-analysis of deterrence theory in information security policy compliance research. Information Systems Frontiers, 21, 1265–1284.

Wu, C. et al. (2021). Does enterprise social media usage make the employee more productive? A meta-analysis. Telematics and Informatics, 60, 101578.