According to research published by the American Psychological Association, employees who frequently interact with artificial intelligence (AI) systems are more likely to experience loneliness, which can lead to insomnia and increased drinking after work. The study, conducted across different cultures in the U.S., Taiwan, Indonesia, and Malaysia, consisted of four experiments and was published in the Journal of Applied Psychology.
Lead researcher Pok Man Tang, PhD, who previously worked in an investment bank and used AI systems, became interested in studying this issue due to the rapid advancement of AI systems and their impact on the workplace. Tang, an assistant professor of management at the University of Georgia, explains that while AI systems bring many benefits and reshape the workplace, they also pose potential dangers to employees’ mental and physical well-being. Humans are social beings, and isolating work with AI systems can have negative effects on their personal lives.
However, the researchers also found that working with AI systems can have some positive aspects. Employees who frequently used AI systems were more likely to offer help to their colleagues. This response might be triggered by their loneliness and the need for social interaction.
The studies further revealed that individuals with higher levels of attachment anxiety, which refers to feeling insecure and concerned about social connections, responded more strongly to working with AI systems. They displayed both positive reactions, such as helping others, and negative ones, such as experiencing loneliness and insomnia.
One experiment involved surveying 166 engineers at a Taiwanese biomedical company who used AI systems. The participants were asked about their feelings of loneliness, attachment anxiety, and sense of belonging over three weeks. Coworkers rated each individual on their helpful behaviors, while family members reported on their insomnia and after-work alcohol consumption. The results showed that employees who interacted more frequently with AI systems were more likely to experience loneliness, insomnia, and increased alcohol consumption after work. However, they also demonstrated some helping behaviors towards their coworkers.
Another experiment involved 126 real estate consultants at an Indonesian property management company. Half of the participants were instructed not to use AI systems for three consecutive days, while the other half were encouraged to work with AI systems as much as possible. The findings for the latter group were similar to the previous experiment, except there was no association between the frequency of AI use and after-work alcohol consumption.
Similar findings were observed in an online experiment with 214 full-time working adults in the U.S. and another experiment with 294 employees at a Malaysian tech company.
It is important to note that the research findings establish correlations and do not prove that working with AI systems directly causes loneliness or the other observed responses. However, the associations between them are evident.
Tang suggests that developers of AI technology should consider incorporating social features, such as a human-like voice, into AI systems to emulate human interactions. Employers can also limit the frequency of AI system usage and provide opportunities for employees to socialize. Tang also emphasizes the importance of involving humans in team decision-making and other tasks that require social connections, while AI systems can handle repetitive and tedious tasks.
To address loneliness, Tang proposes the implementation of mindfulness programs and other positive interventions. Since AI will continue to expand, action is necessary now to mitigate the potentially detrimental effects on individuals who work with these systems.
For more information, please contact Pok Man Tang, PhD, at [email protected].
ARTICLE: “No Person Is an Island: Unpacking the Work and After-work Consequences of Interacting with Artificial Intelligence” by Pok Man Tang, PhD (University of Georgia), Joel Koopman, PhD (Texas A&M University), Ke Michael Mai, PhD, and David De Cremer, PhD (National University of Singapore), Jack H. Zhang, PhD (Nanyang Business School), Philipp Reynders, PhD (Cardiff University), Chin Tung Stewart Ng,
MSc, and I-Heng Chen, PhD (National Sun Yat-sen University). Published online in the Journal of Applied Psychology on June 12, 2023.