The "Artificial" Colleague: Evaluation of Work Satisfaction in Collaboration with Non-human Coworkers

Title of the Paper

The "Artificial" Colleague: Evaluation of Work Satisfaction in Collaboration with Non-human Coworkers

Bibliographic Information

• Field of Study: Artificial Intelligence and Human-Computer Collaboration
• Keywords: Human-AI collaboration, work satisfaction, meaningful work, work automation, human-machine teamwork

Research Background and Problem

• Identified Issues:
  • With the development of artificial intelligence (AI) technologies, work patterns are undergoing profound changes. However, existing research primarily focuses on the efficiency and performance of AI collaboration, with limited attention to the impact of AI on work meaning and job satisfaction.
  • Collaboration with AI can influence human autonomy and social relationships at work, thereby affecting the sense of work meaning in different ways.
• Research Significance: Work constitutes a significant part of human life, and its sense of meaning directly affects overall well-being. Understanding the impact of AI collaboration on work meaning is crucial for designing workplaces that promote human welfare.
• Research Motivation and Related Work:
  • Current AI technology design emphasizes efficiency and performance, without adequately considering how to foster meaningful and experience-oriented work design.
  • There is a lack of clear understanding regarding the social role of AI. For instance, should AI be regarded as a team member or merely a tool?

Solution

• Research Methods:
  • A virtual scenario-based experimental questionnaire method was employed, designing two specific scenarios to explore how changes in task allocation and collaboration partners (human or AI) influence job satisfaction and collaborative relationships.
  • Scenarios simulated collaboration with either humans or AI to complete two types of tasks (meaningful tasks and routine tasks).
• Research Innovation:
  • The study approaches human-AI collaboration from the perspective of experience and meaning, rather than focusing solely on performance.
  • It investigates the different relational roles AI can play as a "colleague" in the workplace (e.g., subordinate, partner, superior).
• Implementation Steps and Techniques:
  • Surveys were used to collect participants' satisfaction ratings for different tasks and collaborators.
  • The Job Diagnostic Survey (JDS) was employed to measure job characteristics and perceptions of collaborative roles.
  • Statistical analysis was conducted to compare the effects of collaboration partners (human/AI) and task allocation on key variables such as satisfaction and relational perception.

Research Findings

• Specific Findings:
  • Even under identical tasks with positive outcomes, collaboration with humans was perceived as more meaningful and satisfying than collaboration with AI.
  • More meaningful tasks (e.g., content preparation tasks) generally resulted in higher job satisfaction.
  • AI was more often perceived as a subordinate, while humans were more frequently regarded as team partners.
  • People rated AI performance lower, even when it was designed to perform at a human-equivalent level.
• Experimental Results:
  • Tasks with high meaningfulness (e.g., content preparation) significantly increased the Motivating Potential Score (MPS).
  • Human colleagues were perceived as more competent and provided greater skill variety, task significance, and autonomy compared to AI.
  • Task allocation had no significant impact on perceptions of human-AI relationships, but AI's "instrumental" nature was more pronounced.
• Comparison with Existing Approaches:
  • Beyond the traditional automation design focus on efficiency, this study highlights the importance of experience-oriented goals (e.g., work meaning and satisfaction).
  • By addressing psychological needs in work (e.g., self-worth, social connection), the study emphasizes more human-centered AI collaboration design.
• Limitations and Future Directions:
  • The reliance on virtual scenario studies captures only anticipated experiences, rather than real-world workplace feelings.
  • The tasks involved were relatively independent, lacking scenarios with higher interactivity in collaboration.
  • Future research should further explore how natural interaction designs for AI influence job satisfaction and social relationships.

Conclusion

Through experimental research, this study reveals that, under equivalent conditions, human colleagues bring more meaning and satisfaction to work compared to AI. By optimizing task allocation and designing more human-centered AI interactions, it is possible to better meet psychological needs for autonomy, skill value, and social connection, thereby enhancing the experiential outcomes of human-AI collaboration in the workplace. Future research should delve deeper into design principles for more complex collaboration scenarios to balance the dual goals of performance and work meaning.