Computed compatibility: examining user perceptions of AI and matchmaking algorithms

Aditi Paul,Saifuddin Ahmed

DOI: https://doi.org/10.1080/0144929X.2023.2196579

2023-04-16

Behaviour and Information Technology

Abstract:Artificial intelligence (AI) driven matchmaking algorithms are at the core of modern-day dating. Millions of users rely on these algorithms used by online dating platforms for successful matchmaking. However, a scholarly understanding of user perceptions of AI-driven matchmaking algorithms is limited. We explore the factors affecting users' perceived effectiveness of matchmaking algorithms and analyze how users' perception of AI's fairness, social presence, and the threat posed by AI are associated with their perceived effectiveness of matchmaking algorithms. We also investigate if their previous relationship initiation experience through online dating platforms further moderates the studied relationships. An analysis of survey data from Singapore suggests that those who perceive AI to be fair and have higher levels of social presence are more likely to showcase a higher degree of perceived effectiveness of matchmaking algorithms. Moreover, those who have previously been successful in online dating relationship initiation are also more likely to believe in the perceived effectiveness of these algorithms. Further, previous experience of relationship initiation conditionally impacts the relationship between users' general AI perceptions and perceived effectiveness. We also find that males and younger respondents are more likely to believe in the efficacy of matchmaking algorithms. Practical implications are offered.

ergonomics,computer science, cybernetics