Implicit theories of malleability in machines influence the perception and trust repair of intelligent agent
After an intelligent agent makes an error, trust repair can be attempted to regain lost trust. While several ways are possible, individuals' underlying perception of malleability in machines--implicit theory-- can also influence the agent's trust repair process. In this study, we investigated the influence of implicit theory of machines on intelligent agents' apology after the trust violation. A 2 (implicit theory: Incremental vs. Entity) X 2 (apology attribution: Internal vs. External) between-subject design experiment of simulated stock market investment was conducted (N = 150) via online. Participants were given a situation in which they had to make investment decisions based on the recommendation of an artificial intelligence agent. We created an investment game consist of 40 investment opportunities to see the process of trust development, trust violation, and trust repair. The results show that trust damaged less severely in Incremental rather than Entity implicit theory condition and External rather than internal attribution apology condition after the trust violation. However, trust recovered more highly in Entity-External condition. We discussed both theoretical and practical implications.