How perceptions of intelligence and anthropomorphism affect adoption of personal intelligent agents

A personal intelligent agent (PIA) is a system that acts intelligently to assist a human using natural language. Examples include Siri and Alexa. These agents are powerful computer programs that operate autonomously and proactively, learn and adapt to change, react to the environment, complete tasks within a favorable timeframe and communicate with the user using natural language to process commands and compose replies. PIAs are different from other systems previously explored in Information Systems (IS) due to their personalized, intelligent, and human-like behavior. Drawing on research in IS and Artificial Intelligence, we build and test a model of user adoption of PIAs leveraging their uique characteristics. Analysis of data collected from an interactive lab-based study for new PIA users confirms that both perceived intelligence and anthropomorphism are significant antecedents of PIA adoption. Our findings contribute to the understanding of a quickly-changing and fast-growing set of technologies that extend users’ capabilities and their sense of self​.