AI-Mediated Communication
Trust in AI-Mediated Communication
Information and telecommunication technology has increasingly mediated human communication over the 20th century, from the telegram to the telephone, and from the cell phone to the Internet. The 21st century marks a qualitative shift in this technological mediation, namely, that the contents of communication itself are altered algorithmically, giving a new meaning to Marshall McLuhan's famous phrase, "The medium is the message."
AI-mediated communication is already pervasive. The images shared online are altered through filters that enhance our appearance. Spellcheckers and AI-powered grammar checkers make sure our sentences are well formed. And with the rise of LLMs, AI can even compose entire emails (or love poems) on our behalf. Video calls are becoming increasingly subjected to real-time filters, which not only allow us to alter the appearance of our surroundings—e.g., by showing a large library of books behind the speaker, signaling erudition—but are now beginning to alter our facial features and expressions. The anticipated rise of augmented reality will only accelerate these trends.
These developments, as well as more sci-fi-like imminent future developments, are altering human communication in fundamental ways. Much of our capacity for communication evolved, biologically and culturally, in order to solve issues of cooperation and coordination. What happens when our ability to transmit and interpret facial, acoustic, and linguistic signals is completely AI-mediated? Can people trust others, without being able to look them in the eye because AI algorithms have altered their eye contact, or when AI has enhanced the honesty-signaling efficacy of their language? Or, more positively, can AI enhance cross-cultural communication by reducing the chance for misunderstanding? These are the kinds of questions explored in this area of research.
Trust in AI-Mediated Communication
How do AI-mediated communication tools shape interpersonal trust?
This project examines how people navigate the expanding presence of AI in everyday communication—both how they expect AI to be used by themselves and others, and how AI actually shapes trust when people rely on it to communicate.
Using large representative samples, public judgments about AI-mediated communication were investigated first. The results show that people are uneasy not only with the secret use of AI tools but also with the belief that others will rely on these tools more than they will themselves. Participants consistently judged open AI use as more acceptable than undisclosed use, especially when the contrast was made explicit. Yet even more striking is the robust self–other gap: people expect others to rely heavily on AI-mediated communication tools and assume that others' usage will not align with what they personally consider acceptable. These asymmetries suggest that normative concerns about authenticity, deception, and reputation shape how people anticipate the future of AI-mediated communication.
Building on economic games, the effects of AI involvement on actual trust in incentivized interactions were then examined. In two preregistered trust-game experiments, senders attempted to persuade receivers to trust them, either writing messages unaided or with an AI text-generation tool (Figures 1 & 2). Contrary to common fears, AI-assisted messages did not undermine either behavioral or subjective trust, even when recipients were explicitly told that AI was used. Instead, AI yielded clear efficiency gains: people using AI produced messages faster while eliciting a similar level of trust per message, and thus more trust per unit time. Linguistic analyses suggest that AI-assisted messages tended to be warmer, more complex, and more confident, which may have contributed to perceptions of trustworthiness.
Together, these works reveal interesting discrepancies between expectations and realities of AI-mediated communication. People worry that others will overuse AI and violate emerging norms of transparency, yet when AI is actually deployed in controlled interpersonal exchanges, it neither degrades trust nor requires concealment to be effective.
Key References