Regular interaction with artificial intelligence and social robots can gradually affect how people communicate, perceive themselves and shape their behavior, according to a study by scientists from Great Britain, Denmark and Sweden, published on August 19, 2026 in the journal AI & Society.
The authors of the work from the University of Birmingham, Aarhus University and Linnaeus University proposed the term "robotoid humanness" — "robotic humanity". By this term, researchers understand a possible process during which a person, in the course of repeated communication with AI, begins to adapt his own behavior to the logic of the machine — becoming more predictable, standardized and convenient for algorithmic processing.
This effect can potentially be especially noticeable in the service sector, where AI is already used in trade, hotel business, tourism and health care. Today's digital assistants and robots mimic human speech, emotional cues, and personalize responses to instill trust and facilitate customer interactions.
Researchers suggest that this process may be two-way. Machines are becoming more and more like people, but a person can also unconsciously adopt the features of communication with a machine.
"The consumer acts, the robot responds, and during the repeated interaction, the person learns this communication model over time," explains one of the authors of the study Selchen Öztürkcan from Linnaeus University. According to her, the natural tendency of people to copy the behavior of the interlocutor can lead to the fact that users begin to reproduce the communicative features of AI.
The authors describe a three-stage mechanism. At first, a person begins to perceive interaction with AI as social. The algorithm then builds a simplified digital model of the user and returns it to them in the form of personalized responses. In the future, the user can gradually adjust self-presentation and behavior to those forms that the system more easily recognizes and encourages.
One of the risks scientists call the formation of a kind of feedback, in which the algorithm constantly confirms the user's already known features. Unlike human communication, where the interlocutor may unexpectedly object, misunderstand, or offer a completely different perspective, algorithmic systems are usually tuned for consistency, convenience, and personalization. This can potentially reinforce the confirmation of existing views and reduce the variety of ways of self-expression.
At the same time, the authors emphasize an important limitation of the work: this is a theoretical study, not an experiment on users. Scientists have not yet proven that prolonged interaction with ChatGPT, robots, or other artificial intelligence systems actually makes people more "robotic." The proposed concept should be tested in further empirical studies.
The scientific article "Robotoid humanity: when selfhood becomes machine-legible" was published on August 19, 2026 in the journal AI & Society. A ScienceAlert article highlighting the research was published on August 31.
year
