New Study Suggests that Chatbot Companions Can Leave us Lonelier

Jean Rhodes

In a new study in Psychological Science, researchers Folk and Dunn (2026) followed more than 2,000 adults (average age 40) in the US, UK, Canada, and Australia for over a year. Every four months, participants reported on their use of chatbots for things like reflecting on their day, seeking advice on life decisions, and seeking companionship. They also answered questions about their feelings of isolation and connection to others.

The findings suggested some reasons for concern. The authors state, “Turning to AI for companionship predicted increased emotional isolation 4 months later” (Folk & Dunn, 2026, p. 284). There was also some evidence that feeling isolated predicted greater chatbot use, although this finding was less consistent across analyses. This raises the possibility that people who are struggling with loneliness may turn to chatbots for support without finding lasting relief.

The findings were somewhat different when the researchers used a broader measure of social connection. People who felt less connected to others later turned to chatbots more often. But using chatbots more often did not predict a later decline in their overall sense of social connection. Nor did it predict “subsequent changes in either social support or number of close friends” (Folk & Dunn, 2026, p. 284). So, although greater chatbot use predicted feeling more isolated, the study did not show that people were actually replacing their human relationships with chatbot conversations.

The authors themselves note, “We urge caution, however, in drawing strong conclusions given the exploratory nature of our analyses” (Folk & Dunn, 2026, p. 276). The associations were small, and only 46% of participants completed all four surveys. Given participants’ average age, the findings may not generalize to the young people served by mentoring programs. And although following people over time helps clarify the sequence of events, it does not establish that chatbot use caused greater isolation. Still, the findings give us reason to ask whether turning to AI for companionship is meeting the needs that bring people to it in the first place.

The findings help put earlier research in perspective. De Freitas et al. (2026) found that AI companions could provide momentary reductions in loneliness, including after repeated use over a week. More recently, Li et al. (2026) randomly assigned first-year university students to exchange daily messages with a fellow student or a supportive chatbot, or to write brief journal entries. After 2 weeks, the human-partner group reported less loneliness than either comparison group, while the chatbot group did not differ significantly from the journaling group (Li et al., 2026). These studies suggest that feeling better after a conversation and developing a more satisfying social life may follow different paths.

References

De Freitas, J., Oğuz-Uğuralp, Z., Uğuralp, A. K., & Puntoni, S. (2026). AI companions reduce loneliness. Journal of Consumer Research, 52(6), 1126-1148. https://doi.org/10.1093/jcr/ucaf040

Folk, D., & Dunn, E. (2026). How does turning to AI for companionship predict loneliness and vice versa? Psychological Science, 37(4), 276-286. https://doi.org/10.1177/09567976261427747

Li, R.-N., Folk, D., Singh, A., Ungar, L., & Dunn, E. (2026). Is a random human peer better than a highly supportive chatbot in reducing loneliness over time? Journal of Experimental Social Psychology, 125, Article 104911. https://doi.org/10.1016/j.jesp.2026.104911