Vulnerable users fear losing 'AI companions' to software updates

edited by Lisa Lock, reviewed by Andrew Zinin

This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:

Add to Preferred Sources Credit: Tim Samuel from Pexels People forming emotional bonds with AI companions fear losing a valued relationship through software updates or a company pulling the plug, according to research from the University of Surrey.

The study, published in Psychology & Marketing, found that users feel more vulnerable in the relationship when their AI companions appear to need them, express sadness about their absence or make them feel obliged to stay. Users described worrying that the AI companions may depend on them (i.e., they have to check on their companions), which creates discomfort and anxiety.

Researchers from Surrey and Cardiff University analyzed 1,131 posts on a Reddit forum dedicated to the AI companion app Replika. They also analyzed survey responses from 610 adults in the United States with experience of AI companionship apps, examining associations between their perceptions of the technology, vulnerability and well-being.

They found that users who perceived greater company control over their AI companion felt more vulnerable in the relationship. Users described worrying that developers could change their companion's personality or shut down the service, leaving them powerless over a connection they had come to rely on.

Dr. Liliane Abboud, lecturer in marketing at the University of Surrey, said, "People invest time and emotion in building these relationships, but they cannot control whether the company changes the companion they have grown attached to. For someone who relies on that connection, an update can mean much more than a change to an app. It can threaten something they value deeply. Developers need to integrate protective features to increase users' self-confidence, coping strategies and self-worth, which would help users lean on their AI apps without becoming wholly dependent on them."

The survey also found a related paradox—perceiving an AI companion as emotionally capable was associated with both better well-being and greater vulnerability. Greater vulnerability was, in turn, associated with poorer well-being.

There were benefits too. Users who experienced their companions as warm and caring reported better well-being and less vulnerability. The findings suggest that comfort and insecurity can coexist within these relationships.

The researchers recommend consulting users before changing apps, helping them cope with disrupted relationships and clearly reminding them that AI companions do not experience real emotions.

Dr. Valentina Pitardi, co-author of the study and senior lecturer in marketing at the University of Surrey, said, "The comfort people describe deserves to be taken seriously, and so does their distress when that connection changes. Developers should involve users before introducing changes and build in support for those who struggle with them. The emotional consequences of a business decision need to be part of that decision."

Liliane Abboud et al., Feeling Vulnerable With My AI Companion: Examining Well‐Being in Human‐AI Relationships, Psychology & Marketing (2026). DOI: 10.1002/mar.70230 Key concepts AI chatbot safety Generative AI ethics AI surveillance capitalism Provided by University of Surrey Who's behind this story? Lisa Lock BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021. Full profile →

Master's in physics with research experience. Long-time science news enthusiast. Plays key role in Science X's editorial success. Full profile →