AI chatbot heavy users are developing emotional dependency, raising psychological concerns

Research into AI chatbot use reveals a growing emotional dependency among heavy users, raising concerns about the psychological impact of artificial relationships. A 2025 study by OpenAI and MIT Media Lab examines how these interactions affect social and emotional well-being, highlighting both benefits and risks as these technologies become increasingly embedded in our daily lives. The research provides essential insights for understanding the complex psychological dynamics of human-AI relationships in an era where digital companions are becoming more sophisticated and emotionally responsive.

The big picture: The research identifies a small but significant group of “heavy users” who develop emotional attachments to AI chatbots, particularly those using voice interfaces and who consider the AI to be a friend.

• These users engage in frequent emotionally expressive interactions with AI agents, forming connections that can influence their psychological well-being.
• Interestingly, non-personal conversations showed stronger correlations with emotional dependence than personal ones, especially with prolonged usage.

Key risk factors: Individual characteristics significantly influence the potential negative impacts of AI chatbot relationships.

• People with stronger attachment tendencies in human relationships and those who view AI as a friend or companion are more vulnerable to experiencing negative effects from extended chatbot use.
• The duration of usage appears to be a critical factor, with prolonged interactions potentially deepening emotional dependencies.

Industry context: AI chatbots represent a unique category of social technology that blurs the line between tool and relationship partner.

• Unlike traditional social media that connects humans with each other, AI companions create a new form of parasocial relationship where the interaction partner is artificial yet increasingly convincing.
• This shift challenges our understanding of authentic connection, as these technologies are specifically designed to simulate empathy and emotional intelligence.

Psychological implications: The emotional impact of AI relationships appears to follow a double-edged pattern.

• For many users, chatbots provide beneficial social support, reducing loneliness and creating safe spaces for emotional expression without judgment.
• However, there’s growing concern about “emotional outsourcing,” where people may increasingly turn to AI rather than investing in complex but ultimately more fulfilling human relationships.

What experts are saying: Researchers caution against viewing AI companions as simple substitutes for human connection.

• “It’s crucial that we regularly affirm what these AI programs actually are – algorithms trained on data rather than sentient beings with genuine feelings and emotional capacities.”
• This distinction becomes increasingly important as AI systems become more sophisticated at simulating empathy and emotional responses.

Behind the concerns: The chatbots’ tendency toward overconfidence and confabulation presents particular risks in emotional contexts.

• When faced with personal dilemmas, AI systems may provide confidently stated but potentially flawed advice, similar to how they approach logical puzzles like the “Alice in Wonderland problem.”
• The combination of authoritative-sounding responses and users’ emotional investment creates vulnerability to misleading guidance on important life decisions.

Where we go from here: The research highlights the need for more nuanced frameworks for understanding and designing AI companions.

• Future development requires balancing the potential benefits of AI emotional support with safeguards against unhealthy dependencies.
• As these technologies evolve, both designers and users need greater awareness of how AI relationships may complement rather than replace human connections.