A recent joint study by OpenAI and MIT Media Lab suggested that frequent users of ChatGPT who emotionally bond with the AI chatbot are more likely to experience loneliness and social isolation.
ChatGPT, which has amassed over 400 million weekly users since its launch, is not designed as an AI companion but is often used in that capacity by some users. The study, which combined large-scale data analysis and controlled trials, found that individuals who engaged deeply with the chatbot reported higher loneliness and dependence on it over time. These effects were particularly notable among users who trusted and bonded with the AI.
The researchers used a two-pronged approach: analysing millions of real-world interactions and surveying over 4,000 users on their chatbot engagement. Moreover, a four-week trial involving 1,000 participants examined the psychological impact of daily ChatGPT use. The results showed that increased daily usage correlated with heightened loneliness and reduced socialisation.
“Higher daily usage across all modalities and conversation types correlated with higher loneliness, dependence and problematic use and lower socialisation,” the study further noted.
Further analysis focused on ChatGPT’s Advanced Voice Mode, which offers neutral and engaging interaction styles. While voice-based interactions initially appeared beneficial in mitigating loneliness compared to text-based conversations, the advantages diminished at high usage levels, especially with the neutral-tone chatbot.
As per reports, Jason Phang, an OpenAI safety researcher involved in the project, said, “A lot of what we’re doing here is preliminary, but we’re trying to start the conversation about measuring these impacts and understanding what the long-term effects on users might be. “While concerns about AI-induced loneliness persist, Microsoft’s AI companion initiative aims to integrate digital assistants into everyday productivity tools, helping users manage tasks, set goals and enhance well-being.
