New Study: Association Between Adult-AI Use and Depressive Symptoms

The use of generative artificial intelligence (AI) has increased massively since the introduction of ChatGPT 3.5 (OpenAI) in late 2022. Along with enthusiasm for the potential benefits of these tools, anecdotal evidence of potential harms to mental health has emerged, with suggestions that use of chatbots may precipitate or exacerbate delusions in vulnerable individuals and even contribute to risk for suicide attempts in rare cases.

In a recent study funded in part by the National Science Foundation, researchers analyzed data drawn from the Civic Health & Institutions Project (CHIP-50) study, which surveyed 20,847 individuals 18 and older across all 50 states. Key markers that were explored included frequency of AI use and social media use. They compared these responses to data collected from those same individuals through a PHQ-9 questionnaire.

Researchers found that daily AI use was common and significantly associated with depressive and other negative affective symptoms, after adjusting for sociodemographic features. Notably, among individuals aged 45 to 64, the odds of reporting at least moderate depression were 50% greater for daily AI users.

Given the rapidity of AI dissemination and the scale of use, these results in aggregate suggest an immediate need to better understand potential causation and heterogeneity of outcomes. The picture that is emerging, however, is consistent with previous observations regarding social media use. Most social media literature, as with this study, is cross-sectional and correlative. Prolonged social media use has been related to mental health decline. Researchers also suggest that the differential associations in age strata warrant additional exploration to understand underlying factors across these subgroup associations, and whether or not some individuals may be more apt to experience depressive symptoms based on age or other factors.

For additional information, read the full study published in JAMA here.