A large new U.S. study has raised concerns about the potential mental health impact of frequent artificial intelligence (AI) use, finding that people who use AI tools daily are significantly more likely to report symptoms of depression, anxiety, and irritability compared with those who never use them.
The research, published in JAMA Network Open, analyzed survey responses from 20,847 adults across all 50 U.S. states. Participants were asked how often they used AI technologies—ranging from “never” to “multiple times a day”—and whether their use was for personal, work, or educational purposes.
Key findings from the study
Researchers found that 10.3% of respondents used AI every day, including 5.3% who used it multiple times daily. Among these frequent users, nearly nine out of ten reported using AI for personal reasons, while about half used it for work and a smaller share for education.
After adjusting for income, education, age, gender, and location, the results showed a clear association: daily AI users had about a 30% higher likelihood of experiencing at least moderate depressive symptoms. Similar trends were seen for anxiety and irritability scores.
Age matters
Interestingly, the association was not uniform across all age groups. The link between frequent AI use and depressive symptoms was strongest among adults aged 25–44 and 45–64. Younger adults under 25 and seniors over 64 did not show the same pattern, suggesting that how and why AI is used at different life stages may play a role.
Personal use vs. work use
When researchers examined the purpose of AI use, they found that mental health risks were linked primarily to personal use, not to AI used for work or school. This distinction suggests that casual or social reliance on AI—rather than productivity-focused use—may be more closely tied to emotional well-being.
Experts urge caution, not panic
Speaking to Medical News Today, study author Roy H. Perlis, a professor of psychiatry at Harvard Medical School, emphasized that the findings do not prove AI causes depression.
“It’s entirely possible that people who are already experiencing depressive symptoms may be more likely to turn to AI,” he noted, adding that only long-term or randomized studies can clarify cause and effect.
Mental health professionals not involved in the study echoed this caution, suggesting that AI may sometimes replace real human interaction, potentially increasing isolation and loneliness—both known risk factors for depression.
AI as both risk and opportunity
Despite the concerns, experts stress that AI is not inherently harmful. When thoughtfully designed and used alongside human support, AI tools may help expand access to mental health resources, particularly for people who cannot easily reach a therapist.
The researchers conclude that future studies should examine specific types of AI use, track mental health changes over time, and explore how AI can be deployed responsibly to support—not replace—human connection.
