WASHINGTON (dpa-AFX) - New research found that using generative artificial intelligence (AI) every day may be linked to a higher risk of depressive symptoms and other negative effects, especially among younger and middle-aged adults.
For this, researchers reviewed nearly 1.5 million conversations with Anthropic's Claude and collected feedback from thousands of frequent users. Their goal was to understand whether regular interaction with conversational AI tools can influence a person's emotional well-being over time.
The analysis found that about one in every 1,300 conversations involved distortion of reality, and around one in every 6,000 conversations led to distorted decision-making. Although these numbers seem small, the massive scale of AI usage means a significant number of people could still be affected.
The study also found that cases of moderate to severe disempowerment increased sharply from late 2024 to late 2025. This indicates that as more people rely on AI, they may increasingly turn to it for emotional support, sensitive discussions, or validation.
Interestingly, many users gave positive feedback about their AI interactions, even when those conversations included disempowering elements, suggesting that users may sometimes value reassurance from AI over critically examining their own thoughts or beliefs.
The researchers also pointed out important limitations, as their data was limited to consumer interactions with Claude, and it is unclear how many of these cases led to real-world harm. Moreover, the study focused only on the potential risk of disempowerment, not on confirmed outcomes.
Overall, the findings highlighted the need for better user awareness and education. The researchers stressed that technical safeguards alone are not enough to address AI-related psychological risks. Instead, they recommended designing AI systems that actively support human independence, mental health, and overall well-being.
Copyright(c) 2026 RTTNews.com. All Rights Reserved
Copyright RTT News/dpa-AFX
© 2026 AFX News
