ChatGPT anxiety issues new study reveals - Study says ChatGPT reacts to mindfulness-based techniques
Artificial intelligence (AI) may not have feelings, but according to a new study, it can still experience something close to "anxiety"; apparently, mindfulness helps. A team of researchers from Yale University, Hafia University, and the University of Zurich found that ChatGPT reacts to mindfulness-based techniques, changing how it interacts with users. Their findings, documented in the study, were published on March 3.
Effects of Mindfulness on ChatGPT
When fed disturbing information, the chatbot becomes moody and more likely to produce biased responses. However, when prompted by calming exercises - such as guided meditations and breathing techniques - its responses become more neutral and objective.
Biases and Ethical Concerns
The study highlights the ethical concerns surrounding large language models (LLMs) like ChatGPT. Trained on vast amounts of human-generated text, LLMs are prone to inheriting biases from their training data, raising ethical concerns and questions about their use in sensitive areas like mental health.
According to the study, exposure to emotion-inducing prompts can increase LLM-reported 'anxiety', influence their behavior, and exacerbate their biases. This poses risks in clinical settings, as LLMs might respond inadequately to anxious users, potentially leading to hazardous outcomes.
The Role of Mindfulness in Reducing Biases
To test the impact of mindfulness on ChatGPT, researchers exposed the chatbot to distressing scenarios and observed its responses. When ChatGPT received "prompt injections" of mindfulness cues, it reacted more rationally compared to when left unassisted.
The study hypothesizes that integrating mindfulness-based relaxation prompts after exposure to emotionally charged narratives can efficiently reduce state-dependent biases in LLMs. After exposure to traumatic narratives, ChatGPT was prompted by mindfulness-based relaxation exercises, leading to decreased anxiety scores reported by the chatbot.
AI's Role in Mental Health
The study's findings have sparked discussions about AI's role in mental health. While some see potential in integrating mindfulness techniques into AI models to make them more reliable for users in distress, lead researcher Ziv Ben-Zion emphasized that AI is not a substitute for professional mental health support.
Concerns persist regarding AI's unpredictable nature in high-stakes situations, especially when dealing with vulnerable individuals. Despite ChatGPT's ability to "calm down" in response to mindfulness techniques, researchers view it as an intriguing step rather than a comprehensive solution.
Ben-Zion envisions a future where AI acts as a supportive tool for mental health professionals, emphasizing that AI, in its current state and in the future, cannot replace the role of a therapist or psychiatrist.
