Development, system design, safety, and performance metrics of a conversational agent for reducing depressive and anxious symptoms based on a large language model: The MHAI study

Key Points

• The research aims to develop and evaluate a conversational agent that can reduce depressive and anxious symptoms using a large language model.
• Designing a conversational agent with multiple functionalities.
• Comparing performance metrics of GPT-4o and Llama 3.1-8B.
• Evaluating safety and efficacy of the agent for mental health outcomes.
• GPT-4o shows superior performance compared to Llama 3.1-8B.
• The implementation of this conversational agent may enhance traditional therapeutic approaches.