User experience and safety of generative AI-based mental health chatbots: Scoping review protocol

Area of Science:

• Digital Health
• Artificial Intelligence in Mental Health
• Human-Computer Interaction

Background:

• Mental health issues present a global challenge with a significant treatment gap.
• Digital Mental Health Interventions (DMHIs), particularly chatbots, offer scalable solutions.
• Generative Artificial Intelligence (GenAI) enhances chatbot capabilities but raises safety concerns.

Purpose of the Study:

• To conduct a scoping review of GenAI-based mental health chatbots.
• To identify and describe existing GenAI mental health chatbots.
• To focus on user experience, safety, and risk mitigation strategies.

Main Methods:

• Adherence to Joanna Briggs Institute (JBI) and PRISMA-ScR guidelines.
• Systematic database searches across multiple academic and research engines.
• Screening and data extraction by two independent reviewers, employing descriptive and thematic analysis.

Main Results:

• A comprehensive overview of current GenAI mental health chatbots.
• Identification of innovative practices in user experience and safety.
• Highlighting of knowledge gaps in the field.

Conclusions:

• Findings will guide the ethical development and implementation of GenAI mental health interventions.
• Informing best practices for user safety and experience in AI-driven mental health tools.
• Addressing the need for further research into the safe application of GenAI in mental healthcare.