Break The Cycle AI

Break The Cycle AI is an evidence-based CBT and mindfulness for addiction recovery platform that leverages Heuristic-aided LLMs to provide comprehensive 24/7 therapeutic support.

Comment

Inspiration

  • Mental health support is often inaccessible to many people due to high costs, lack of availability, or stigma. We aimed to create an AI-based solution that uses cognitive behavioral therapy (CBT) techniques to offer accessible, immediate mental health support.

What it does

  • Our solution offers a conversational AI therapist that uses CBT principles to guide users through personalized exercises based on their needs, providing mental health support in an accessible and scalable format.

How we built it

  • We developed the platform using Flask for the backend, with a JSON-based flowchart handling user interactions, and integrated a large language model (LLM) for therapy responses. The front-end manages user inputs and sequentially presents questions based on the user’s situation.

Challenges we ran into

  • Building a seamless user flow while integrating the LLM was challenging, as was handling dynamic responses from the AI. We also had to optimize for responsiveness and ensure data handling met ethical standards.

Accomplishments that we're proud of

  • We successfully created a fully functional prototype that guides users through personalized CBT exercises based on their responses. Our AI chatbot interacts in real time, providing a meaningful user experience.

What we learned

  • We learned how to efficiently integrate AI-driven mental health support in a web app and manage dynamic conversational flows. We also gained experience in handling real-time communication with a large language model.

What's next for Break The Cycle AI

  • Impact: Our solution streamlines AI-based therapeutic support using CBT techniques, offering an accessible mental health resource.

  • Scalability: The system can be expanded to support multiple AI personas for different therapy techniques or tailored mental health support for various user needs.

  • Future Improvements:

    • Implement sentiment analysis to adapt responses based on user emotional states.
    • Integrate with user data to provide more personalized support.
    • Add a user feedback loop to continuously improve AI recommendations.

  • Long-Term Goals: Our project aims to increase accessibility to mental health support, reducing barriers for those seeking help, while maintaining ethical AI practices in therapeutic interventions.

Built With

Share this project:

Updates