InsureWise AI

Inspiration

We were inspired by how confusing health insurance can be for everyday people. Many individuals choose plans based mainly on price, advertising, or advice from friends, without fully understanding what they are covered for. Terms like deductibles, co-pays, exclusions, waiting periods, and provider networks can make insurance feel overwhelming. We wanted to build a tool that helps people feel more informed and confident before making such an important financial and healthcare decision.

What it does

InsureWise AI is an AI-powered Health Coverage Navigator. It uses conversational AI to ask users simple questions about their lifestyle, health habits, family medical history, financial priorities, current healthcare needs, and future plans. Based on their responses, the system creates a personalized health and lifestyle profile.

The AI then identifies current risk factors, possible future healthcare needs, and coverage priorities. Instead of recommending a specific insurance company, InsureWise AI explains what types of coverage features may matter most to the user, such as maternity benefits, outpatient care, chronic disease management, preventive healthcare, specialist access, or broader hospital networks.

It also generates personalized questions users can ask insurers, helping them compare plans more clearly and avoid unexpected costs.

How we built it

We built InsureWise AI using a combination of conversational AI, secure API integration, and a simple user-friendly interface. As a team without formal computer science backgrounds, we focused on rapid learning, research, and iteration throughout the development process.

We began by designing the user experience and chatbot flow, making sure the conversation felt intuitive and easy to follow for people with little or no insurance knowledge. Once the frontend was in place, we integrated the Gemini API to power the AI-driven analysis and personalized guidance.

One of the key technical challenges was managing API security and deployment. We explored several approaches before rebuilding the application using Streamlit, which allowed us to securely store API credentials through Streamlit Secrets rather than exposing them in the codebase. We also adapted our implementation during development to work around API quota limitations and model availability constraints.

Throughout the project, we used AI-assisted development tools to help us learn unfamiliar concepts, troubleshoot issues, and explore implementation options. While these tools helped accelerate development, all product decisions, user flows, and feature designs were created and evaluated by our team.

The final result is a working prototype that combines personalized user input, AI-generated insights, and educational guidance to help users better understand their health insurance needs.## Challenges we ran into

One challenge was finding the right balance between being helpful and being responsible. Health insurance decisions can have major financial and personal consequences, so we did not want the AI to make decisions for users or recommend a specific insurer.

Another challenge was simplifying complex insurance language without losing important meaning. We had to make sure the system could explain terms like deductibles, exclusions, waiting periods, and provider networks in a way that users can actually understand.

We also had to think carefully about privacy, because users may share sensitive information about their health, family, and finances.

Accomplishments that we're proud of

We transformed a traditionally complex and jargon-heavy insurance evaluation process into an accessible AI-guided experience that helps users identify coverage priorities before speaking with insurers. The project does not just provide information; it helps users understand their own needs, prepare better questions, and compare plans more confidently.

We are also proud of the explainable output. Users are not just told what to consider — they are shown why certain coverage features may matter based on their personal circumstances and future plans.

Most importantly, we built the project in a responsible way. InsureWise AI supports decision-making without replacing professional advice, official policy documents, or insurer confirmation.

What we learned

We learned that AI can be most valuable when it helps people understand complex systems, not when it makes decisions for them. For health insurance, the real problem is not only lack of information, but lack of clear, personalized guidance.

We also learned that responsible AI design is essential in high-impact areas. A tool like InsureWise AI must be transparent, privacy-conscious, and careful about the limits of its recommendations.

What's next for InsureWise AI

Next, we would like to improve InsureWise AI by expanding the insurance knowledge base, testing it with more realistic user scenarios, and refining the chatbot experience.

Future versions could include plan comparison templates, multilingual support, and more detailed education about insurance terminology. We would also like to add stronger safeguards, regular human review of the knowledge base, and clearer pathways for users to verify information directly with insurers.

By helping users better understand their coverage needs before purchasing insurance, InsureWise AI has the potential to reduce confusion, improve financial decision-making, and encourage more informed healthcare planning.

Our goal is for InsureWise AI to become a trusted educational guide that helps people make more confident and informed health insurance decisions.