WDAIT

Inspiration

AI gives us answers every day, but we rarely understand what influences them. We built Why Did AI Do That? (WDAIT) to make AI interactions more transparent and understandable.

What it does

WDAIT is a Chrome extension that explains ChatGPT responses by highlighting possible influences, assumptions, and information that could change the answer—all in one sidebar.

How we built it

We used JavaScript, Chrome Extension Manifest V3, Python, FastAPI, and the Gemini API to capture conversations, analyze responses, and display structured explanations through manual and automatic analysis.

Challenges we ran into

Capturing conversations reliably, detecting completed responses, connecting the extension to the backend, and generating useful explanations without claiming access to hidden AI reasoning were our main challenges.

Accomplishments that we're proud of

We created an AI response debugger designed to make AI interactions more transparent, accessible, and actionable without requiring users to copy and paste conversations into another tool.

What we learned

We learned to build browser extensions, integrate APIs, connect frontend and backend systems, and structure AI-generated responses. We also learned the importance of explaining observable influences rather than claiming to reveal an AI's private reasoning.

What's next for WDAIT?

We aim to improve analysis accuracy, add full-conversation context, strengthen automatic detection, and eventually support more AI platforms while maintaining simplicity and privacy.

Built With

• chrome-extension-api-(manifest-v3)
• css
• fastapi
• gemini-api
• github
• html
• javascript
• pydantic
• python