AI Debate

Inspiration

Debates are one of the best ways to develop critical thinking, but practicing debates alone is difficult because people often struggle to find someone to debate with or to quickly generate strong arguments.

We wanted to explore how AI could make debating more accessible and interactive. The idea was to create a platform where two AI agents debate a topic while humans can actively participate and support either side. This allows users to observe structured arguments and also practice generating their own counterpoints in real time.

What it does

AI Debate Arena is an interactive web application where two AI agents debate a given topic from opposite perspectives.

Users can join the debate by supporting either side and submitting their own arguments through text or voice input. The AI then incorporates the user's argument into its next response, creating a collaborative human-AI debate experience.

The platform also includes animated robot debaters and voice narration, making the debate feel more engaging and dynamic.

How we built it

We built the project using Next.js for the frontend and backend API routes. The AI debate responses are generated using the Gemini API, which allows us to create two agents arguing different viewpoints.

The frontend manages the debate state and displays the conversation round by round. Users can submit arguments through a text input or voice input using the browser's Speech Recognition API, while responses are read aloud using the Speech Synthesis API.

To ensure reliability during demos, the system also includes a fallback mode that generates local debate responses if the AI API becomes unavailable or quota limits are reached.

Challenges we ran into

One challenge was designing prompts that consistently generated clear arguments for both sides of the debate without repeating ideas.

Another challenge was managing the debate history so that the AI remained focused without sending too much context to the API.

We also had to handle voice input and speech synthesis carefully to avoid conflicts when the microphone was listening while the system was speaking.

Finally, ensuring the application continued working even when the AI API failed required implementing a fallback debate system.

Accomplishments that we're proud of

We’re proud that we built a fully interactive AI debate system that combines multi-agent AI, human participation, and voice interaction into a single experience.

The platform allows users not only to watch AI debate but also to influence the discussion directly, which makes the experience much more engaging than a typical chatbot.

We’re also proud of making the demo reliable by including fallback responses and designing the debate flow to work smoothly in real time.

What we learned

Through this project we learned how to design effective prompts for multi-agent AI interactions and how to manage conversation history to keep responses relevant.

We also gained experience integrating voice interfaces in web applications, which added a more natural way for users to interact with the system.

Most importantly, we learned how AI can be used not only for answering questions but also for creating interactive experiences that help people practice skills like argumentation and critical thinking.