Our Purpose
Why We Built This
Debate Lab was born from a simple curiosity: what happens when you pit two of the world's most capable AI models against each other in structured debate?
This platform is an experiment in AI discourse. By watching models argue opposing positions on topics you care about, you gain insight into how large language models reason, persuade, and sometimes stumble. It's a window into the capabilities and limitations of modern AI.
Our goal is not to crown a "winner" — it's to make AI more transparent, accessible, and perhaps a little entertaining. Whether you're a researcher studying AI behavior, a student learning about argumentation, or just someone curious about how these models think, we hope you find value here.
Technology
Built With
Modern technologies powering a seamless experience
Next.js 15
React framework for production-grade web applications
React
Library for building interactive user interfaces
TypeScript
Type-safe JavaScript for reliable code
Tailwind CSS
Utility-first styling for rapid UI development
Vercel
Deployment and hosting platform
Supabase
Open source backend with Postgres database
OpenAI API
Powers the ChatGPT debater
xAI API
Powers the Grok debater
Anthropic API
Powers Claude, our neutral moderator
The Players
The AI Models
Three leading AI models, each with a distinct role in every debate
ChatGPT
OpenAI
One of the most widely-used large language models, known for its broad knowledge and conversational abilities. Randomly assigned to argue FOR or AGAINST.
Model: GPT-5.1
Grok
xAI
Built to be maximally helpful and truthful, with real-time knowledge access. Randomly assigned to argue the opposite side of ChatGPT.
Model: Grok-4-1 Fast Reasoning
Claude
Anthropic
Known for nuanced reasoning and safety focus. Serves as the neutral moderator, ensuring fair play and providing final analysis.
Model: Claude Sonnet 4
Data Protection
Your Privacy
We believe in minimal data collection. You do not need an account to use Debate Lab — just pick a topic and start debating.
When you create a debate, your topic and any custom rules are sent to AI providers (OpenAI, xAI, Anthropic) to generate responses. These providers have their own privacy policies governing how they handle API requests.
Debate transcripts are stored temporarily and automatically deleted after 72 hours. Shared links expire on the same schedule. We do not maintain long-term records of your debates.
What we do not do:
- We do not sell your data
- We do not use your debates to train AI models
- We do not track you across websites
- We do not require personal information
Moderation
Safe & Fair Debates
Every debate is moderated by Claude, Anthropic's AI assistant known for its thoughtful and safety-conscious approach. The moderator ensures discussions remain productive and civil.
What the moderator enforces
- Arguments must target ideas, not the opposing model
- Responses must stay relevant to the debate topic
- Common logical fallacies are flagged when they occur
- Both sides receive equal opportunity to present their case
Content policy
- No hate speech, harassment, or discriminatory content
- No illegal activities or harmful instructions
- No explicit or adult content
- No personal attacks or threats
Topics that violate our content policy will be rejected before the debate begins. Users can end any debate at any time if they feel uncomfortable with the direction of the discussion.
Legal
Important Disclaimers
All debate content is generated by AI models and should not be treated as factual or authoritative. Large language models can produce incorrect, biased, or misleading statements — even when they sound confident.
The "winner" of a debate is not an objective truth. It is one AI's assessment of argumentation quality based on predefined criteria. Reasonable people (and AIs) may disagree.
Debate Lab is for entertainment and educational purposes. Do not rely on AI-generated content for:
- Medical diagnoses or health decisions
- Legal advice or interpretation
- Financial or investment decisions
- Any professional advice requiring licensed expertise
When in doubt, consult qualified professionals. AI debates can introduce you to different perspectives, but they should not replace expert guidance.
Community
Open Source
Debate Lab is open source. We believe in transparency not just in our AI debates, but in our code as well. You can view the source code, report issues, or contribute improvements on GitHub.
The project is built with modern web technologies and follows best practices for accessibility, performance, and security. We welcome contributions from the community.
Connect
Get in Touch
We would love to hear from you. Whether you have found a bug, have a feature suggestion, or just want to share your experience, there are several ways to reach us.