DebateZone is built on fair, honest debate. These rules keep competition meaningful for everyone. Violations are detected automatically.
You will be randomly assigned FOR or AGAINST the topic. Debate that position — even if you personally disagree. This is the core of the game.
All arguments must relate to the debate topic. Off-topic messages will score you lower with the AI judge and may be flagged as spam.
Rebut what your opponent actually said. Ignoring their points and repeating your own argument is penalised by the judge.
Make your points clearly and let the debate flow naturally. Sending walls of text or flooding the chat to block your opponent counts against you.
Using ChatGPT, Claude, Gemini, or any other AI to write your messages is strictly forbidden. Our system detects AI-generated text automatically. Only flag when certainty is very high — good human arguments are never penalised.
Sending the same message repeatedly, copy-pasting external text, or spamming to run down the clock will result in a rate-limit block and a lower judge score.
Attack arguments, not people. Slurs, personal attacks, or content targeting someone's identity will result in an immediate account suspension.
Disconnecting to avoid a loss counts as a forfeit. You lose the match automatically and your opponent gets the win.
Every message is scanned by a lightweight heuristic that looks for structural AI patterns — bullet lists, numbered lists, stacked transition words ("Furthermore… Moreover… Additionally…"), signature phrases, and markdown formatting. These are things humans almost never use in live chat.
If enough suspicious messages accumulate, a second AI model analyses your full conversation and decides with at least 90% confidence before any penalty is applied. Good human arguments — even very sophisticated ones — will never trigger a ban. The system is tuned to avoid false positives.
Cheating undermines the purpose of the platform. If you need to use AI to win, you haven't learned anything.
| Offense | Consequence | Severity |
|---|---|---|
| AI usage detected (1st time) | 1-hour queue suspension, no Elo or stats awarded | medium |
| AI usage detected (repeat) | 12-hour queue suspension, no Elo or stats awarded | high |
| Spam / rate limit | Message blocked, lower judge score | low |
| Forfeit / intentional disconnect | Automatic loss, Elo penalty (ranked) | medium |
| Hate speech / harassment | Permanent account suspension | high |
After each debate, an AI judge evaluates both participants on six criteria (0–10 each): Relevance, Clarity, Logic, Rebuttal, Persuasiveness, and Consistency. The highest total score wins. Ranked matches also update your Elo rating.
The judge penalises spam, repetition, ignoring the opponent, and off-topic arguments. It rewards direct engagement, logical structure, and strong rebuttals.
Ready to debate?