A new website called DetectArena is crowdsourcing the ranking of AI detectors by letting real users vote on which tool is more accurate. The platform uses an Elo rating system—similar to chess rankings—to create a live leaderboard of AI detection tools like GPTZero and Originality.ai.
Users simply submit a piece of text, and the site pairs two detectors head-to-head. The community then votes on which one correctly identifies whether the text is AI-generated or human-written. Over time, the Elo scores update to reflect each detector's win-loss record against opponents.
The project aims to provide a transparent, community-driven benchmark for AI detection accuracy. "Whether you're a student, writer, or content creator, knowing which AI detector is most accurate matters," the creator explains. DetectArena is live now and already attracting users who want to test how well different detectors handle AI-generated content.
As AI writing tools become more common, reliable detection is increasingly important. DetectArena offers a dynamic, real-world test instead of relying on static academic benchmarks.