🔍 Unlocking the AI Detection Challenge!
I’ve created a crowd-sourced AI detection benchmark designed to differentiate between human-generated and AI-generated content. Dive into this engaging challenge where your insights matter!
Key Highlights:
- Dataset:
- 16K human posts sourced from Reddit, Hacker News, and Yelp.
- Paired with AI responses from 6 models across Anthropic and OpenAI.
- Methodology:
- Responses matched for prompt and length.
- No coaching; just natural model performance in context.
- Early Insights:
- Reddit: Human posts are more casual, easier to detect.
- Hacker News: Detection proves significantly challenging.
Your feedback is invaluable! Participate in our project by testing pairs and identifying the subtleties. I’ll be releasing the full dataset on HuggingFace and plan to publish a paper based on our findings.
👉 Join the conversation! Share your thoughts and help us calibrate AI detectability.
