LLMMedium
AI-Generated Text Detectors Reportedly Misclassified Human and AI-Written Text
- Occurred
- Jan 31, 2023
- Published
- Jan 31, 2023
- Source
- AI Incident Database
Summary
AI-generated text detectors, including OpenAI’s AI Text Classifier, GPTZero, and Originality.ai, were publicly tested in early 2023 and reportedly produced false positives, false negatives, and easy evasion. OpenAI said its classifier identified only 26% of AI-written samples as likely AI-written and falsely labeled 9% of human-written text.
Developers
- OpenAI
- Edward Tian
Deployers
- Originality.ai
- OpenAI
- Edward Tian
Harmed parties
- Teachers
- Students
- Educators
- Educational communities
Classification
Category confidence: 90%
Severity confidence: 85%
Rationale: AI text detection tools show high error rates, causing false accusations of AI usage.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8