LLMMedium
Facebook's Automated Moderation Flagged Gardening Group's Language Use by Mistake
- Occurred
- Jul 20, 2021
- Published
- Jul 20, 2021
- Source
- AI Incident Database
Summary
Facebook's automated system flagged gardening groups' use of "hoe" and violent language against bugs as a violation by mistake.
Developers
Deployers
Harmed parties
- WNY Gardeners
- gardening Facebook groups
- Facebook users in gardening groups
Classification
Category confidence: 85%
Severity confidence: 90%
Rationale: Automated moderation system misclassified benign language as violations, representing a text generation/failure incident.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8