BiasMedium
Social Media's Automated Word-Flagging without Context Shifted Content Creators' Language Use
- Occurred
- Mar 15, 2017
- Published
- Apr 8, 2022
- Source
- AI Incident Database
Summary
TikTok's, YouTube's, Instagram's, and Twitch's use of algorithms to flag certain words devoid of context changed content creators' use of everyday language or discussion about certain topics in fear of their content getting flagged or auto-demonetized by mistake.
Developers
- YouTube
- Twitch
- TikTok
Deployers
- YouTube
- Twitch
- TikTok
Harmed parties
- YouTube content creators
- Twitch content creators
- TikTok content creators
- Instagram content creators
Classification
Category confidence: 85%
Severity confidence: 80%
Rationale: Automated flagging without context causes disproportionate impact on creators, reflecting algorithmic bias in content moderation.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8