Policy · 2h ago
Meta’s Content Seal lags behind Google’s SynthID watermarking
Meta introduced Content Seal, an invisible watermark for AI images, in July after pressure from its Oversight Board. The system only works with Meta’s own AI model, unlike Google’s SynthID which is open and model-agnostic. Critics say Meta’s closed approach limits effectiveness against deceptive content across platforms.
Meridian48 take
Meta’s proprietary watermarking is a half-measure that ignores the cross-platform nature of AI-generated misinformation.
Read the full reporting
Meta made its own AI detection system. It should have just used Google’s →
The Verge AI
ai-watermarkingcontent-authenticity