Trivial to Remove: Expert Says C2PA AI Watermarks Offer No Real Protection
CodeByPoonam · x · 2026-08-13
A recent tool claims to remove AI-generated image watermarks from models like OpenAI and Gemini. However, developer tests reveal that removing C2PA metadata is trivial and can be easily done with existing command-line tools, offering no real security.
He emphasizes that the real technical challenge lies in steganographic watermarks embedded directly into image pixels, such as SynthID. Only robust underlying watermarking technologies capable of surviving image processing and malicious removal can effectively counter deepfakes.
More from Safety
- Linux Kernel CVEs Surge From ~500 to 1500+ Per Release, LLMs Blamed for Bulk of the Rise — burny_tech · 2026-10-03
- COLM 2026 Launches DAIH Workshop on Deploying LLMs/VLMs Responsibly in Healthcare — StellaLisy · 2026-10-03
- Trillium Labs wants to do open research on recursive self-improvement and agents — nordicinst · 2026-10-03
- Trillium Labs Wants to Research Self-Improvement and Model Behavior in the Open — Wired AI · 2026-10-03
- Cloudflare Turnstile everywhere: anti-AI scraping walls now hit human users — sethlazar · 2026-10-02
- Filler tokens let frontier models reason invisibly: 13-point gains undetectable by CoT monitoring — PandaAshwinee · 2026-10-02