Why text watermarking is hard: expert explains discrete data challenge and AI Act implications
antoine_chaffin · x · 2026-08-11
AI researcher @antoinechaffin discusses the difficulty of watermarking LLM outputs: text is discrete, so you can't simply add noise like with continuous data (images, videos, audio). This is similar to the problem with text GANs. He notes that while you can't freely navigate a continuous space, there are still methods, but exposing encryption keys or detection tools carries risks. He also recalls discussions around the AI Act and recommends Meta watermarking expert @pierrefdz.
More from Safety
- [un]prompted 2026 Announces First Speakers: AI x Cybersecurity — dyn___ · 2026-08-11
- LLM Watermarking Can Be Repurposed for Imperceptible Text Steganography — dyn___ · 2026-08-11
- AI Labs Pivot to Offensive Use Cases to Mask Poor Reliability, Says Researcher — mer__edith · 2026-08-11
- Mapping the AI Agent Governance and Security Landscape — serendip-ml · 2026-08-11
- CMU Introduces WeClawArena: Benchmark for Cross-User Agent Collaboration and Security — CarnegieMellonU · 2026-08-11
- AI Safety: Can 'Lab Spoofing' Bypass Model Alignment? — IasonGabriel · 2026-08-11