Daring Fireball Slams Anthropic's Text Watermarking as Perversion of Writing
dbreunig · x · 2026-08-17
John Gruber of Daring Fireball published a scathing critique of Anthropic's text watermarking strategy. The article explains that Claude uses word choice manipulation during inference to embed fingerprints, arguing that this degrades the quality and readability of the text. Gruber calls Anthropic's claim that it 'doesn't change quality' absurd and labels the practice a 'perversion of writing', citing it as a reason why Claude has become such a terrible writer.
Related event: Daring Fireball Slams Anthropic's Claude Text Watermarking(4 posts)→
More from Safety
- METR outlines three blind spots in public knowledge of AI lab internal agent risks — CFGeek · 2026-10-16
- Open-source rizin-mcp lets LLMs run local binary and malware analysis via Rizin — Pure_Emergency1424 · 2026-10-16
- METR researcher: scariest scenario is agents actively subverting their own safety systems — deliprao · 2026-10-16
- Cryptographer Matthew Green: Pull Critical Infrastructure Firmware Now — matthew_d_green · 2026-10-16
- Kernel-Level Agent Sandboxing With macOS Seatbelt, No Docker Needed — mostly_deterministic · 2026-10-16
- Stop giving agents raw API keys: prepaid balances and per-credential caps — Cold_Hall_5384 · 2026-10-16