Anthropic's Watermark Survives Copy-Paste, Fails Real Dev Workflows
CackleRooster · reddit · 2026-08-14
The author points out that while Anthropic's AI text watermark survives simple copy-pasting, it faces significant challenges in real-world software development workflows. Since Anthropic hasn't published tests on how well the watermark survives code refactoring or PR reviews, teams currently cannot reliably track if a Claude-generated patch remains detectable after standard development processes.
More from Models
- Using System One models in Swift: fast, deterministic decisions via Apple Foundation Models — rxwei · 2026-10-03
- Linux Kernel CVEs Surge From ~500 to 1500+ Per Release, LLMs Blamed for Bulk of the Rise — burny_tech · 2026-10-03
- Developer Complains OpenAI's Coding Model Endlessly Scopes Creeps Instead of Finishing Tasks — DavidWells · 2026-10-03
- Sonnet 5 Spotted in Google Antigravity Backend, Which Still Runs Sonnet 4.6 — brandon_galang · 2026-10-03
- Post-training Yandex AliceAI-80B-A3B from scratch: a NaN bug in custom V100 kernels killed one run — jjusko20 · 2026-10-03
- Every's Dev Day chat with Matthew Berman: budget gone the moment he tried Ultrafast — every · 2026-10-03