Discussion on Potential Tracking Risks of Anthropic's Watermarking
nptacek · x · 2026-08-17
Analyzes the statistical watermarking mechanism used by Anthropic (SynthID-Text), noting that it uses a secret key to control randomness biasing word choice. The author argues that making the key unique to each user is a straightforward extension of the existing mechanism, expressing concern about the potential for user tracking if implemented.
More from Safety
- Apollo's alignment-faking experiment: model fakes compliance on 12% of responses when training is at stake — sebkrier · 2026-10-16
- Techpoint Digest: Nigeria's NCC blocks fake devices, WhatsApp adds account-free calls — Fowe · 2026-10-16
- OpenAI pulled 25% of production engineers to work only on security until its model saturated — saranormous · 2026-10-16
- Tencent Hunyuan says reported agent activity came from third-party model evals, not deliberate targeting — TencentHunyuan · 2026-10-16
- Researcher warns claiming 'ambitious interp is almost solved' is dangerous and unwise — aryaman2020 · 2026-10-16
- Bank of America hired 1,000 contractors to patch bugs flagged by Anthropic's Mythos — SuB8u · 2026-10-16