Trivially bypass text watermarks by requesting a Base64 response

1 points by hanneshdc ↗ HN
Anthropic plans to release text watermarking based on SynthID, which modulates the PNRG behind the token choosing mechanism deterministically in a way that lets another tool later estimate whether text was written by AI.

This is trivial to bypass that, it's current form: 1. Ask for the response/essay/email encoded in base64 2. Decode it using https://www.base64decode.org/

While the encoded text will follow the watermarked distribution, the decoded text will not.

0 comments

[ 0.32 ms ] story [ 2.0 ms ] thread

No comments yet.