cd /news/artificial-intelligence/trivially-bypass-text-watermarks-by-… · home topics artificial-intelligence article
[ARTICLE · art-100021] src=news.ycombinator.com ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

Trivially bypass text watermarks by requesting a Base64 response

Anthropic plans to release text watermarking based on Google DeepMind's SynthID, which deterministically modulates the pseudorandom number generator behind token selection to allow later detection of AI-written text. However, the watermark is trivially bypassed by requesting the response encoded in Base64 and then decoding it, as the decoded text no longer follows the watermarked distribution. The bypass was highlighted in a Hacker News comment, though no official statement from Anthropic has been made.

read1 min views12 publishedAug 17, 2026

| |||||||||||| 1 point by | 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. | ||||||||||| |

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @anthropic 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/trivially-bypass-tex…] indexed:0 read:1min 2026-08-17 ·