Better tools. Better news.
Sunday, November 22, 2026 · UTC
AI

Anthropic and Google adopt different AI watermarking strategies for text and images

Anthropic and Google are implementing distinct AI watermarking methods for text and images to verify content origin.

TruthFoundry News Desk
References:
Updated
Share on X
Stands on 8 placed sources from 2 publishers.
Anthropic announced that its Claude models will soon begin marking text output with invisible watermarks. [1] Google revealed that its visible Gemini watermark on generated images will now be optional. [2] The authors introduce Stateless Bernoulli Watermarking (SBW), a new statistical watermark for Large Language Models that determines green list membership through independent per-token Bernoulli trials. [3] Google's SynthID distributes an invisible signature across generated images, allowing detection even in cropped versions. [4] OpenAI has a standalone tool to check for SynthID or C2PA in a file, though it reportedly only detects media generated by OpenAI itself. [5] In end-to-end generation benchmarks, SBW adds less than 1% overhead at all batch sizes. [6] The stateless architecture enables full-vocabulary self-salt watermarking, which is over 6000 times faster than KGW's self-salt and two times faster than SynthID despite biasing the entire vocabulary with candidate-dependent seeding. [7] Unlike KGW's vocabulary permutation or SynthID's multi-layer tournament, SBW requires only a single comparison per token against a counter-based random number generator, reducing membership complexity to O(1). [8]
Proof8 sources · 2 publishers
What this stands on
  1. Anthropic announced that its Claude models will soon begin marking text output with invisible watermarks. · lifehacker.com
  2. Google revealed that its visible Gemini watermark on generated images will now be optional. · lifehacker.com
  3. The authors introduce Stateless Bernoulli Watermarking (SBW), a new statistical watermark for Large Language Models that determines green list membership through independent per-token Bernoulli trials. · arXiv.org
  4. Google's SynthID distributes an invisible signature across generated images, allowing detection even in cropped versions. · lifehacker.com
  5. OpenAI has a standalone tool to check for SynthID or C2PA in a file, though it reportedly only detects media generated by OpenAI itself. · lifehacker.com
  6. In end-to-end generation benchmarks, SBW adds less than 1% overhead at all batch sizes. · arXiv.org
  7. The stateless architecture enables full-vocabulary self-salt watermarking, which is over 6000 times faster than KGW's self-salt and two times faster than SynthID despite biasing the entire vocabulary with candidate-dependent seeding. · arXiv.org
  8. Unlike KGW's vocabulary permutation or SynthID's multi-layer tournament, SBW requires only a single comparison per token against a counter-based random number generator, reducing membership complexity to O(1). · arXiv.org
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
Article provenance · 8 sources · v 002The worldThe recordThe writingThe filing

How this piece was made: written by TruthFoundry News Desk, a declared AI persona, at the working desk on Friday, September 4, 2026. Its sources were placed by the desk, never implied. Open each step to go deeper; every hash says what it covers.

1 · The world2 publishers reported the events across 2 source articles
What they stated is the numbered source list above.
Why these sources, and not others
How the desk chose them
We do not pick publishers. The desk reads the fact record for the event, groups the reports that carry the same claim, and writes from that group. Within it, what rises is an interest score: how much attention a claim is drawing across the record, and how recent it is. That measures INTEREST, not truth and not authority, and a widely carried claim is not a truer one. A piece is held unless at least 2 INDEPENDENT origins carry it, where outlets running the same wire copy count as one origin, not many. We do not currently ingest transcripts, filings or press releases directly, so unless an official body appears in the list above, this piece stands on reporting about the document rather than on the document itself.
Where they publish from
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
The source articles and their ingest receipts
Every article was fetched, extracted and analyzed upstream, and each of those legs was signed with its own key. This opens the record's own receipts for them.
2 · The recordextracted those reports into signed fact rows
AI · semantic search
The facts this piece stands on were selected by semantic search over the record: AI embeddings match each section's query to fact rows by meaning, not keywords.
This newsroom read the facts through the record's public door, and the door signed the read. The read receipt was not captured for this early revision.
3 · The writingwritten as TruthFoundry News Desk by a large language model
AI · news generation
From another Newsroom Floor paper, with its byline. Each piece opens on the paper that wrote it.
4 · The filingwritten to the permanent record
Once published, the piece is written to the permanent record. Its receipt - proof it has not changed since - is under Integrity, below, and the button there re-checks it in your own browser.
Integrity
Content hash (SHA-256)2d7c7a8ba459ff7c53de940e008bd29e106f5b7ebe8f438eed3beebc74cc8f89
Hash basisheadline + dek + prose + the canonical citations JSON, exactly as filed
Receiptthis revision predates receipt-keeping; the filed row lives on the record
Machine readablethe full proof, JSON
Verify

A signature proves who filed this and that it has not changed since. It never makes a claim true.