Aug 24, 2026 · 30 min · 9 segments
Anthropic just announced they're baking invisible watermarks directly into Claude's generated text — and while everyone else was busy having opinions about it, we were busy asking the more interesting…
Katie MaloneHost
As I am recording this, it is the middle of August 2026, and what that means is that about a week ago, maybe a bit more, Anthropic had an, an announcement that actually got a lot of airtime.
It was about a change that they're doing to, uh, their generation algorithm to put a watermark into their text.
A lot of people were like, "Thank goodness." A bunch of other people got real mad [laughs] and immediately started trying to figure out how to get rid of the watermark.
But what I was excited about is this episode that we are recording right now because I was like, "An invisible watermark in text? Oh, heck yeah.
[rock music] The general idea of a watermark, just to back up in case this isn't something you've been following super closely, is it's a way that you can tell whether a given piece of text was generated by an AI or not.
We'll talk a little bit more about what that means and how true it might be over the course of this episode.
Um, but the idea is [laughs] it's not like, uh, I don't know, a giant tag at the beginning that says, "AI-generated text," because that would obviously give itself away.
Or there were some jokes about maybe if you took the first letter of every sentence, it spells out a secret message, and that's how you know that it was watermarked by an AI.
There was some speculation that maybe what they were doing was they were encoding, uh, special invisible Unicode characters, like maybe special white space characters that when you render the, the text on the screen for your eyes as a human, it doesn't look like anything, but it's to the machines that are reading it, maybe a signature of the fact that it was generated by an LLM.
So the choice of words that Claude makes when it's generating an, an answer is explicitly impacted and affected by this algorithm.
In particular, they say that, "Our text watermark is a version of the SynthID text approach published by Google DeepMind in Nature in 2024." So I looked it up, and this is what we're gonna be talking about today.
Read the full transcript.
Create an account to read the whole episode, search across every transcript, and follow the shows you care about.
As I am recording this, it is the middle of August 2026, and what that means is that about a week ago, maybe a bit more, Anthropic had an, an announcement that actually got a lot of airtime.
It was about a change that they're doing to, uh, their generation algorithm to put a watermark into their text.
A lot of people were like, "Thank goodness." A bunch of other people got real mad [laughs] and immediately started trying to figure out how to get rid of the watermark.
But what I was excited about is this episode that we are recording right now because I was like, "An invisible watermark in text? Oh, heck yeah.
[rock music] The general idea of a watermark, just to back up in case this isn't something you've been following super closely, is it's a way that you can tell whether a given piece of text was generated by an AI or not.
We'll talk a little bit more about what that means and how true it might be over the course of this episode.
Um, but the idea is [laughs] it's not like, uh, I don't know, a giant tag at the beginning that says, "AI-generated text," because that would obviously give itself away.
Or there were some jokes about maybe if you took the first letter of every sentence, it spells out a secret message, and that's how you know that it was watermarked by an AI.
There was some speculation that maybe what they were doing was they were encoding, uh, special invisible Unicode characters, like maybe special white space characters that when you render the, the text on the screen for your eyes as a human, it doesn't look like anything, but it's to the machines that are reading it, maybe a signature of the fact that it was generated by an LLM.
So the choice of words that Claude makes when it's generating an, an answer is explicitly impacted and affected by this algorithm.
In particular, they say that, "Our text watermark is a version of the SynthID text approach published by Google DeepMind in Nature in 2024." So I looked it up, and this is what we're gonna be talking about today.
