As is being reported pretty much everywhere, Anthropic is adding watermarks to AI-generated content. Will other AI models follow suit?
As discussed in this article on Anthropic’s website, Anthropic has signed the EU AI Act’s Article 50(2) Code of Practice on Transparency of AI-Generated Content, as a provider of both generative AI models and generative AI systems. So, they are putting those commitments into practice. As they note in the article, here’s what their marking commitments mean for Claude:
- New models will mark AI-generated content from day one. Claude models launched in the EU on or after August 2, 2026 will support machine-readable marking at launch. Generated text will carry embedded watermarks, and generated files will include digitally signed provenance metadata where supported.
- Marking works everywhere you use Claude. Marks will apply to output from supported Claude models across Claude Platform (API), Claude, Claude Code, Claude Cowork, and Claude Tag, and wherever Claude is offered, worldwide. Some platforms or features may not support certain marking types.
- We’ll help you detect Claude’s marks. We’ll support users and other third parties to detect Claude’s marks, as the Code requires, and we’ll share details in forthcoming documentation.
- Existing models are in progress. The law includes a transition period for Anthropic models launched before August 2, 2026, and we’re working to add marking support for those models as well.
How will Claude mark content? Through “two complementary techniques”, as follows:
1. Embedded watermarks in text
When a supported Claude model generates text, it weaves an imperceptible watermark directly into the text itself. You won’t see it, and it doesn’t change the meaning, quality, or readability of Claude’s response.
Because the watermark is part of the text, it will travel with the text when it’s copied and pasted elsewhere, and may persist through some editing. Watermarking will be applied at the model level, which means it will be present no matter which Claude product or surface the text comes from.
2. Signed provenance metadata
When Claude generates a supported file type, such as a .svg, .png, or .jpg, it will attach signed provenance metadata. This metadata follows the Coalition for Content Provenance and Authenticity (C2PA) open standard, which is used across the industry to record information about content provenance. If a signed metadata label is present, it signals that a file was processed by Claude and lets you detect whether the file has been tampered with.
There are some limitations. Content generated by Claude may not carry a detectable mark if, for example:
- It was generated by a model released before marking was supported;
- The text has been heavily edited, paraphrased, translated, or mixed into other writing;
- The passage is very short, leaving too little text for a reliable signal;
- A file’s metadata was stripped through format conversion, re-saving, screenshots, or other means;
- It was produced through a platform, feature, or file type where a particular marking type wasn’t supported.
That last one may be key as it relates to eDiscovery and other solutions where Claude is being used. It sounds to me as though the watermarking won’t extend to those platforms – at least yet.
Anthropic also states: “We’re also working to enable users and other third parties to detect Claude’s embedded watermarks and provenance metadata. Detection checks whether a piece of text or a file carries a supported Claude mark. If a supported mark is found, it indicates that the content may have been processed by Claude… We’ll share details on detection mechanisms in forthcoming technical documentation.”
It’s important to note that users can’t opt out of watermarking. The New York Post noted (rather cynically, I might add) in the title of their coverage piece on it: “Sorry, students: Anthropic adding watermarks to AI-generated content, potentially making cheating harder”.
🤣
Not surprisingly, many users aren’t happy that Anthropic is adding watermarks to AI-generated content. As noted in the NY Post article, one person wrote on Reddit that “This is such a massive blunder on Anthropic’s part… That mark will be the kiss of death on any piece of text that people can sell. People won’t want to pay for it. It will be a scarlet letter.” Another commenter bluntly complained, “Claude watermarking our work is unethical and disgusting.”
Unethical? 🤯 Really?
Given that Anthropic has made it clear their commitment to adhere to the EU AI Act’s Article 50(2) code, the obvious question is: what will the other AI models do? Will they follow suit? Bury their heads in the sand and wait for potential action from the EU? Push back against the EU AI Act’s Article 50(2) code?
Imagine if all the models follow suit – and make the detection mechanisms available as well. How many people are putting content out there claiming it to be their original works, when (in fact) AI has written much or all of it? Too many to count.
Of course, if the other models don’t follow suit, Anthropic’s mark could be the kiss of death…for Anthropic. Users will flee in droves for other models where they can continue to flaunt transparency – for as long as they can.
I can’t decide if this is a groundbreaking moment in AI development or much ado about nothing. But it’s intriguing.
So, what do you think? Are you surprised that Anthropic is adding watermarks to AI-generated content? Please share any comments you might have or if you’d like to know more about a particular topic.
Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by my employer, my partners or my clients. eDiscovery Today is made available solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscovery Today should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.