[Interview] Does copyright protect your AI-generated content in Europe? Let’s find out

20 Aug 2026, 10:39 by https://euobserver.com/author/alejandro-tauber/ · EUobserver

Copyright scholar Daniel Gervais: ‘When you put your name on an article that’s written by ChatGPT or Claude, you’re basically putting a provenance mark on it saying: I take responsibility for this. I haven’t written it, but I’m putting my name on it. That doesn’t give you copyright, but it does give you liability for the content’ (Source: Courtesy Daniel J. Gervais)

Europe today

Unlock article and share
By Alejandro Tauber,
Amsterdam
,

To remain compliant with the EU’s AI Act, AI-giant Anthropic announced they would be adding an invisible watermark to text generated by its chatbot Claude.

There was much despairing.

Especially on LinkedIn, a social platform for professionals, which recently added an option to flag content as AI slop – after research estimated over one third of the copy posted on the site was AI-generated.

Some came up with clever ways around the watermark (no, copy-pasting into plain text does not work). Others cancelled their Claude account.

Yet others demonstrated how the watermark is a feature of the LLM’s next word prediction and would require full rewrites to get rid of. A small group, most likely using AI-generated text, argued we should just accept AI-written content as a fact of life, perhaps even as a form of avant-garde.

Underneath all the content, however, there is an actually interesting question. A question that, when definitively answered, could destroy billions of euros of value created in the past few years overnight.

It’s not the stock bubble.

Intellectual property (IP) and copyright laws were developed to protect what people make. So what happens when a machine ‘makes’ something? Or when it is involved in making something? Who owns what a machine makes? How much human input is needed to make machine-generated content ‘human’?

In 2019, Vanderbilt Law copyright scholar Daniel Gervais published an influential – and prescient – paper which concludes that works that do not result from human creative choices belong in the public domain – i.e. no copyright.

More recently, he proposed a framework, based largely on existing authorship law, for answering the harder question raised by today's LLMs: when a human and a machine both contribute to a work, has the human done enough for copyright to attach?

Let's take a very simple case. I use Claude to write me a LinkedIn post promoting this interview and post it verbatim. Do I own the text?

No. Nobody does.

Then the question is: why would you want to own a LinkedIn post? And that goes to the question of why we have copyright.

Copyright applies to all this type of content, but when you write a blog post or LinkedIn post, usually it's not done to monetise that text. When I write a law review article, I don't get paid. I peer-review articles for journals and don't get paid. It's part of my job.

So copyright is not really a motivator there. But if you're a professional writer, journalist or songwriter, of course the answer matters a heck of a lot more.

What if I feed the recording of this interview, my research and the papers I've read into an LLM and ask it to generate the article in my tone of voice? Do I own that?

No.

And in fact, the recording you're making of what I'm saying, in some countries, may be sufficient to give me copyright in what I've said. You might quote it and then infringe my copyright.

If you were to publish that article under your name, then you're the infringer.

Ok, but if I then choose to hand that AI-generated article to my publisher, does the publisher own the copyright?

No. There's nothing to transfer. There's no right to transfer.

When you put your name on an article that's written by ChatGPT or Claude, you're basically putting a provenance mark on it saying: I take responsibility for this. I haven't written it, but I'm putting my name on it.

That doesn't give you copyright, but it does give you liability for the content.

So the short answer is: there's no right to transfer. Putting your name on it doesn't give you a right. It might give you liability, but no right.

Is there a cut-off point — 20 percent my work, 50 percent, 80 percent — where an AI-generated work becomes my work? Or is it just kind of up in the air now?

To read this story, log in or subscribe

Enjoy access to all articles and 25 years of archives, comment and gift articles. Become a subscriber for as low as €1,75 per week.

Read without limits
Already a subscriber? Login
Unlock article and share

Latest from AI

Reverse centaurs: EU approach to AI risks dystopia for human workers

Russia’s faith in ‘error-free’ killer robots and global divisions expose the moral vacuum in AI warfare

Latest from Digital

After France’s failed attempt, who will be first to shape Europe’s answer to children and social media?

EU action against ‘The Com’, a nihilist online extremist network, shows that security policies might be outdated

Latest from Interview

[Interview] From cheap drones to AI targeting: Ukraine’s robots still answer to human commanders

[Interview] German MEP Damian Boeselager: Locking 27 leaders in a castle and thinking they will deliver a good outcome is a categorical mistake

Latest from Europe today

Russia is adapting its drone-warfare strategy, with faster and deadlier ‘Shaheds’ (Ukraine Battlefield update, Day 1,636)

[Interview] Ukrainian frontline reporter Diana Bucko: ‘I have learned not only to report on the war, but also to live in it’

Copyright scholar Daniel Gervais: 'When you put your name on an article that's written by ChatGPT or Claude, you're basically putting a provenance mark on it saying: I take responsibility for this. I haven't written it, but I'm putting my name on it. That doesn't give you copyright, but it does give you liability for the content' (Source: Courtesy Daniel J. Gervais)