Creating content is the bread and butter for some. But Anthropic’s new watermarks have many content creators shivering. AI has made it so much easier, leaving us time to add that extra human soul to our organic ideas, which AI then turns into content following precise guidelines. Anthropic has rolled out watermarks, and for many, it feels like the Roman persecution. The echo created by influencers capitalizing on people’s fear and anxiety over this topic is making things worse.
Their message is plain: “I have a thing that lets you remove the watermarks.” The implication is that Anthropic’s watermarking is bad… and that in order to keep benefiting from content production, you need their repos, their solutions. But, as this article will show you, there is more to the story. We’ll unwrap the true core of the matter, and what Anthropic rolling out watermarks really means for you, whether you’re a local business with a blog or a content creator using AI to generate faceless content.
The tired eye doesn’t see the copycat — the tired ear doesn’t hear the echo
My social media feed is full of influencers talking about how Anthropic has watermarks now and how to remove them. Most of the videos end the same way: “There is a repo that can fix it. Comment ‘xyz’ and I will send it to you.” I want to make this article about my long-standing SEO experience. I have been doing SEO for more than 13 years now. Back in the day, I did it because I wanted more people to find my articles. I started a blog from scratch and looked for ways to get attention. I discovered a two-sided hat and a three-sided story: black, white, and gray.
The three ways of SEO
White hat SEO follows guidelines and tries to adhere to policies. Then there is black hat SEO. These are people, in every age and phase of the internet, who find ways to game the algorithms. I say “algorithms” in the plural because, contrary to popular belief, more than one algorithm works in SEO. Several algorithmic systems evaluate internet content and index or rank it for searches.
Black hat SEO has only one goal: game the algorithm to get a better search ranking so content lands in front of people. That content is either meant only to game the algorithm, or to take people out of their hard-earned money. Either way, there is something ironic here. I will show you why, in principle, there’s no difference between many of today’s influencers and the spam content creators of the past who used black hat SEO. But more on that further down.
Then there is gray hat SEO. This is the line between white and black, where the transition, or gradient, is hard to discern.
What Google actually says about SEO
Most people watch YouTube videos made by people who rely on buzz topics to draw attention to their channel. They also capitalize on people’s mental biases to bypass our reasoning and get engagement. It’s not inherently bad or good. I would consider it a gray zone. Advertising uses it, marketers use it, and magicians have always used it. But when it comes to content and what standards are expected of content creators, why not hear it straight from the horse’s mouth: Google. They have a whole blog section on SEO-, AI-, or automation-created content.
They state clearly that: “Google’s ranking systems aim to reward original, high-quality content that demonstrates qualities of what we call E-E-A-T: expertise, experience, authoritativeness, and trustworthiness. Our focus is on the quality of content, rather than how content is produced.”
Why doesn’t Google Search ban AI content?
Publishers have long used automation to create useful content. AI can assist with and generate useful content in exciting new ways. That means Google doesn’t care whether you write the content or your automation writes it, as long as it isn’t made to manipulate rankings. Google looks at other things, and even many so-called SEO experts don’t rely on them enough. It’s like a mental framework, a mentality, a paradigm.
On the one side are people who think: “SEO will help me get my content seen by more people, because it’s on page #1 or surfaces, and thus I will earn more money.” Or the other version: “The more I appear on page #1 of Google, the more traffic and clicks I can attract, and the more I will earn from my website’s banner ads, via Google ad cents.” On the other hand, some people have a different mentality. They think: “If I can optimize my content for the exact person who is my ideal client, so that they find my content so helpful that they can’t avoid browsing my website for more, and eventually become my customer after they go through know-like-trust.” Both mentalities do SEO. Can you figure out which one? If you said the second mentality, then you are right.
There is no formula, no deterministic formula, to beat Google’s algorithm. Searching for one means sliding along the SEO gray-zone gradient toward black-hat SEO. Before you know it, you are creating content focused only on schmoozing the algorithm, completely disregarding what readers and your target audience actually want and need when they search online. But again, most social media influencers love to ride trends. In fact, any course or book like the one written by a good friend of MrBeast will tell you that trends and patterns are your best friend. They say: just follow the trend, copy what works, tweak it a bit to make it yours.
The loud echo of copycats and the demise of useful insights
Because every influencer analyzes content only for performance metrics, most content is shallow. In its essence, it is a bunch of colorful eggs with different ribbons, but the content is the same; the message is the same. Multiply that across thousands of influencers (and I am only counting the top 10%), and you get a kind of message echo chamber. And the result is me, swiping through social media and hearing the new message: “Anthropic just added watermarks to all their content. I know of a GitHub repo that can help you remove it. Comment xyz, and I will send it to you.”
Now watch how they use automation to serve you the content after you comment. But before that, a reframe: in the old internet days, you used deterministic, algorithm-based software to spin content on websites and produce variations to game the algorithm. Fortunately, I never went down that rabbit hole. But being in this game as long as I have, you get around the block and meet people.
Anyway, these content creators, by analogy, use content not for the benefit of their audience (I dare say) but for the sole purpose of getting more views, more likes, more reach, and more… well, more of whatever lets them monetize their online presence. And therein lies the issue. This is the same difference we discussed earlier, when I asked you to choose a mentality. One of them was a mentality that made content only for its own benefit and crossed the gray line, creating content with zero nutritional value but high clickbait appeal to game the algorithm. Now, the algorithm in question is not Google’s. It’s Meta’s, Instagram’s, X’s, YouTube’s… bottom line.
There are only a few creators who sit down and say: “How can I knock my audience out of their seats? How can I create content that wows them, that enlightens them, that feeds and nourishes them and satisfies their hunger?” Unfortunately, those people are becoming fewer, because the misconception is created that gaming the algorithm is the way to win, and that doing it the white-hat way only gets you… well, nothing. But that is wrong.
What the algorithm looks for can only be found with effort and virtue
Google is very clear about what it looks for in content, or considers the content it rewards and does not penalize: “Our focus on the quality of content, rather than how content is produced, is a useful guide that has helped us deliver reliable, high-quality results to users for years. […] When it comes to automatically generated content, our guidance has been consistent for years. Using automation — including AI — to generate content with the primary purpose of manipulating ranking in search results is a violation of our spam policies. […] This said, it’s important to recognize that not all use of automation, including AI generation, is spam.”
How will Google Search prevent poor quality AI content from taking over search results?
Poor-quality content isn’t a new challenge for Google Search. We’ve been tackling poor-quality content created by both humans and automation for years. We have existing systems to determine content helpfulness. Other systems work to elevate original news reporting. Our systems continue to be regularly improved.” In other words, Google is telling us that it uses its algorithms to prevent spammy content from flooding the internet and tainting search results. As a business whose core purpose is to serve you content and help you find information online and offline, their greatest asset is your trust.
If you don’t trust the quality of their searches, and if you don’t get meaningful satisfaction from engaging with the search engine or a Google product like YouTube, then they lose you as a customer. Said another way, Google cares about the people who use Google.
Say it to the person, not the search algorithm
Google cares about the people who use Google, and wants more people to trust and use it. That also means Google earns more money, because advertisers will want their ads on its platform. That said, serving you the best-quality content for your search intent also means Google needs to remove the “bad” content. That is content that does not align with Google’s interest, which means content that is not created out of care for another human but purely to game the system. That’s the type of content Google and its SEO engine, which we call the algorithm, focus on finding, penalizing, and removing.
The stick: Google’s warning
Google will warn you by penalizing your content. Then Google will remove your content if it crosses the line. But why? Because Google needs a clean search engine. Clean means a search engine free of spammy content not made for humans, and Google cares about humans. Do you see how mentality serves you here, and either puts you on Google’s friend list or scratches you from it?
Does Anthropic’s watermarking mean the age of AI-generated content is coming to an end?
No, I say, far from it. But don’t take my word for it. Here is what Google has to say: “How will Google address AI content that potentially propagates misinformation or contradicts consensus on important topics? These issues exist in both human-generated and AI-generated content. However content is produced, our systems look to surface high-quality information from reliable sources, and not information that contradicts well-established consensus on important topics. On topics where information quality is critically important — like health, civic, or financial information — our systems place an even greater emphasis on signals of reliability.”
Google is handing us the ultimate SEO playbook for the AI era. But it comes at a price, and it is only visible to those who look at SEO through the right lens. Because what Google is saying is: you can create AI content, but we will penalize it for several reasons, whether it is human-made or AI-made. The reasons for getting penalized by Google are:
- Your content is scammy: You are literally trying to play the system and the users. Your content is meaningless, like paper versus an apple. Your content doesn’t care who reads it or what the reader gets from it.
- Your content is gaming the algorithm: Back in the day, people would do what is called keyword stuffing. They would stuff a bunch of keywords into the page’s meta description. The human would not see it, but the crawler, which was much simpler a decade ago, would think, “Cool.” And these people would win the search race. But then things changed. The algorithm and neural networks became more advanced. People started to experience a drop in traffic, a drop in ad revenue, a drop in the outcomes they expected from not caring about the reader. I will not go into the debate about censorship, because going down the next reason for getting penalized inevitably opens that debate.
- Misinformation: The most obvious form of misinformation is in marketing. People make unsubstantiated claims and often straight-out lie in their content and messaging. For example, someone might claim that their supplement can produce a certain result. If that claim isn’t substantiated, it can also violate legal boundaries, not just Google’s content guidelines. You can also fall into this trap if you don’t proofread and fact-check what you write. That is why professional teams have an editor and a chief editor to review the content and make sure it follows their guidelines. You could be writing an article about something, and the AI could hallucinate and confidently say something wrong. If you don’t double-check the content against the facts, that’s on you, not on the AI.
How will AI content rank on Google if Google knows it’s AI-generated? (Good or bad?)
I think we are getting to the heart of things now. The fear trigger and anxiety source behind influencers currently pushing out five different angles of the same message about Anthropic and its new watermarks is based on one question: Will Google punish me if it finds out my content is AI-generated? To answer this question satisfactorily, and leave nothing to speculation, we need to put a few things in order.
Google does not care if you create content with AI. And it will penalize you if your content is spam, even if it is 100% human-written. “Appropriate use of AI or automation is not against our guidelines. This means that it is not used to generate content primarily to manipulate search rankings, which is against our spam policies.” That means that as long as you don’t intentionally manipulate search rankings or violate Google’s spam policies, then creating AI-generated content is fine.
Does it matter whether that content is watermarked? No. Unwatermarked content can be spammy and violate Google’s spam policies. Watermarked content can be valuable to readers and meet Google’s spam policy standards. In fact, even though AI is the new kid on the block: “Automation has long been used in publishing to create useful content. AI can assist with and generate useful content in exciting new ways.”
Notice the two words highlighted twice in one Google quote? “Useful content.” …That’s all Google expects. That’s all you expect when you have a health goal and search online for supplements that could help you get there. So, bottom line: Google doesn’t care whether your content is AI-generated or watermarked. In fact, the article I wrote before feeding it to AI to tighten it up or flesh it out was longer than the article Anthropic’s AI printed out. That means I still put a whole load of time, work, and effort into my content.
Why then is Anthropic adding watermarks to content?
To understand why the watermarks exist, we need to understand why anyone would want to signal: “This was made with AI.” We live in a time when many bad actors use deepfakes, spam calls, and impersonation of relatives in need of help to con people. Regulators have decided to put their foot down, and the EU AI Act is the result. The AI Act aims to level the playing field by preventing AI misuse and protecting vulnerable consumers.
Its labeling rule for text is narrower than most people think. It covers AI-generated text published to inform the public about matters of public interest, such as health, politics, or the environment. Even then, text that has undergone human review or editorial control, with a person taking responsibility, is generally exempt. (Heuking law firm analysis) Anthropic’s own support page makes a second point. A Claude mark can be present even when the ideas and words originated elsewhere. So if you write something entirely yourself and use Claude to fix your punctuation or translate it, the text can come back carrying a mark that a detector reads as AI. (Anthropic: How Claude marks AI-generated content)
In the end, it comes down to how and why you make your content, not what you use.
What the watermark really changes
So what does Anthropic’s watermark actually change? For creators who care about readers, nothing. It’s a response to regulation, a signal that a model was involved somewhere along the way. It isn’t a verdict on your content’s quality, and it isn’t proof that Claude wrote it. Google has been telling us the same thing for years. It doesn’t ask how your content was made. It asks why. Content made for a person who needs it- content that is useful, original, and honest- is what gets rewarded, whether a human typed it or automation helped. Content made to game the system gets penalized, with or without a watermark.
After more than 13 years in SEO, that is the lesson I keep coming back to. White hat, black hat, and gray hat were never about tools. They were about mentality. Spinning software yesterday, a “comment xyz” repo today, a new trend tomorrow: the tool changes, the mentality stays the same. So before you hand your attention to anyone selling a way to remove a watermark, ask yourself the only question that has ever mattered: who is this content for?
Answer that honestly, and the watermark is the least of your worries.
Follow us on X, Facebook, or Pinterest
