AI “slop” has a new enemy. On Tuesday, Substack said it wants to fight back against the flood of AI-generated clutter that it believes is overwhelming the internet, and it is starting with a practical move: giving Substack users access to a tool created by Pangram, an AI-detection startup.

This is not a vague manifesto. It is Substack CEO Chris Best using the platform’s product surface to answer a worry many users already have: that more AI use means less human signal. For now, the company’s effort amounts to detection. Down the line, Substack says it may introduce other features, including the ability for Substack readers to keep AI-generated content out of their feeds.

To understand why this matters, you have to zoom out to the bigger timeline the source points to: AI is probably inevitable, and most systems of communication are heading toward being intertwined with AI in some way. At the same time, consumer sentiment is increasingly split. Some backlash has targeted the infrastructure side, like the anti-data-center push picking up steam. Other reactions have been cultural and brand-based, like people scolding A24 for working with Google’s DeepMind AI lab. The key point from the Business Insider piece is that even people who you might not assume are worried about AI are signaling they understand the concern.

Substack’s positioning is designed to sit inside that tension instead of detonating it. Best takes pains to say Substack is not opposed to AI use in general or in the creation of Substack content. The company’s language is about choice, not purity. On the platform, some creators are “human-writing maximalists,” while others are publicly exploring the frontier of using AI tools while sweating the details to make work they stand behind. Substack says it uses AI already at the company level, including to write software, do research, and build product features like clipping and translations.

So what is Substack actually afraid of? The source frames the giveaway as a specific reference to where things could go if AI growth is not curtailed. Best uses the “L-word” to describe a future where Substack becomes LinkedIn. It is an easy trigger because it is loaded. LinkedIn is not just another platform. It is shorthand for career content that can feel formulaic. The source immediately complicates that instinct by pointing out that the internet is full of both crummy Substack writing and occasionally insightful LinkedIn posts. In other words, the platform is not making an argument that all AI is bad or all human writing is good. It is betting that the mix of what shows up in people’s feeds will shape their experience.

And Substack is not operating in a vacuum. The source notes that a couple months ago, LinkedIn said it was trying to fight AI slop, too. That matters because it suggests the “label and filter” approach is not just a Substack quirk. It is turning into a broader competitive pattern: platforms treating AI-generated content as an inventory problem. What do you serve? What do you demote? What do you label? Who gets to decide?

That gets to the deeper debate the article raises. Are people opposed to any use of AI in writing or communication? How much is too much? Who decides how much of it people get to see? And honestly, the piece challenges the idea that there is a single moral line everyone agrees on. It asks whether someone at the SVP level who uses ChatGPT to write a boring widget workflow post should really be a crisis. The underlying reality is that many people do not want a full philosophical fight. They want control over the texture of what they consume.

This is why the “AI-free” angle could become a business model, not just a culture war. The source’s argument is that if AI is unstoppable, then the likely outcome is a labeling regime. In that world, the average person spends a lot of time encountering AI, often without thinking too hard about it. But for some consumers and some brands, an “AI-free” label becomes a selling point, similar to how some customers pay a premium for organic food while others buy whatever they find. The article even uses a personal analogy: loving organic hot dogs while eating plenty of junk food. Translation for executives: preferences can be contextual, not binary.

So the second-order implication is straightforward. Even if your product team believes AI integration is inevitable, you still need to plan for consumer choice, labeling, and feed-level controls. Because the moment platforms offer filters or labels, “human signal” becomes a feature customers can buy with their attention. Substack is betting that this is where differentiation will live.

For boards and leadership teams across media, creator economy tools, and social distribution, the strategic stake is time. If you wait to react, you will end up building features after competitors already trained users to expect labeling and filtering. And if you ignore the demand entirely, you risk building an experience that feels like LinkedIn slop to your audience, even if your internal AI use is responsible. Substack is not proving that AI-free is morally superior. It is showing that consumers will pay, click, and return to products that make AI visibility a choice.