5 Disturbing Truths About ‘AI Slop’ That Go Deeper Than Bad Content
If you’ve spent any time on social media lately, you’ve probably stumbled into it: a bizarre, uncanny landscape of AI-generated images. Maybe you saw “Shrimp Jesus,” a surreal depiction of Christ fused with crustaceans, or a photoshopped LeBron James inexplicably milking a cow.
These images are often nonsensical, slightly off, and yet they rack up millions of views. They feel lifeless and strange because they are. What we’re witnessing is a new, unsettling digital folklore being written not by communities, but by algorithms.
This phenomenon has a name: “AI slop,” a term so fitting for the deluge of low-effort, mass-produced digital material that one could easily imagine Merriam-Webster declaring it the 2025 Word of the Year.
But this trend is more than just quirky, low-quality content cluttering our feeds. It’s a visible symptom of a fundamental and troubling transformation of the internet. Here are five disturbing truths about AI slop that reveal the deeper issues at play.
1. It’s Not Just Annoying, It’s a Symptom of the “Dead Internet”
The flood of AI slop is seen by many as concrete evidence for the “Dead Internet Theory”—the assertion that organic human activity online is being displaced by bot activity and automatically generated content.
More than just a feeling of emptiness, the theory posits this is part of an “intentional effort to manipulate algorithms, influence consumers, and control the population.” What was once a paranoid conspiracy theory is becoming a visible reality. One user, as noted by the Substack Etymology Nerd, counted 16 out of 20 consecutive Instagram Reels as AI-generated.
The internet doesn’t just feel dead; corporations are actively building the infrastructure to make it so, turning it from a place of human connection into a landscape of artificial content engineered solely to capture attention.
“We expect these AIs to actually, over time, exist on our platforms, kind of in the same way that accounts do…They’ll have bios and profile pictures and be able to generate and share content powered by AI on the platform.”
— Connor Hayes, Meta’s VP of Generative AI
2. The Name “Slop” is Intentionally Visceral
As we navigate this increasingly artificial environment, the language we use to describe it matters. The term “slop” wasn’t chosen by accident over words like “junk” or “trash.” It was selected for its powerful physical and emotional connotations.
As both Merriam-Webster and Psychology Today point out, the word summons images of pig feed or heaps of congealed, unappetizing food being shoveled into troughs. The term’s power lies in its appeal to our primal senses of purity and contamination, a linguistic immune response to a digital pathogen.
“Like slime, sludge, and muck, slop has the wet sound of something you don’t want to touch.”
— Merriam-Webster’s Word of the Year announcement
3. This Content is Engineered to Make You Anxious and Numb
This visceral revulsion isn’t just an aesthetic judgment; it’s the first sign of a deeper psychological toll this content is engineered to exact. Being inundated with AI slop can trap us in what humanities scholar Laura Glitson calls a “Slop-Doom Feedback Loop.”
According to an analysis in Psychology Today, the endless stream of surreal, meaningless content creates “affective noise” that overwhelms our attention, contributing to a generalized sense of doom, paranoia, and anxiety.
This leads to a “Normalization Paradox.” On one hand, the “mere exposure effect” suggests that the more we see AI content, the more we get used to it. On the other, the “black box effect”—our unease from not understanding how AI works—fuels a sense of foreboding.
“The internet makes me miserable for 80% of the time I’m on it, but I just can’t get out… Have I developed mild to moderate anxiety from constant exposure to news and social media that indicate we’re headed to unavoidable collapse? Sure have.”
— Reddit post cited in Psychology Today
4. AI Training on AI Data Creates a Degenerative Loop Called “Model Collapse”
The psychological decay we experience is mirrored by a technical decay in the very systems producing the slop. Researchers have identified a phenomenon called “Model Collapse,” which is like making a photocopy of a photocopy.
The first copy looks sharp, but with each successive generation, the image becomes blurrier, the colors fade, and the details in the shadows vanish. The AI, learning only from its own faded copies, forgets the richness of the original, human-created world.
In technical terms, models trained on data generated by other AIs begin to forget “improbable events.” The research paper on the topic states that the “tails of the original content distribution disappear.” This leads to a blander, distorted version of reality where common things are over-represented and everything else—the unique, the rare, the unusual—is forgotten.
“We find that use of model-generated content in training causes irreversible defects in the resulting models, where tails of the original content distribution disappear.”
— From “The Curse of Recursion: Training on Generated Data Makes Models Forget”
5. It’s Not an Accident, It’s a Business Model: “Slop Capitalism”
This technical degradation is not an unfortunate byproduct of new technology; it is the direct consequence of deliberate economic incentives. The term “slop capitalism” describes a business model where platforms like Meta actively encourage AI-generated content.
The primary motivation is to “crowd out actual human voices” and, in doing so, reduce the revenue paid to human creators. This is also seen on Spotify, which stuffs playlists with AI music, and in Google’s AI Overviews, which replace links to creators.
This business model doesn’t just entertain us into passivity; it preys on the same psychological vulnerabilities identified in the “Slop-Doom Feedback Loop.” It profits from the anxiety and division fomented by “affective noise,” because engagement is engagement—whether it comes from a bizarre image of Shrimp Jesus or from overtly racist content that the same algorithms promote because it generates clicks.
Conclusion: Rescuing Reality from the Slop Pile
AI slop is not merely a technical artifact; it is a cultural one. It reflects a society grappling with the degradation of its information ecosystems.
As our digital world becomes increasingly artificial, how do we ensure we don’t lose the original, human-made data that’s necessary to teach AI what reality even looks like?