We talk about slop like it's new. Like AI invented it. But slop has been with us for as long as there's been an incentive to produce more and care less. What changed isn't the existence of slop. It's the cost of making it. And now that AI has pushed that cost to nearly zero, something interesting is happening. People aren't just annoyed. They're hungry for the opposite. They want raw, unpolished, unmistakably human things. The rising popularity of "boring videos," people just talking to a camera with no edits, no script, no production, is one of the clearest signals of that hunger. Slop won Merriam-Webster's word of the year in 2025, but the thing it describes has been here all along.

The word is new, the problem isn't

Merriam-Webster defines slop as "digital content of low quality that is produced usually in quantity by means of artificial intelligence." That definition captures the current moment, but the pattern is older than the internet. The first spam email was sent in 1978 on ARPANET. A marketer at Digital Equipment Corporation blasted 400 users with an unsolicited pitch for computers. It reportedly made millions and annoyed everyone who received it. By the mid-1990s, spam was already a growing plague, and email providers were scrambling to build filters. Then came the content farms. In the late 2000s and early 2010s, companies like Demand Media and Associated Content hired armies of freelancers to churn out SEO-optimized articles designed to rank on Google, not to inform anyone. The articles were technically written by humans, but they were hollow. They existed to capture search traffic and serve ads. Google's Panda update in 2011 was a direct response to this flood of low-quality content. Clickbait followed. Upworthy-style headlines. "You won't believe what happened next." Listicles engineered for shares. Facebook's algorithm rewarded engagement above all else, and publishers optimized accordingly. The content wasn't AI-generated, but it was slop by every meaningful measure: produced cheaply, distributed at scale, and designed to extract attention rather than deliver value. Social media platforms then became breeding grounds for engagement bait. Stolen videos re-uploaded with new captions. Rage-farming posts designed to provoke comments. Fake accounts posting recycled content to build follower counts for sale. The playbook was always the same: minimize effort, maximize distribution. Slop has always been the natural byproduct of systems that reward volume over substance.

What AI actually changed

If slop existed before AI, why does it feel so much worse now? Because AI didn't create the incentive to produce slop. It obliterated the friction. The constraint on slop was never desire, it was effort. Even content farms needed to pay humans a few dollars per article. Even spam required some infrastructure. AI removed those constraints almost entirely. Today, a single person with a laptop can generate thousands of blog posts, social media captions, product descriptions, and fake news articles in hours. Generative image tools like Midjourney and DALL-E produce visuals that used to require a designer. AI voice cloning can narrate podcasts and videos without a human ever speaking. The LA Times reported that nearly 75% of new web content is at least partially AI-generated. The result is a volume problem unlike anything we've seen. Mentions of "AI slop" across the internet increased ninefold from 2024 to 2025, according to Meltwater. Facebook feeds are clogged with AI-generated images of shrimp Jesus, vegetable sculptures, and disabled veterans designed purely to farm engagement. Amazon's Kindle marketplace is flooded with AI-written books. AI-generated podcasts are pushing out independent creators. Stanford researchers documented over 120 Facebook pages posting at least 50 AI-generated images each, collectively receiving hundreds of millions of engagements. One AI-generated image post was among the 20 most-viewed pieces of content on all of Facebook in Q3 2023, with 40 million views. The quality floor hasn't changed much. Slop was always low quality. But the quantity ceiling has been blown off completely. And because platforms still reward engagement over authenticity, the economics favor flooding the zone.

Why people hate it

The backlash against AI slop isn't really about AI. It's about the feeling of being disrespected. When someone encounters AI-generated content that's pretending to be human, there's an implicit message: your attention isn't worth real effort. The content exists to extract something from you, a click, a view, an ad impression, without offering anything genuine in return. As one etymology writer put it, slop is "form without content." A pig's fodder is simply meant to be food, not a type of food that does something. Instagram slop is simply meant to go viral. It's molded in the image of the medium, recreating an idealized "successful message" rather than being a message itself. This is what people are really reacting to. Not the technology, but the hollowness. The sense that more and more of what they encounter online was never meant for them. It was meant for an algorithm. And the problem compounds. As AI slop crowds out human-created content, the signal-to-noise ratio degrades. Finding genuinely useful information, real recommendations, honest perspectives, becomes harder. Trust erodes. The internet starts to feel less like a place made by people and more like a machine talking to itself.

The authenticity correction

Here's where the story gets interesting. The same wave of AI slop that's degrading online spaces is also creating a powerful counter-movement. People are gravitating toward content that couldn't have been made by a machine. Raw, imperfect, unmistakably human content. And the market is responding. One of the clearest examples is the rise of what people are calling "boring videos." These are exactly what they sound like: someone sitting in front of a camera, talking. No jump cuts. No background music. No graphics. No script. Just a person sharing a thought, telling a story, or working through an idea out loud. These videos are thriving precisely because they're the opposite of slop. They require a human to show up, be present, and say something real. You can't fake the pauses, the half-formed thoughts, the moments where someone catches themselves mid-sentence. The "imperfection" is the proof of authenticity. This isn't limited to video. Podcasts and long-form interviews are experiencing a resurgence. Newsletters written in a distinct personal voice outperform generic content. Substack's growth is partly a story of people seeking out individual writers they trust over anonymous content mills. Forbes noted that lower-quality, authentic content often outperforms highly produced videos on platforms like TikTok. Reddit threads consistently show users saying the same thing: raw content feels like a real person talking, polished content feels like a brand performing. Imperfection signals honesty. Relatability beats perfection. Even AI search systems are starting to reward this shift. Content with a clear author, a strong point of view, and external validation is more likely to be cited by AI tools selecting sources. Generic AI-written content underperforms with both humans and machines. Authenticity is becoming a discoverability factor, not just a brand value.

The cost of being real just went up

There's an irony in all this. As AI makes it cheaper to produce content, it makes authenticity more valuable. Being real is becoming a premium. This is already playing out in marketing. When McDonald's and Coca-Cola used AI to generate holiday ads, the backlash was swift. Audiences felt the absence of human creativity in an emotional context. The "authenticity premium," the intangible value that human effort provides, turned out to be very tangible when it was missing. For individual creators, this is both a challenge and an opportunity. The challenge is that standing out in an ocean of slop requires more intentionality. You can't just make content. You have to make content that carries evidence of a human behind it. A perspective. A voice. A willingness to be imperfect. The opportunity is that the bar for "real" hasn't actually moved. It's the same as it always was: say something you mean, in a way that's yours. The difference is that in a world drowning in generated content, that simplicity has become rare enough to be remarkable.

Slop is a mirror

Slop didn't start with AI, and it won't end with it. It's a reflection of incentive structures. Wherever there's a system that rewards attention over substance, slop will fill the gap. Email inboxes in the '90s. Google search results in the 2000s. Facebook feeds in the 2010s. The entire internet in the 2020s. What AI did was hold up a mirror and make the pattern undeniable. The volume is so extreme now that we finally gave it a name. But the instinct to produce cheap, high-volume, low-effort content for profit has been with us from the start. The real question isn't how to stop slop. It's what we choose to make instead. The boring videos, the personal newsletters, the unscripted conversations, these aren't a trend. They're a correction. A reminder that what people actually want hasn't changed: something real, made by someone who cared enough to show up. The tools for creating slop will only get better. But so will our ability to recognize it. And every time someone chooses to sit in front of a camera with nothing but a thought and hits record, they're making a small bet that authenticity still matters. It does.

References