Why the internet makes more money when it captures your attention and refuses to let go
A thirteen year old shouts “six seven” while lifting one hand and lowering the other, for reasons nobody, including the rapper who wrote the line, can fully explain. By the time most adults learn what it means, it will already have been declared dead by the same children who made it popular. In the same week, a toilet with a human head sings on a loop to an audience in the hundreds of millions, and a stream of AI generated cats narrates events that never happened in voices that belong to nobody.
Almost everyone who watches this content will tell you they find it stupid. Teachers complain about it. Parents worry about it. Oxford University Press named the phenomenon its word of the year for 2024, choosing the term “brain rot” after a public vote of more than thirty seven thousand people, a term that had first appeared in print not on the internet but in Henry David Thoreau’s Walden, written in 1854, when Thoreau used the phrase to criticize a society that let demanding, complex ideas lose out to simple ones, and saw it as a sign of declining intellectual effort.
So here is the puzzle. If nobody wants this content, if even the people making it will admit under pressure that it is meaningless, why does there keep being more of it?
The internet became remarkably good at measuring attention. It never became equally good at measuring value. That single gap, more than any platform, any app, or any generation’s supposed decline in taste, explains why the content everyone calls brainrot keeps winning.
What gets counted
A click can be counted. A second of watch time can be counted. A share, a comment, a return visit the next day, all of these can be counted, timestamped, and fed back into a system that adjusts what it shows you next. What cannot be counted nearly as easily is whether a piece of content taught you something, whether it made you calmer or more anxious, whether you will remember it in a week, or whether your life is in any way better for having seen it.
Faced with that gap, the same logic became increasingly dominant across the largest digital platforms: optimize for what can be measured. Not because engineers set out to make people worse off, but because a company cannot manage a number it cannot see, and understanding was never a number anyone could see.
This is not a new problem, and it did not begin with social media. The economist Herbert Simon described it in 1971, decades before a smartphone existed, when he observed that a wealth of information creates a poverty of attention, and that any system flooded with information has to find some way to decide what deserves a person’s limited time. Simon was writing about libraries and organizations drowning in paperwork. He could not have known that his observation would become the design brief for the most profitable machines ever built.
Once attention became the resource in short supply, it also became the resource worth selling. Every platform’s business model runs on the same basic sequence: attention becomes engagement, engagement becomes data about what keeps a person watching, and that data becomes the raw material for advertising that gets more precisely targeted and more valuable with every session. Nothing in that sequence requires the content to be good. It only requires the content to hold.
Why boring loses
In 2012, YouTube changed how it decided which videos to promote. Until then, the platform had mostly ranked videos by view count, a metric that rewarded a compelling thumbnail and a misleading title more than it rewarded the video itself. People clicked, watched a few seconds, and left. YouTube’s own data showed that total time spent on the site was growing more slowly than the number of clicks, which meant the platform was succeeding by a metric that no longer reflected whether anyone was actually satisfied. So YouTube switched to ranking videos by watch time instead, a change the company later described publicly as a way to reward videos that kept people watching rather than videos that merely got clicked.
That single decision reshaped what an entire generation of creators made. It is also the clearest example of the deeper pattern behind brainrot. The system cannot directly measure whether a video is good, so it relies on behavior as a proxy, most often whether people keep watching.
A calm, careful explanation of a complicated idea competes for attention against a clip that resolves in three seconds and requires nothing from the viewer. A nuanced argument competes against outrage, which processes faster and provokes an immediate reaction. Under those rules, content drifts toward whatever is fastest to understand, easiest to feel something about, and most repeatable. Nobody needed to decide that simple content should beat complex content. The system was built to notice, automatically and constantly, which one people kept watching, and to show more of it.
The feed has no floor
A film gets one chance to hold your attention before the credits roll and you have to decide what to do next. A short form feed never gives you that decision point. One video ends and another has already begun, chosen by a system that has been quietly noticing what makes you stay through every video that came before it.
This is a genuine structural break from how media used to work, not just a difference in video length. Regulators have started to treat it that way. In February 2026 the European Commission issued preliminary findings that TikTok’s infinite scroll, autoplay, and recommendation system violate the EU’s Digital Services Act, arguing that the constant stream of new content shifts users into what the Commission called an autopilot mode and reduces their ability to stop on their own. The complaint was not about any individual video. It was about the absence of a natural stopping point, a feature the Commission said the platform would need to redesign.
That absence is not a bug that slipped through testing. Removing the stopping point gives a platform more opportunities to extend a session and learn what keeps a person there. The product being built was never really the individual video. It was the sequence, and a sequence that never stops has more chances to earn a longer one.
Why nonsense travels
This is where “six seven” belongs in the story, and where it becomes useful rather than merely baffling. The phrase comes from “Doot Doot,” a song by the Philadelphia rapper Skrilla that circulated informally in December 2024 before its official release in February 2025. Nobody, including Skrilla, has ever settled on what the line means. Some listeners connect it to a police radio code for a death report, others to a street in his neighborhood, and the rapper himself has said he never wanted to pin the phrase to one meaning at all. What actually made it spread was a basketball player listed at six foot seven, a hand gesture a teenager on his team improvised to go with the sound, and millions of people who could copy both in under two seconds without needing to understand any of it.
That is precisely why it worked. Nonsense has always traveled, from novelty songs to absurdist catchphrases, long before there was a feed to spread it on. What is different now is how efficiently a platform can reward that kind of content. Instant recognition and instant repeatability convert directly into reach. A viewer does not need to resolve what “six seven” means to imitate it, remix it, or attach it to their own video, and every imitation counts as another data point telling the algorithm this content deserves more distribution. People spent months arguing over what the phrase actually meant. None of that argument was necessary for it to spread. The imitation was.
The creator is caught in the same current
It is tempting to make platforms the only villain in this story, but the people making the content are working inside the same incentive structure, watching the same numbers everyone else watches. A creator posts something thoughtful and it gets a modest response. They post something simplified, louder, or stranger, and it outperforms the thoughtful version by a wide margin. Repeat that pattern a few dozen times and the lesson is not subtle.
Some creators resist it anyway, and pay for that resistance in reach. Others adapt within a few uploads, not because they stopped caring about the work, but because the dashboard in front of them updates in real time and the rent does not wait for a slower kind of content to find its audience.
Who actually gets paid
Somewhere behind every viral clip sits a business model, and it is worth following the money before concluding that any of this is accidental. In its full year 2025 results, Meta reported advertising revenue of roughly 196 billion dollars, accounting for close to 98 percent of the company’s total revenue for the year, driven in large part by an increase in ad impressions delivered as user engagement rose. Meta’s own quarterly reporting has tied that growth directly to time spent on its apps and the volume of content its recommendation systems can serve in that time. More time inside the platform creates more opportunities to serve advertising and more behavioral signals that make that advertising easier to target and price.
Around that core business sits a wider industry of advertisers, ad exchanges, and measurement firms, all of whom are compensated in some way by how much attention passes through the system, whether or not any of that attention was well spent. The older boundary between content and advertising has thinned as well. Global spending on influencer marketing alone reached an estimated 33 billion dollars in 2025, a figure that captures only part of the broader sponsored content economy, much of it aimed at posts built to look and feel like ordinary content rather than an advertisement, because the same design choices that make content engaging also make it an effective vehicle for commerce. Brainrot fits inside that same machine, a visible, low cost byproduct of a much larger system built to convert attention into revenue.
When content becomes infinite
Until recently, there was a natural limit on how much content could exist, because making it required a person’s time. That limit is disappearing. Google reported that its Veo 3 video model had been used to generate more than 70 million videos in the roughly two months after its May 2025 launch, a volume that took human creators on early web video platforms years to reach. The tools keep accelerating. Generative video, narration, and music can now be produced at a speed no group of humans could match, and the cost of producing another clip keeps falling sharply with each new model.
The real danger here goes beyond sheer volume. It is content produced specifically to be tested against an individual viewer’s own attention patterns, adjusted in real time based on what that particular person has already proven they will watch. A feed stops being a single broadcast that everyone sees and becomes something closer to a personal instrument, tuned to whichever version of a person is easiest to hold onto.
Human attention is not expanding to match this. People still have the same number of waking hours they had a decade ago, while content supply can now grow far faster than any audience’s ability to absorb it. That mismatch is where the real scarcity sits. Content used to be the limited resource. Now a mind willing to sit still with one thing at a time is the limited resource, and there is no update that manufactures more of it.
Children make the pattern visible, not exceptional
Children are usually where this argument goes first, and for a reason. They have more free time, attention and self-control still developing, and decades of potential engagement ahead of them, which makes them valuable to any system built around long term retention, and they have the least practice recognizing when a piece of content was built specifically to hold them.
Australia responded to this directly. On December 10, 2025, the country enforced a nationwide ban preventing anyone under sixteen from holding accounts on major platforms including TikTok, Instagram, Snapchat, YouTube, and Facebook, becoming the first country to implement restrictions at that scale. Within about a month, Australia’s eSafety Commissioner reported that age restricted platforms had removed access to close to 4.7 million under-16 accounts nationwide. Whether the ban actually protects the children it targets is still being tested, and some early reports describe children working around the age checks. But the law itself is an admission that the design of these platforms, not merely what individual children choose to watch on them, was the thing that needed to change.
Still, this is not only a children’s issue. Adults spend hours inside the same systems, built on design principles that remove stopping points for a developed brain almost as effectively as for a developing one. Children just make the mechanism easier to see, because the gap between what a platform optimizes for and what is good for the person using it shows up faster in someone who has had less time to build resistance to it.
Attention becomes something people pay to protect
A strange market has grown up around this problem, made up of people paying money to get back a resource that used to be free. Industry researchers put the global market for app blockers alone at roughly 2.4 billion dollars in 2024, a figure that does not even include the separate, harder to size businesses built around screen time trackers, digital detox retreats, and subscription products explicitly marketed as free of ads and algorithmic feeds. First, an entire industry profits from making attention difficult to keep. Then a second industry profits from helping people take it back.
That second market is a signal worth taking seriously. When focus becomes something people have to pay to protect, it has effectively become a luxury good. Deep, undivided attention is turning into a form of advantage, useful to a student trying to learn something complicated, an employer looking for someone who can sit with a hard problem, or a creator whose audience trusts them enough to sit through something longer than a few seconds. In a media environment flooded with cheap, abundant, algorithmically optimized content, the ability to hold a person’s attention honestly, without tricks, is becoming rare enough to be valuable in its own right.
Why nobody can simply switch it off
Governments are trying to respond, and the TikTok case in the European Union shows both what regulation can do and where it runs into trouble. Regulators can order a single company to change a specific feature, disable infinite scroll, or build in mandatory screen time breaks. What they cannot easily do is regulate a pattern that exists identically across nearly every major platform. The harmful outcome emerges from millions of individually unremarkable pieces of content, produced by millions of different creators, all competing under the same measurable rules.
A law can be written against a chemical, a weapon, or a specific defective part. It is much harder to write one against a pattern that shows up, in slightly different form, across much of the industry’s product design, produced by millions of separate creators none of whom individually did anything a court could point to.
What the incentives actually built
None of this required anyone to want a worse internet. It required a system that could count clicks and watch time with precision, and that had no equally precise way to count understanding, memory, or wellbeing. Given that gap, the system did exactly what any well optimized system does. It produced more of what it could see and less of what it could not, and it did this relentlessly, at a scale no single person or company fully controls anymore.
Brainrot is the visible output of a system that became extremely efficient at pursuing the one goal its business model gave it a precise way to measure: attention, not the value that attention was supposed to serve. The machine was never broken. It has been working, with real precision, toward a target that was never quite the same thing as a better life, and it will likely keep doing so for as long as attention stays the easiest thing in the room to count.
