But the discoverability crisis isn’t the only shift happening. There’s a second one running alongside it—quieter, but arguably more consequential for anyone trying to make a living as a creator.

This has massive implications for how I’ve built my business—and probably for how you’ve built yours. Feel free to read the sentence again and sit with it for a minute before continuing…Because it is painful in ways that only creators will truly understand.

For the past decade, digital content—ebooks, courses, newsletters, information products—could function as both the thing that built your audience and the thing that made you money. Write the book, sell the book. Build the course, sell the course. That was the model, and it worked.

I built an entire career on it. I have 30+ published nonfiction books. I know what it feels like when someone buys your knowledge and it pays your bills.

That business model is becoming much harder to sustain, though. Content is rapidly becoming the price of admission to the online world—the thing you must produce to build trust, demonstrate expertise, and attract an audience. But increasingly, it is not the thing you sell.

Not because content is becoming bad, unimportant, or “slop.” It’s actually the opposite—content is becoming foundational, and foundational is not necessarily directly monetizable. The money is moving upstream—from content itself to everything content enables—brand, intellectual property, relationships, experiences, community, tools, and more.

And if your business model still depends primarily on selling digital content as a standalone product...The math is getting harder by the month.

How Content Lost Its Pricing Power

Content has been losing its pricing power in stages for decades, but this stage has expanded content availability faster than anything I’ve seen in the past.

Each stage was triggered by a technology that made content production dramatically cheaper and faster, which overwhelmed the monetization model that had been working at the previous scale. When the supply becomes unwieldy, demand destabilizes and prices plummet.

In the beginning, distribution was controlled, so content commanded a price. For most of human history, producing and distributing content required real infrastructure. Put another way, production and distribution were prohibitively expensive for the average creator. Thus, publishing a book meant working with a publisher, and distributing educational material meant working with an institution, recording music meant working with a label, and so on.

Content wasn’t scarce in absolute terms—there were always more ideas than any individual could consume, even then. But the barriers to production and distribution kept supply limited and monetization controlled. People paid for information because access to it was expensive to provide.

More recently, production barriers fell, but discoverability systems kept content monetizable. The internet lowered production and distribution barriers dramatically. Blogging, self-publishing, online courses, YouTube, podcasting—suddenly anyone could produce and distribute content. Supply exploded. But discoverability systems—search engines, then algorithms—were still effective enough to connect good content with paying audiences. You could build a real business selling ebooks at $9.99, courses at $497, and programs at $1997. The content surplus was enormous, but the filtering systems made it navigable enough that quality content could still find paying buyers. Many people built businesses this way—including me!

Now, Generative AI has collapsed the final cost barrier, and pricing power has eroded further. Generative AI triggered the next supply explosion—one that hit content production across every format simultaneously. When anyone can generate articles, tutorials, summaries, educational material, videos, audios, and marketing copy at near-zero cost, the supply of information content becomes functionally unlimited relative to demand. And when supply is unlimited, price trends toward zero. The cost advantage that justified the price no longer exists.

This is where we are now. And the effects are showing up everywhere.

Content Leverage is Uneven Across Industries

A few caveats to this—not all content is commoditizing at the same rate. The commodity crisis is real, but it’s not flattening everything equally.

What it’s doing is flattening everything enough, and significantly, to make the default monetization model—package information, sell at scale—much harder to sustain for most creators in most niches.

Generic, procedural, and interchangeable information content is under the most pressure, of course—no surprise there.

And of course the large, distinctive voices with huge emotional resonance and intimate creator-audience relationships are likely to become more valuable as generic content floods the market—no surprise there either.

The edges are not the new part of the problem—it’s the broad middle that is where the problem lies.

And while this is bad news for most creators, it’s also bad news for platforms…Which means everyone is incentivized to solve the problem, and power shifts can happen within that solution.

The Collapse in the Middle

The clearest evidence that the middle of everything—the midlist authors, the non-blockbuster movies, the non-charting musicians, the character actors—is collapsing…

…Is in nonfiction education and information products—the sector that built the modern Creator Economy.

I’ve been watching six-figure revenue streams in several sectors of the information economy get wiped out over the past year, and not slowly. Entire business models that were printing money as recently as 2024 are struggling to break even in 2025 and 2026. I’m talking to creators I respect, people who were doing everything right by the old playbook, and the message is the same: What happened?

The collapse is not happening evenly, either. It’s concentrated in the $200–$2,000 course tier—the backbone of the information product economy—where value is hollowing out.

Generic educational content, mid-tier nonfiction books, and self-study digital products are all under pressure. AI can now provide good-enough versions of that information for free. Why pay $997 for a course on email marketing when you can ask an AI to build you an email strategy, write the sequences, and explain the reasoning?

Now, I’m all for innovation and I would like to stop paying $1k for courses on email marketing and just have a tool do it for me.

But think about how fast this happened. In 2024, a well-positioned creator could launch a $500 course to a warm audience and reliably generate five or six figures. By 2025, many of those same creators are seeing launch revenue drop by half or more—not because their audience left—no, the audience is growing, usually!—but because the perceived value of packaged information cratered. The information didn’t get worse, it just stopped being scarce and hard to find.

The biggest names in information and education businesses are responding by going upmarket—high-touch consulting, masterminds, advisory work with successful individuals and companies who can afford premium pricing. These are smart moves. But they’re also an acknowledgment that the old model—package knowledge, sell at scale—is breaking down.

Meanwhile, at the low end, content is becoming free or near-free—given away to build audience, not to generate revenue directly. On subscription platforms like Substack and Patreon, middle-tier creators are responding by moving their subscription offerings upmarket too, bundling coaching, calls, live courses, and more into their ongoing subscriptions.

The middle is where the pain is sharpest. And the middle is compressing fast.

Done For You (DFY) Is Replacing Do It Yourself (DIY)

There’s a deeper shift underneath the pricing collapse that I don’t think enough people are talking about yet.

People don’t want to learn things. They want to do things. Or more accurately, they need to get things done to survive in the current economy.

I know that sounds harsh. But sit with it for a second.

The entire information product economy was built on the assumption that people would pay to learn how to do something—that if they could just get the right (insider) information, they’d be able to implement the right solution and see success. Buy the course, study the material, implement the strategy. That was the model.

Instead of a book about how to build a sales funnel, people want a prompt or a tool that builds the funnel for them. Instead of a course on copywriting, they want software that writes the copy. Instead of learning email marketing strategy, they want Kit or Aweber or ConstantContact to write email sequences for them.

Done-for-you is replacing do-it-yourself, because increasingly, people can’t afford the learning curve. They need to execute now to keep things moving. They don’t have six weeks to work through a course—they need the funnel built by Friday.

This is structural, not temporary.

Again, the Shift is Uneven Across Industries

The content that’s losing the most value is procedural teaching—step-by-step instruction on how to perform a specific, repeatable task.

Teaching that involves judgment, strategic thinking, and contextual application is under different pressure to go upmarket. For this type of teaching, frameworks still matter, knowing which funnel to build and why still matters, but the delivery mechanism for that knowledge—the course, the book, the tutorial—is competing with tools that deliver the outcome directly.

Still, in both cases, information products, self-study, anything that was touted to generate “passive income,” is in big trouble.

From Content To Tools

The nonfiction creators who are adapting fastest are the ones who’ve noticed this. They’re replacing books with tools as their lead generation—instead of publishing a guide that teaches you how to do something, they’re building a prompt library, a template, or a vibe-software tool that does it for you, and using that as the entry point to their world.

I’ll be honest—this one hits close to home for me. I’ve spent twenty years building a body of nonfiction work that teaches creators how to do things. The catalog size was supposed to be the moat.

Yet, I’m watching the floor shift under that model in real time. The delivery mechanism? That’s changing fast, and many of us are starting over in a world where that 20 years and catalog we’ve worked so hard to build is no longer leverage.

That hurts—and I get it if you’re feeling it now. For me, the thing I thought would give me safety and stability (a large catalog) is not necessarily going to do so in the future. That’s real disruption with real consequences.

What This Means for Other Creative Fields

Creators who work primarily in storytelling and entertainment instead of education and transformation are following the same trajectory I’ve just laid out, but they have (maybe) a little more time, and the dynamics look slightly different in ways that matter.

In fiction, the unit of value has already shifted once. Traditional publishing made the individual book the unit. Indie publishing made the series the unit—because algorithms rewarded volume and read-through, and having a long series became a near-necessity for sustainable income.

The next shift will make the world—the intellectual property itself—the unit of value.

We already see this in the largest entertainment ecosystems—Marvel, Star Wars, One Piece, anime, gaming franchises, webtoon universes. The individual movie or book is not where the money lives. The money instead lives in the world—in licensing, merchandise, adaptations, experiences, and the fandom that forms around the IP.

Much of this is due to the massive collapse of production and distribution costs in film and television.

For example, a fiction author mostly had to go to a publisher if they wanted their books turned into a film or series. There was still a gate there…But those gates are rapidly opening to indie film and television makers, as well as all other storytellers.

These Collapsing Gaps Will Change Several Industries

I used to call this gap between books and film the Novelist’s Dilemma—the problem that independent fiction authors often do not have enough profitable offers to raise customer lifetime value (LTV) beyond a few hundred dollars. Even using direct sales, reader might buy a discounted bundle, maybe another bundle, maybe a new release, and then they effectively own the full catalog of the author’s work.

Because most fiction products are relatively low-priced and physical add-ons often eat up margins, even a strong ebook-focused business can hit a ceiling where lifetime value stays too low to support rising acquisition costs over time.

That dilemma is disappearing as publishing and other industries—Youtube, podcasting, and more—builds models to break through the gates that were once closed.

Books and other forms of static content aren’t going to disappear. But they’re becoming the entry point into a larger story ecosystem—the content-as-entry-fee principle applied to fiction. The money is moving to everything the book makes possible.

And honestly? For those of us who’ve been building sprawling fictional worlds for years, this might be the first time the industry structure actually matches what we’ve been doing. That’s...Exciting in many ways. Terrifying and exciting.

That said, for twenty years, creators have focused on productivity tactics—how to build more content faster because a large catalog used to be a moat. Now, it’s not a moat, just the dirt path through the forest to the clearing where the moat lives.

And for many creators, that will be an exhausting recognition.

The Third Path

Most of the conversation around the nonfiction collapse focuses on two responses:

• Keep selling content and watch helplessly as margins shrink, or hope that things stabilize)

• Go upmarket to new formats—consulting and services for transformation facilitators, or filmmaking and television for storytellers, which may not scale for certain niches, certain creators)

But there’s a third path that I think is potentially more interesting: become a trusted curator and tastemaker in your domain.

When information is infinite, the person who produces information competes with AI…But the person who filters information meaningfully—who curates, synthesizes, makes sense of a chaotic domain, and helps people navigate it—becomes more valuable. In an environment of overwhelming content, the curator’s judgment is the scarce resource.

This is what I’ve been building toward with Letters on the Future, even before I had this language for it. Not just producing content about the creator economy—but filtering, synthesizing, and making sense of a domain that’s moving faster than any individual can track without tools. The Signals sections at the end of these posts? That’s curation in action.

The New Math

Let me be direct about what all of this adds up to, because the core logic is simple even though the implications are enormous.

If content becomes easier to produce and less defensible as a paid product, then value moves to what content builds, activates, or unlocks.

The business models of the online world are swinging—or swinging back—to brand trust, relationship, connection, collaboration, community, large intellectual properties, and fandom.

Building an entire business primarily on selling digital content, courses, books, and information is becoming much harder to sustain.

And yet—you can’t build brand trust, relationship, connection, community, and fandom without a large body of content that explains your brand.

That’s the tension. And it’s the tension that defines this moment for creators.

Is Any of It Worth It Anymore?

This is the quiet question that some creators have started asking themselves.

Content used to be the product, and now it’s the entry fee, the marketing asset. The creators who understand this distinction—and build accordingly—have a significant advantage over those still trying to make the old math work. What you build on top of the content—the relationships, the community, the experiences, the IP, the tools—that’s the business.

But this is so uncomfortable. It’s uncomfortable for me, too—I’ve built a lot of what I have on the old model, and reorganizing around a new one is real work with real risk.

And it’s not just risk—what if you don’t have the content yet? I work with so many creators who are still at the phase where they are figuring out how to create consistently and build their catalog. For the last two decades, I’ve had to tell them that the one book they want to write—a huge accomplishment in itself—is not enough to make money as a writer anymore. And now I’m telling them that the catalog they’ve worked to build might not be enough either in the Infinity Era.

That’s a tough pill to swallow, and I can understand why anyone would struggle with whether their efforts are worth it.

That said, I’d rather see the shift clearly—and unveil the truth to those ready for it—than pretend the ground isn’t moving.

And…There is hope for what to do next, too. If a large catalog is no longer a moat, there may be paths forward that don’t require a large catalog. If long-time independent creators are getting disrupted too, you could be the disruptor.

Signals I’m Watching

Content Supply Explosion

AI-generated content has reached roughly 50% of new articles online. Graphite’s analysis of 65,000 URLs found that AI-generated and human-written articles hit a 50/50 split—up from about 10% before ChatGPT launched. Europol had estimated that up to 90% of online content could be synthetically generated by 2026. (Futurism / Europol via Futurism)

YouTube purged 16 major AI channels with 4.7 billion views and $10M+ in annual revenue. In January 2026, YouTube terminated or wiped content from channels that had built empires on mass-produced AI content—weeks after CEO Neal Mohan pledged to “reduce the spread of low quality AI content” in his annual priorities letter. A Kapwing study found 278 channels producing nothing but AI slop had collectively amassed 63 billion views and an estimated $117M in annual ad revenue. The platform is simultaneously encouraging creators to use AI tools while cracking down on AI as a content farm—a tension that highlights the instability of algorithmic discovery when the supply explosion hits platforms, not just creators. (Search Engine Journal / The Media Copilot)

The Collapse in the Middle

Course creators are reporting sharp revenue declines. A finance educator saw course sales drop 60% in three months; creators across niches are watching the “six modules + lifetime access” model collapse as AI provides free alternatives. The consensus: selling packaged information alone no longer works. (OllyRichards / Boss Project)

Marie Forleo, one of the biggest names in online education, is pivoting from courses to AI tools. Marie Forleo—whose B-School has served 80,000+ students—now sells standalone AI-powered products alongside her courses. "Revenue on Repeat" is six interactive AI tools with the tagline: "No modules. No curriculum. No homework. You bring your business. The tools do the heavy lifting." "Build Your $250k Offer" is a guided app that builds your offer, sales page, and emails for you in 90 minutes. (Marie Forleo Shop)

Amy Porterfield's Digital Course Academy now includes an AI toolkit in every module. The most established course-about-courses creator in the industry—$120M+ in course sales, 100,000+ students—has embedded a custom-trained AI assistant throughout her flagship program. The AI drafts sales page copy, brainstorms webinar talking points, simplifies tech setup, and clarifies next steps. The program has also added expert-reviewed feedback submissions and live Q&A, shifting the value proposition from information delivery toward transformation and high-touch support. (Sell Courses Online / Amy Porterfield DCA)

DFY Replacing DIY

Stanford’s Erik Brynjolfsson argues value is shifting from execution to judgment. As execution becomes commoditized by AI, the bottleneck moves to “asking the right questions and evaluating results”—what he calls the rise of the “Chief Question Officer.” (TIME)

79% of companies are now leveraging agentic AI for task execution. PwC’s survey of business executives shows that AI agents performing tasks—not just providing information—is already mainstream. The consumer version is following. (PwC)

Collapsing Gates Across Industries

AI tools are cutting indie film production costs by 50% or more. Indie filmmakers report that shorts that once required £40,000–£80,000 can now be completed for under £8,000–£16,000 using AI across script-to-screen processes. The barrier between “author with a book” and “creator with a world” is collapsing in real time.

Amazon launches AI Studio to cut film and TV production costs. Amazon MGM Studio’s new “AI Studio” aims to bridge the gap between existing AI tools and what directors actually need—with a closed beta launching in early 2026. The same infrastructure compression that reshaped publishing is now hitting film. (The Decoder)

Bain & Co. finds AI can cut blockbuster film costs by 15-20% without replacing creatives. A $200M sci-fi film could save $30-40M through AI-assisted pre-production and post-production, while a $100M family film could cut 20% off time and budget. The study explicitly warns studios to “spurn” the idea of replacing creative talent—the savings come from streamlining production processes, not eliminating people. This is the same infrastructure compression dynamic that’s reshaping publishing, now hitting film at institutional scale. (The Wrap)

An AI short film now costs $80-$130 to produce. The traditional equivalent: $5,000-$30,000. MindStudio’s detailed cost breakdown for a 3-minute narrative short shows that video generation, AI voice, AI music, scripting, and post-production can be done for under $200. Even a bare-minimum traditional DIY production struggles to stay under $1,000. The financial barrier to indie filmmaking has dropped by an order of magnitude—which is exactly the kind of gate collapse that opens new paths for fiction creators who want to expand their worlds beyond books. (MindStudio)

YouTube predicts 2026 is the year creators break into prestigious film and TV awards. YouTube’s head of creator partnerships says creators are “the new Hollywood,” pointing to Emmy and Golden Globe nominations as evidence that the gates between creator content and traditional entertainment are opening. (The Wrap)

Podcasting is evolving into a multi-format, multi-platform medium. 51% of Americans have now watched a podcast. YouTube users streamed over 700 million hours of video podcasts on TVs in October 2025—nearly double the prior year. Netflix is launching 50–75 original podcasts. The gates between “podcaster” and “media company” are dissolving. (The Wrap / eMarketer)

The Rising Value of Human Trust

Marketers increased AI content spend by 79%—while consumer preference for AI content collapsed from 60% to 26%. Billion Dollar Boy’s survey of 6,000 consumers, creators, and marketers reveals the widening gap between what producers think is working and what audiences actually want. 73% of marketers believe AI content performs better; only 26% of consumers prefer it. Meanwhile, 87% of creators increased their AI output, 52% report career burnout, and 37% are considering leaving the profession. The supply side is accelerating. The demand side is pulling away. That’s the commoditization dynamic in a single data set. (Net Influencer)

Nieman Lab's 210 predictions for 2026 converge on one thesis: loyalty, not scale, wins. Across the most respected prediction series in journalism, the loudest signal is a pivot away from volume-first content toward trust, community, and direct relationships. "The publishers who win in 2026 won't be the ones with the most clicks—they'll be the ones with the most habituated, trusting, and engaged communities." Multiple predictors argue that audiences follow individuals rather than institutions, that "owning a recognizable story world, personality, or aesthetic becomes the new distribution strategy," and that "communities will save journalism—not influencers, not algorithms, not corporate platforms." The factory model is dying. The curation model is forming. (Nieman Lab Predictions 2026)

The Association of National Advertisers named “authenticity” as co-Word of the Year for 2026—alongside “agentic AI.” The industry recognizes the twin forces reshaping content economics: AI production explosion and the human trust response. (Phys.org)

This is Post 2 of The Infinity Era, an eight-part series on how AI, curation, and creator ecosystems are reshaping how the internet discovers and values creative work. The full series publishes on Letters on the Future.

Hi! I’m Monica Leonelle, futurist, storyteller, and context curator exploring how technology and AI are reshaping the future of creative work. I’m a USA TODAY bestselling author, former software engineer, and Chicago Booth MBA who has published 50+ books. I’ve spent the last two decades building, analyzing, and writing about the creator economy from the inside, and my work has been featured in outlets including Forbes, Inc., Newsweek, AdAge, and The New York Times.

• Reach is declining

• Traffic is unpredictable

• Advertising costs are climbing

The systems that used to reliably connect creators with audiences feel...Unstable.

I’ve been building online since the mid-2000s. I’ve watched multiple waves of “everything just changed” sweep through this industry—and usually, it’s a platform update, a market correction, something that settles back to normal if you figure out the new rules fast enough.

So when creators in my circles started saying several years ago that the playbook that defined the 2010-20s wasn’t working anymore, part of me wanted to file it under, “algorithm shifts, adjust and move on.” And honestly, I did just that for a long time, especially because over in the publishing world, I taught direct sales around crowdfunding, website stores, and subscriptions and memberships. My direct sales bubble was thriving, and I thought it was primarily exclusivity to Amazon’s KDP Select program that was causing authors many of their problems.

And that was part of the story, but not the full story. I’ve spent the last several months—if not years—trying to understand what is really happening and why it’s happening now.

What I’ve landed on is a pattern—one that’s been repeating long before the internet existed. Once I saw it, everything I’d been experiencing stopped looking chaotic and started looking almost boringly predictable.

This essay is about the structural pattern underneath the chaos—and why it matters for every creator trying to build something sustainable right now.

We All Agree…Something Is Breaking

Let’s start with what creators are actually experiencing, because the specifics matter.

Organic reach on social platforms has been declining for years—not gradually, but in large lurches:

• A video hits millions one week and 6 months later, identical content dies at two hundred views

• Facebook organic reach for pages dropped from roughly 16% in 2012 to under 2% by the early 2020s

• Amazon’s algorithms swing in ways that destabilize book sales that had been steady for years

• Newsletter deliverability erodes quietly as inboxes get more aggressive about filtering

• Customer acquisition costs across digital advertising have roughly doubled since 2019

In my publishing bubble, I’ve talked to authors—smart, strategic authors with strong backlists—who watched their book royalties drop 30-40% in a single quarter with no clear explanation. But it’s not just Amazon that’s experiencing challenges…It’s every social, retail, and advertising platform too.

As algorithms throttle human-made content, some estimates suggest AI-generated content online has increased by over 1,000% since 2023.

Creators experience all of this as enshittification—a term coined by Cory Doctorow that perfectly captures what it feels like when a platform that once served you starts serving itself at your expense—and despite all your efforts to build the platform in the first place. Declining reach, inconsistent performance, higher acquisition costs, the constant sense that the rules changed overnight with no notice and keep getting further and further from helping you.

Despite the obvious enshittification of the system, most creators do not rage against the machine. Instead, they interpret these symptoms as a personal failure or a puzzle to solve. “I need to post more.” “I need better content.” “I need to figure out the new algorithm’s rules.”

I get the impulse. I’ve had that impulse. For me, it has always been because others are succeeding still—so success must still be possible, I just don’t have the right winning combo of content, timing, and luck.

Of course, this is illogical in some sense—someone is always winning an algorithm because it’s a system that must have winners. What’s changing is not if there are winners or not, but more subtle factors that are harder to track:

• If the winnings are getting better or worse

• If the churn of who wins at any given time is more distributed across creators

• How much it costs to win (and if it even makes economical sense to win)

• The underlying business model or the “why” of winning and how the creator is actually getting paid

The ecosystems around algorithms are actually quite complex—something I learned when doing a deep dive and writing over 1000 pages on retailer algorithms at Amazon, Barnes and Noble, Google Play, Apple, and Kobo. And social and advertising algorithms are experiencing similar complexities as retailer algorithms.

For example, a Youtuber who is winning may be getting paid through advertising revenue and brand sponsorships. But someone using Youtube as a marketing channel may be getting paid through products and services off-platform. Those are different motivations that may give different creators different levers to pull within the algorithms. That complicates the analysis and breaks down the playbooks too.

In truth, when there’s change in the air, the issue is not the content, and the algorithms aren’t failing randomly. They’re optimizing exactly as the platforms designed them to—for ad revenue, for session time, for platform-native content, for engagement loops that keep users inside the walled garden. Those are rational business decisions from the platform’s perspective, they just happen to be terrible for creators.

Hence, enshittification. We know it’s happening, but it’s still not the whole story.

The real question is: why is enshittification happening everywhere, all at once, across every major platform? The answer is something that most people aren’t looking at.

What Is Actually Breaking Across All Algorithms

The real bottleneck in any media ecosystem has never been production, but human attention.

A person has roughly sixteen waking hours, which gives them limited reading and watching capacity, and emotional investment. There is a hard cap on how much culture any individual can consume. That cap hasn’t changed since the invention of the printing press—and it’s not going to change now.

The internet has always had more content than anyone could consume. There was already more content than a person could process in 2005, and the same trend happened in 2010, and 2015, and 2020, and 2025.

Content surplus is not new. We’ve been drowning in content for decades. So what is it then?

What actually breaks the systems is when a technology comes along and makes content production dramatically cheaper and faster, creating a fresh explosion of supply that the current filtering tools weren’t built to handle.

The problem isn’t that we moved from scarcity to abundance. We’ve been in abundance for a long time. The problem is that we keep experiencing new scales of abundance—content explosions so large that the discovery systems managing the previous scale can no longer keep up.

And that is what is happening now with generative AI.

Luckily…There Are Answers in History Around How We Respond

After the printing press exploded book production in Europe, readers couldn’t evaluate everything. New professions emerged to handle the surplus—editors, critics, publishers. These people didn’t create art, but they did create value as filterers and curators. Publishing houses became the discovery layer for literature.

This same pattern has happened across all creator industries:

• The rise of mass-market magazines created editors and editorial brands that decided what was worth reading

• The emergence of radio, and music becoming effectively free to the listener, created an ecosystem of DJs, programmers, and labels that shaped what people heard

• The expansion of film and television created studio executives, distributors, festival programmers, and critics who filtered what reached audiences and what mattered culturally

• The rise of blogging and online media created aggregators, tastemakers, and niche editorial hubs that helped readers navigate the flood of digital writing

• The social media era created influencers, recommendation networks, and algorithm-amplified human tastemakers who could surface what spread through culture

Every time a technology triggered a fresh supply explosion, the existing filtering layer buckled—and new curators emerged to help people navigate the flood.

The internet is not exempt from this pattern. It’s just moving through it faster than any medium before it.

And once you see the pattern, the current moment stops looking chaotic. It starts looking like a very specific phase of a very specific cycle.

The Hidden Pattern Underneath the Chaos

Here’s where it gets interesting—and honestly, this is the part that made me sit back and go, oh.

The internet has always relied on both automated discovery and human curation running in parallel. Even during the peak of algorithmic feeds, people still discovered culture through critics, tastemakers, communities, and word of mouth. The difference between eras isn’t whether humans or machines do the filtering—it’s which system dominates and where power sits.

What shifts is the balance. And that balance is driven by one thing—the relationship between the current scale of content supply and the dominant filtering system’s capacity to manage it.

Here’s how that has played out.

The Directory Era (1990s). The earliest web was small enough to organize by hand. Yahoo Directory, DMOZ, and similar projects had editors manually categorizing websites into browsable lists. This worked because the web was still small—a few hundred thousand sites could be meaningfully sorted by teams of humans. Automated search existed, but human curation dominated. Then millions of websites appeared—a fresh supply explosion—and human-scale curation couldn’t keep up. The balance tipped toward automation.

The Search Era (2000s). Google’s PageRank replaced editorial judgment with mathematical ranking—the more websites link to something, the more important it probably is. Search became the dominant discovery layer for nearly two decades. Human curation didn’t disappear—book reviews, music critics, magazine editors all continued to work—but search dominated. Google search specifically also heavily weighted inbound links—a human curation signal—through a proprietary scoring called Pagerank. All of this made the surplus manageable again, until SEO manipulation, content farms, and sheer web scale overwhelmed mathematical sorting, and the balance began shifting again from individual blogs to social media sites.

The Algorithmic Feed Era (2010s). Instead of users searching for content, platforms like Facebook, Instagram, YouTube, TikTok started allowing creators to publish right there. You didn’t need to run a website anymore—and online branding wasn’t only for the nerdiest among us (tech people, bloggers). Then, social sites started pushing content to users automatically. Discovery became passive as algorithms decided what you saw. This powered the entire Creator Economy boom. Unknown creators could publish something and have a platform’s algorithm carry it to millions. The surplus of content was once again enormous, but algorithmic feeds made it navigable—for a while. Human curation continued underneath—communities, word of mouth, fandom networks—but algorithmic discovery dominated.

The AI Saturation Era (2020s). And this is where we are. AI turns content production into something close to a zero-cost activity. Anyone can now generate articles, videos, images, books, and marketing copy at scale. Generative AI didn’t break algorithmic systems, but instead triggered the next supply explosion—one that stress-tested a system already under strain. Long story short, most of the systems running on algorithms have buckled. The surplus went from overwhelming-but-navigable to overwhelming-and-unnavigable. Discovery systems built for the previous scale of abundance cannot handle this one.

Here’s the important takeaway of this history lesson: none of these systems replaced the previous one. Directories still exist in niche form. Search is still massive. Algorithmic feeds still run social media.

Instead, each era became infrastructure underneath the next. What changed in all cases was which layer held the most power over how people found things.

And further—every time the dominant filtering system strained under a new scale of supply, the same thing happened—trust-based human curation became more valuable again. The automated systems that had been handling the volume were no longer keeping up, and people needed human judgment to help them navigate the new flood.

That’s the pattern:

• Step 1: New production technology triggers a supply explosion; meanwhile, attention stays fixed

• Step 2: Existing discovery and surfacing filters buckle

• Step 3: The filtering layer reorganizes—and trust-based, human curation regains importance

• Step 4: Automation tools catch up on what human signals matter the most

• Step 5: People come to the tool because it’s genuinely useful; human curation wanes

• Step 6: Money flows to the tool; power users come, then celebrities come, then scammers come, then the general population comes

• Step 7: The tool reaches a tipping point of market share, becoming dominant

• Step 8: The tool can’t grow monetization through new users, so it moves to charging creators

• Step 9: Enshittification is declared; the alt-tools flood the market, trying to chip away at market share

• Step 10: The cycle begins again at Step 1

We’ve seen it many times before, and now we’re inside it again.

Why This Moment Feels Different

If this is a repeating cycle, why does this moment feel like more than a normal platform shift?

Because the scale of the supply explosion is unprecedented, and the rate of change itself is destabilizing for Big Tech.

Every previous transition increased content supply by a significant amount. The printing press, the web, social platforms—each one made production meaningfully cheaper and faster…But all of this was happening inside format silos. What was consolidating in film didn’t immediately affect what was consolidating in books, and vice versa.

Previous supply shocks were format-specific:

• The printing press exploded book production

• Radio exploded audio distribution

• The web exploded text publishing

• Social platforms exploded short-form video and image production

Each one overwhelmed its own filtering layer.

In contrast, generative AI is collapsing the cost of content production toward zero across every format simultaneously. Text, image, audio, video, code—all at once, and very quickly—with new collapses month to month. That has never happened before.

That’s why creators across every platform and every format are feeling the instability simultaneously. A single production shock (generative AI) is straining every filtering system at once.

Creators Are the Canary in the Coal Mine

If you’re a creator reading this, you already know something I want to name explicitly: we have been the earliest signal of every one of these shifts.

We were the first to notice SEO degrading. The first to see Facebook reach collapse. The first to feel Amazon’s algorithm swings. The first to watch newsletter deliverability erode.

We detect these shifts first because our income sits directly on top of discovery systems. When the ground shifts, we feel it in our revenue before anyone else sees it in their data.

I felt it. I’ve been watching my own content’s discoverability erode for the last couple of years and asking myself whether I was doing something wrong or whether the ground itself had moved. The answer—which took me longer than I’d like to admit—is that the ground moved. And it’s still moving.

Institutions—traditional publishers, record labels, Hollywood studios—can be less affected at first because they rely on different discovery systems—brand recognition, media relationships, retail distribution, cultural gatekeeping. Those legacy systems are more resilient than algorithms, which usually buys the institutions time. But they’re not immune—Hollywood is fighting over AI, publishing is struggling with copyright law, and the music industry is grappling with AI-generated content flooding streaming platforms.

As the Curation Layer Reorganizes, Who Holds the Power?

So the filtering layer is reorganizing. What does the new mix look like?

I’d rather start with what it’s not. The future is not “humans replace algorithms” or “AI replaces human taste.” It’s a rebalancing of roles. Machine systems and human curation have always coexisted—the question is which one dominates and where power sits. What’s shifting is the balance.

Two developments are converging simultaneously.

The first is trust-based curation regaining importance at scale. People are increasingly relying on trusted creators, communities, newsletters, fandom networks, events, and collaborative ecosystems to decide what’s worth their attention. Instead of trusting an algorithm to surface the right content, they’re trusting people—curators with taste, communities with shared values, creators they’ve built relationships with over time.

You can already see this everywhere. Curated newsletters are booming. BookTok recommendation networks drive more book sales than most publisher marketing departments. Fandom communities on Discord and Reddit organize entire genres. In-person events and festivals are becoming discovery hubs.

This isn’t new—it just got buried under automation for a while. Now that automation is straining under the latest supply explosion, trust-based curation is resurfacing as a primary filtering mechanism—this time with internet-scale tools and reach.

If you’ve been building community, building fandom, building real relationships with your audience—you’ve been building filtering and curation infrastructure for the next era, whether you knew it or not.

The second is AI agentic discovery. Instead of you browsing, scrolling, and comparing, an AI agent will likely soon be doing it for you. AI shopping assistants, AI research tools, AI conversational search—these are early versions of a world where a bot handles your navigation through the surplus. Big Tech is investing heavily here. Google, Amazon, OpenAI, and others are all building toward a future where AI agents are the primary way people move through the internet’s content surplus.

Both are real. Both are already happening. And the most likely future involves both—working together in ways most people haven’t thought through yet.

And it’s not all that different than what we have had. Algorithms have always run on rudimentary signals from humans—viewing, linking, liking, sales, engagement—that are supposed to indicate meaning. And humans have long used machine- or industry-generated filters to curate meaning too.

While AI agents are excellent at navigating large volumes of content and filtering for relevance and efficiency…

…They struggle with taste, identity, cultural meaning, and the emotional resonance that makes someone fall in love with a story or trust a creator. Those things require human judgment operating inside a relationship of trust.

Machines handle the scale. Humans provide the meaning.

But the question of who builds and who controls the curation layers matters enormously. If that layer is entirely owned by the same Big Tech companies that built the algorithmic systems currently straining under their own weight...We’ve just moved the same power dynamics to a new platform.

We currently have a window in which we—individually and/or collectively—can take back some of the power around this dynamic.

As Creators, it’s our time to unenshittify things.

There is One More Layer to This Which Indicates Opportunity

Consumers will consume based on trust—and the largest sentiment out there is people are tired. Trust in journalism and the media has eroded, trust in politicians has definitely eroded. The issue platforms are having is not just that AI is flooding their algorithms and gaining traffic, but that there’s so much of it that is just not good. When a platform delivers even 20% of bad experience and doesn’t have enough value to force people to stick around, people will lose the habit of the platform.

You can likely intuitively feel this too—regular YouTube watchers ditching it because 20% of the videos are low-quality AI content. Readers leaving Kindle Unlimited because of too many bad reading experiences. Scrollers abandoning TikTok because it’s the same tired AI-generated videos again and again.

On a personal level, I’ve pulled back from YouTube because it’s so hard to find creators I am interested in. The same has happened with TikTok—my oldest child and I were using it for a few minutes, a few nights a week to practice identifying AI together, as I thought it was an important skill to discuss with him. After very little time, it was showing us the same few types of videos that had no value to his growing brain, and eventually there was not even interesting AI education to be had.

We have also pulled back on our children watching YouTube for any reason—we are more likely to direct them to a show on Netflix or Disney where we understand the branding and can keep them from watching low quality AI content.

Once those habits are out, it’s hard to reestablish them—especially when the humans don’t really miss the habit + know the habit is not great from them anyway.

We are on platforms almost entirely because of the dopamine hit—so if those platforms can’t give it to us reliably, it’s much easier to check out.

The 10-step process that has dominated how the internet does discoverability may also be breaking down. Because platforms need to become dominant, and domination requires consistent good dopamine hits, and consistent good dopamine hits requires good discoverability and effectively keeping the scammers out.

AI is making life difficult for large platforms on both ends.

Fragmentation is normal in competitive systems, and for creators, the last big fragmentation to tear down major gatekeeping was around institutions. Record labels, radio stations, publishing houses, cable channels, and film studios survived, but they never pieced back together their empires. Instead, Big Tech companies Google, Amazon, Apple, Facebook, and Spotify carved out their own pieces of the market and kept them.

I believe that the rapidity of AI could keep systems destabilized for a while, if not permanently, which means there’s more and longer opportunity for trusted curation to emerge.

What This Means for Creators

Here’s the part that changes what creators should be building.

In the algorithmic era, creators were content producers operating inside someone else’s discovery system. You made the content, but the platform decided who saw it. For a while, that bargain worked and was helping creators rather than squeezing them.

In the emerging era, creators who build trust networks, communities, events, ecosystems, and recommendation loops are no longer just making content inside discovery systems. They’re becoming part of the filtering infrastructure itself.

That’s not a small shift. That’s a fundamental restructuring of where power sits in the Creator Economy. In the algorithmic era, platforms controlled discovery and creators were dependent on their systems. In the emerging era, creators who build their own discovery ecosystems—audiences, communities, events, fandom, collaborative networks—become the discovery layer themselves.

And that...Is a big deal. Because the more content exists, the more valuable trust-based filtering becomes. Creators have been building that trust for years, often without realizing it was going to become the most valuable thing they own.

I don’t say that lightly, because this shift has cost me and a lot of creators I care about real money and real momentum. But here’s what I actually find exciting—what comes next may actually favor creators more than what we’re leaving behind, because our structural position is stronger.

The rest of this series explores what we while discoverability is destabilizing…Because there’s so much we can do to reclaim our space and prepare for the future of the Creator Economy.

Signals I’m Watching

These are real-world developments that reinforce the patterns I’ve described. I’ll be tracking signals like these throughout the series.

Organic Reach is Collapsing

Instagram’s organic reach fell to just 4.0% in 2024—an 18% decline year over year—and dropped further to 2–3% for publishers by mid-2025. If you’ve built an audience of 50,000 followers through years of consistent work, fewer than 1,500 of them see any given post. The algorithm has decided the other 48,500 people who chose to follow you don’t get to see your content. (Blog Herald)

Facebook organic reach has fallen from 16% in 2012 to 1–2% in 2025, Instagram’s reach rate dropped 12% to 3.50%, and LinkedIn saw an even more dramatic 34% slide. The decline is happening everywhere, not just on one platform—the structural shift away from organic distribution is industry-wide. (Hootsuite)

A 2026 survey of 1,000 U.S. creators found that 46.2% of Instagram creators, 76% of TikTok creators, and 59.1% of long-form YouTube creators receive fewer than 1,000 views per post. The vast majority of creators are struggling for visibility across every platform, confirming that building meaningful reach remains the primary challenge. (Influencer Marketing Factory)

Platforms have quietly shifted from distribution engines to discovery engines—prioritizing AI-driven content matching over follower relationships. Up to 50% of the content users now see in their Facebook feeds comes from “unconnected sources,” accounts they don’t follow. The implicit bargain of “build an audience, and the platform helps you reach them” has been quietly voided. (DMNews)

Customer Acquisition Costs are Climbing

Customer acquisition costs have surged 222% over eight years, with the acceleration particularly pronounced in digital channels. The average financial loss per acquired customer jumped from $9 in 2013 to a projected $29 by 2025. Channel saturation, privacy regulations, and diminishing marginal returns are all compounding. (SimplicityDX)

Between 2023 and 2025, customer acquisition costs jumped 40–60%, driven by higher competition, privacy rules, and attribution challenges. Privacy changes like iOS updates and GDPR have made targeting less precise, driving up costs even as platforms improve. (Phoenix Strategy Group)

Growing customer acquisition costs contributed to a 10-point decline in the importance of new customer acquisition as a marketing priority. The IAB’s 2026 Outlook Study found that while 54% of marketers still cite acquisition as a top goal, that figure has dropped significantly—and the focus on driving repeat purchases has nearly doubled since 2024, pointing to challenges in bringing new business in. (IAB)

AI-Generated Content is Flooding the Internet

Europol warned that as much as 90% of online content may be synthetically generated by 2026. This isn’t just text—it includes images, video, audio, and deepfakes across every format simultaneously. (Europol)

Meltwater reports a ninefold increase in mentions of “AI slop” in 2025 compared to 2024, and research by Kapwing estimates that 21–33% of YouTube’s feed may consist of AI slop or “brainrot” videos. “AI slop” was named Word of the Year 2025 by both Merriam-Webster and the Australian National Dictionary. (Future Center UAE)

Content grounded in lived experience, tested knowledge, and real-world work is being buried under an AI flood across Facebook, Pinterest, and beyond. As AI begins training on its own output, distortions will compound, creating an internet that feels more synthetic by the day. Nieman Journalism Lab predicts that in 2026, AI will outwrite humans. (Nieman Journalism Lab)

Trust-Based Curation Is Resurfacing Through Newsletters

The world's largest newsletters now operate like media companies—with multiple publications exceeding 1 million subscribers. 1440, The Rundown, TLDR, Morning Brew, The Hustle, Nice News, MarketBeat, and others have built massive audiences through curation-first models. The common thread: they offer curated value that feels uniquely human in a landscape increasingly saturated with AI-generated content. (Paved)

Readers increasingly prefer newsletters from individuals rather than faceless brands, and personality-led content is the key differentiator from AI-generated competition. Newsletters need to hook people on their unique perspective and style—giving readers something they can't get anywhere else. As HubSpot data shows, 64% of newsletter professionals believe newsletters will be mostly AI-generated by 2030, which paradoxically makes the human-curated ones more valuable, not less. (HubSpot)

Over 17,000 writers now get paid on Substack, and the platform added more than 1 million paid subscriptions between November 2024 and March 2025 alone. Notably, roughly 25% of paid conversions now come from Substack's own internal recommendation and feed features—meaning the platform is building its own trust-based discovery layer. (Backlinko)

Beehiiv housed more than 50,000 newsletters as of 2024, nearly doubling from the year before. The growth isn't just in volume—it's in sophistication. Beehiiv's November 2025 "Winter Release" expanded its vision significantly, adding AI-powered website building, digital product sales, podcast hosting, and advanced analytics—positioning it as a complete creator operating system rather than just an email tool. (IEM Labs)

Creators on Patreon have surpassed $10 billion in cumulative payouts and over 25 million paid memberships, with annual creator earnings now exceeding $2 billion. More than 250,000 active creators monetize on the platform, up 15% from 2023. Patreon has been boosted specifically by the increasing value of smaller creators and the growth of private communities. (Contrary Research)

Community-Driven Curation Is Reforming Alongside Traditional Algorithmic Discovery

More than 50 million books recommended by BookTok were sold across Europe in 2025, generating €800 million in revenue. In Germany alone, 28 million BookTok-recommended books were sold—more than double the 12 million sold in 2023. TikTok announced the expansion of its BookTok Bestseller List into the UK, Italy, and Spain. (TikTok Newsroom)

In 2024, BookTok-driven demand helped generate 59 million print book sales in the U.S., contributing to an estimated $760+ million in revenue tied to TikTok-discovered titles. What began as a grassroots reader movement has evolved into one of the most powerful commercial engines in publishing—and it hasn’t slowed heading into 2026. (WriteStats)

By 2026, BookTok creators have become essential marketing partners for every major publishing house. Publishers now acquire books based on their “TikTokability”—they want plots that creators can summarize in a 10-second emotional hook. Bookstores feature “As Seen on BookTok” sections that have become some of the highest-performing areas in stores. (Rolling Stone)

AI Agentic Discovery is Emerging

Agentic commerce—where AI agents autonomously discover, compare, and purchase products on behalf of consumers—is emerging as one of the defining commerce trends of 2026. Amazon, Google, OpenAI, and Meta have all launched AI shopping tools in the past year. AI platforms are expected to account for $20.57 billion in U.S. retail ecommerce spending in 2026, nearly quadrupling 2025 figures. (eMarketer)

Google launched its Universal Commerce Protocol (UCP) in January 2026, partnered with Walmart, Target, Shopify, Etsy, and 20+ others. Brands now face a three-ecosystem world: Amazon’s proprietary agents, Google’s UCP, and OpenAI’s approach. McKinsey forecasts $3–5 trillion globally in agentic commerce by 2030. (CNBC)

OpenAI pivoted away from its Instant Checkout feature after realizing users researched products in ChatGPT in droves but weren’t completing purchases there. Walmart’s data showed ChatGPT’s integrated checkout performed three times worse than having users complete purchases on the retailer’s own site—highlighting that this space is still being figured out. (CNBC)

The Power Reordering Between Platforms Institutions, and Independent Creators is Building

The number of AI copyright infringement cases filed against AI companies more than doubled in 2025, from around 30 to over 70. The biggest development was the $1.5 billion settlement in Bartz v. Anthropic—and settlements and partnerships were the dominant trend, with more expected to multiply in 2026. (Copyright Alliance)

Warner Music Group settled with both Suno and Udio in November 2025, and Universal Music Group settled with Udio in October 2025. These deals established licensing partnerships where new AI models would be trained on authorized catalogs with artist opt-in provisions—shifting from pure litigation to commercial partnership. (Copyright Alliance)

Disney and Universal filed a first-of-its-kind copyright suit against Midjourney, while Pulitzer Prize-winning author John Carreyrou led a December 2025 lawsuit against six major AI companies. Lawsuits and licensing deals surged in 2025 as authors, journalists, record labels, and Hollywood studios fought over unauthorized use of intellectual property for model training. (Caixin Global)

Consumers Shift Toward Trust and Authenticity

The UK government scrapped plans in March 2026 that would have allowed AI companies to train on copyrighted music without permission. Over 10,000 submissions flooded the consultation, with only 3% supporting the AI-friendly approach—a massive backlash from the creative industries. (Silverman Sound)

Spotify has removed over 75 million tracks it classified as “spammy,” many of which were low-effort AI-generated content. Apple Music launched “Transparency Tags” in March 2026, and Deezer uses proprietary detection technology to automatically tag AI-generated music. The music industry is grappling with AI flooding streaming platforms across every major service. (Jam.com)

Consumer preference for AI-generated content has dropped to 26%, down from 60% three years ago. The Sprout Social Q4 2025 Pulse Survey reinforces that consumers prioritize human-created content in 2026, assigning new value to authenticity as a brand differentiator. (Sprout Social)

While 77% of marketers believe AI effectively crafts emotionally resonant content, only 33% of consumers agree—a 44-percentage-point disconnect. Research also shows 52% of consumers reduce engagement with content they believe is AI-generated, even before confirmation. (NetInfluencer)

The Association of National Advertisers chose “authenticity” and “agentic AI” as its dual Words of the Year for 2026. Marketing professor Colleen Kirk notes that “consumers are becoming ever more skeptical of the human origin of advertisements and marketing messages” while emphasizing that “authenticity is always best.” (Phys.org)

This is Post 1 of The Infinity Era, an eight-part series on how AI, curation, and creator ecosystems are reshaping how the internet discovers and values creative work. The full series publishes on Letters on the Future.

Hi! I’m Monica Leonelle, futurist, storyteller, and context curator exploring how technology and AI are reshaping the future of creative work. I’m a USA TODAY bestselling author, former software engineer, and Chicago Booth MBA who has published 50+ books. I’ve spent the last two decades building, analyzing, and writing about the creator economy from the inside, and my work has been featured in outlets including Forbes, Inc., Newsweek, AdAge, and The New York Times.