AI Slop: An Annoying Short-Term Garbage Heap Trend or the 2nd Coming of Napster?

The internet is drowning in AI-generated content—endless “slop” of dubious quality flooding social feeds. Many dismiss it as a fleeting nuisance: low-effort garbage that will burn out once the novelty fades, if only we can manage to wrangle these AI robber barons and their dirty data centers.

But we won’t wrangle them, since AI is not built primarily to deliver slop. Slop is an off-shoot byproduct, an afterthought for both the barons and the politicians. We can’t stop the data centers, and we shouldn’t. 100s of under-construction data centers are the new table-stakes to feed the brains of millions of drones in 2035.

If we fail to build them, any of several other countries can easily invade us overnight as Ukraine has proven in early examples against Russia. Our best legacy military infrastructure will not stand a chance. It is our responsibility as one of the most prominent democracies on the planet to preserve free and fair elections by being too powerful to attack. Technology and defense have always gone hand in hand. Many people see it as nearly impossible for a foreign force to invade, but a few living relatives still remember Pearl Harbor. To believe it’s unlikely is extremely naive if rivals have millions of AI-driven drones and we don’t.

Furthermore, if we outlaw tech moguls building data centers here, they will migrate into the hands of rivals. It’s much better to keep them close so that we have greater influence over them.

Assume the data centers continue to expand exponentially. Assume regulation prevents them from 10X-ing your power bill. What does this do to AI slop? Does the slop go away? Or does it get better and better until you can’t tell it’s slop anymore in most cases?

Odds are near 100% that slop gets A LOT better. Does the slop get good enough to insert a clip here or there in an edit of a Hollywood show? Today’s slop is already scratching that surface. You may think you can tell which clips are AI because of their low quality, but I have to ask: why are you sure the good clips AREN’T AI? Some of them are.

What if the current AI slop wave is just the opening act of something far more disruptive? What if slightly better generative AI, paired with the resilient infrastructure of peer-to-peer piracy, becomes Napster 2.0—not just for music, but for every form of entertainment: film, television, and gaming?

The “GenAI for Piracy” trend is embryonic. Today’s tools produce mistake-riddled, uncanny, incoherent clips that can rarely pass a Turing test beyond 10 seconds without breaking character or physics. Yet the trajectory is unmistakable. Models are improving at breakneck speed. When they cross the threshold into coherent clip generation, the floodgates will open.

The Coming Revenue Decimation

Streaming giants and digital game distributors have built empires on controlled, subscription-based access to copyrighted content. Netflix, Disney+, Paramount+, Apple TV, Amazone Prime, Steam, Epic Games Store, PlayStation Store, Nintendo eShop, App Store, and Google Play derive massive value from scarcity and centralized distribution. That scarcity is about to evaporate.

Imagine a new Star Wars series drops on Disney+. Within 24 hours, pirate networks host thousands of AI-remixed variations: episodes where dialogue is rewritten to remove “woke” elements or lean harder into them, characters recast with different actors or entirely new interpretations, color grading adjusted to fan preferences, VFX sequences expanded or reimagined, plot branches altered to match viewer fantasies. Early adopters—skilled prompt engineers and tinkerers—stamp their creative mark on the material with hacker call-signs to avoid prosecution. Word spreads on forums and Discord: “Skip the official version; the @ZeroCool cut is way better.” This trend will have legs as the new counterculture, similar to the early days of EDM where everything was on the table for remix. If you’re a hot DJ in that new culture, do you give a damn about copyright?

These remixes won’t come from centralized pirate sites vulnerable to takedowns. They’ll leverage open-source GenAI tools deeply integrated with mature BitTorrent-style P2P networks—the same decentralized architecture that made Napster-era music sharing nearly impossible to eradicate at scale. Prosecuting millions of individual nodes is a fool’s errand, as the music industry learned the hard way.

The result? A tsunami of subscription attrition. Why pay monthly for a dozen services when a free, infinitely customizable version of every show and game exists on the open web? Audiences have long chafed at one-size-fits-all content. They crave stories tailored to their tastes, politics, humor, and wish-fulfillment. GenAI + piracy delivers that at zero marginal cost.

The gaming sector will be a bit more resilient than streaming footage just because GenAI cannot one-shot whole gameloops as easily as replacing film scenes, but gaming will not be immune. Steam already has AI attribution requirements to fight slop, which are probably going to prove useless since they penalize only the honest hard-working devs while lazy slop devs are dishonest anyway. Slop devs are perfectly willing to lie about attribution, so they’ll rarely be caught.

“But we can tell which games are slop just by looking at them.”

No you can’t. Especially not by 2030. GenAI will soon pass Turing tests for 30-second clips. Do you really think it’ll fail to trick Steam mods RE: 3D assets? Do you think it’ll fail to trick you?

More importantly, if you can’t tell anymore, will you really care? Some of you will think, “of course, I care… I want the human artist to get credit for their hard work.”

But what if every human artist is also using 50 AI tools in their pipeline for each asset in each game? Audiences will become increasingly nihilistic about preserving the copyright of their favorite human artists when they realize they never had a chance of discerning which content is “AI-enhanced.” It will become safe to assume it’s all made using AI. The AI tools are, after all, just tools wielded mostly by hacky slop devs by majority today, but that doesn’t mean the tools are bad in the hands of a true artist. Quite the opposite. Eventually, when the audience seems less ready to hang any artist who admits to using AI, every artist will admit to using AI all the time.

The Physical Retreat: Lessons from Music’s Napster Era

The music industry’s response to Napster was sluggish and litigious. Record labels fought tooth and nail to protect CD sales and early digital downloads, largely losing the battle for internet distribution. Metallica wailed against pirates the same way MoistCritical is currently wailing against AI moguls. If you’re wondering why Youtubers might feel they have skin in the game, you can bet every big Youtuber has already issued takedowns to several pirates who deepfake their faces on scam channels, but today they’ll never admit to you that some of them have investigated whether to use an AI avatar of themselves to make faster content on their own channels. Pushing against AI is the Metallica position. Using AI openly would be the RadioHead position. But many artists and creators are pitter-pattering their toes in AI secretly to avoid the ire of the audience. They’re far less bold than RadioHead was. Eventually, a creator is going to come out and say, “I use AI for everything, all the time, and it’s awesome.” And they’ll get lots of love and hate, and then more artists will follow suit, and that will become the norm.

But doesn’t piracy destroy their revenue? Why would they willingly participate in their own demise?

To understand, ask what saved the artists in the Napster Era? They certainly didn’t save their record sales at all. They were saved by live performance. You can’t pirate the real world, only the digital world. Live Nation and the broader concert ecosystem exploded as album revenues collapsed. Touring, merchandise, and experiential revenue became the lifeblood for many artists and labels. Merch pirates and scalpers did appear, but those are far easier to prosecute at scale if they ever dare to scale. New social media channels fed the indie scene simultaneoussly, giving a much needed avenue for new stars to rise on their own while labels circled the wagons and stopped inking risky deals with anyone who wasn’t already proven.

Streaming and gaming face a parallel inflection. Hollywood/theatre revenues were decimated by COVID, but that’s all pre-Piracy-2.0. The AAA games industry is on a layoff spree, crumbling under its own weight. In the 2030s, expect major IP holders—Disney, Netflix, Warner, Ubisoft, EA, etc.—to retreat from pure internet distribution if they’re smart. If they learned from the mistakes of the record labels, they will take large swathes of streaming earnings and jam it into phsycial attraction R&D ASAP. They’ll double down on experiential channels where piracy cannot follow:

• Theme parks and expansions
• Mini-theme parks and mall activations
• Arcades and immersive experiences
• Pop-up events, content-themed carnivals, and branded retail venues

The Piracy Hydra is coming. The escape hatch is clear: revitalization of dying malls into local entertainment hubs. This path to new revenue is extremely cap-ex-heavy relative to digital distro, which is why it’s most compelling as a shared endeavor. Converting a single mall into a high-quality mini-theme park is prohibitively expensive for one company. A consortium of streaming and gaming giants sharing infrastructure costs could create viable large destinations. Netflix House is an early, tentative step in this direction—branded experiential retail and entertainment spaces. Scale that concept aggressively with strength in numbers, and sound the horn to local disparate entertainment venues (concert halls, arcades, escape rooms, go kart tracks, etc.). Rally the entertainment superfriends, and a reskinned mall becomes a feasible entertainment bulwark. Wash, rinse, repeat in every suburb and you have a Rennaissance for the Mall Era of the 80s with more emphasis on entertainment first, commerce second.

Digital pirates can’t steal a theme park ticket, an immersive VR arcade experience, or the atmosphere of a live event. Just as Live Nation thrived amid falling record sales, physical activation networks could preserve value from legacy IP catalogs even as digital revenues hit the chopping block.

The Audience Always Wanted This

No consumer ever demanded ten competing streaming services with fragmented libraries and lowest-common-denominator algorithmic slop. What people wanted was easy access to their favorite universes, the power to remix, extend, and personalize them—true fan fiction at industrial scale. Or at least, they want their favorite online influencers to have the ability to remix on their behalf.

Content companies that treat this desire as illegal will be crushed. The new piracy tsunami will be faster, more creative, and more personalized than anything from previous eras.

There is light at the end of the tunnel, but it requires humility and adaptation:

1. Sanctioned sandboxing. Offer official tools for users to remix, mod, and customize content within controlled environments. Allow GenAI mods with restrictions on over-the-top violence or NC-17 content. The pirates will still have a heyday with the grossest imaginable variations, but a majority of customers don’t want to swim in that cesspool to find “good” content. Yes, sanctioned IP sandboxes will cannibalize some revenue, but it will hurt less than unchecked piracy. Platforms that empower fans to create “better” versions on their own terms will retain loyalty and slow attrition. They’ll offer robust modding tools for films, shows, and games. The S-tier stewards of legacy IP catalogs will emerge from this primordial ooze when they realize their new A-list writers/directors/VFX supervisors are the modders who please the most audience members with this new fan fic era before moving on to author original content.
2. Physical-first strategy for legacy catalogs. Channel resources into experiential distribution that cannot be digitized and remixed into oblivion. Partner across companies to share the burden. The music industry would have preserved far more value if it had aggressively invested in concert infrastructure and venues during the Napster years. Hollywood and gaming should not repeat that mistake by suing GenAI fan-fic authors, hunting dark-web P2P hosts, and pretending streaming networks will last forever in their current form.

AI slop may look like garbage today. But tomorrow it could be the democratizing force that finally gives audiences what they’ve always craved: ownership over the stories they love. The companies that fight this reality with lawyers will fade. Those that fight it with innovation will rise by creating sanctioned sandboxes and bespoke physical experiences.

The Copyright Apocalypse isn’t the end of content industries. It’s a forcing function toward better business models. The question is whether Hollywood and the games industry will recognize it before the tsunami hits.