“World War III will be a guerrilla information war, with no divisions between military and civilian participation.” — Marshall McLuhan, 1970

Ali Baba and the forty bots. A little earlier this month, as Donald Trump turned on Congressman Thomas Massie for reasons I hadn’t been following and — five or six major political controversies into the week — no longer had the energy to care about, Gateway Pundit’s Elijah Schaffer posted: “I would like to confirm that there were financial offers in the past to attack @RepThomasMassie here on X. Some took the money. I got offered the money as well.”

Almost nobody cared.

The problem is it wasn’t drugs or sex or gross corruption. It wasn’t a CIA plot, or a major foreign information op. It wasn’t even a trolling offer to annex Canada. So it barely cracked headlines. But despite the story’s failure to capture the public’s imagination, it was and remains a total scandal, and an important sign of what’s to come for our global information ecosystem.

Elijah didn’t only claim influencers were hired to attack Massie, he implied many of them did so without disclosing the fact that they were paid. Were this only an isolated incident, it would be a problem. But now it’s a trend. Back in November, Pirate Wires discovered the prediction market Kalshi was paying influencers to target rival Polymarket’s CEO Shayne Coplan following an FBI raid on Shayne’s home. A month before that, the Washington Post hinted similar behavior was likely shaping our election. At this point, my sense is most of us are not only encountering paid political propaganda almost every day, but engaging with it.

How many of your favorite crypto Pepes screaming for a CEO’s imprisonment are taking a salary for their opinion? That knitting influencer oddly obsessed with accountability for a Kentucky libertarian in Congress — when and why did she decide she cared about the issue? How many of your favorite fitness instathots currently mad about the president’s immigration policy are collecting a 1099 from a DNC operative? The internet opened up a cartoon hellmouth, and these are only a few of the colorful, laughing faces that now define the seething, 24/7 information war we live inside. They’re also human, which I increasingly believe makes them a minority. Lately, I’ve been wondering —

How much of our internet is real?

Back in 2019, from the bowels of 4chan, the Dead Internet Theory began to gain prominence — a “conspiracy theory,” I learned this week, or at least according to Wikipedia editors. Basically, the notion here is most online activity is driven by bots, AI, and corporate content rather than real human users. We wander about our days online, the theory goes, totally blind to the fact that most “people” we engage with are fake. Interactions are rarely organic, and there are almost no real organic social movements whatsoever. Our entire internet reality is controlled by a mysterious, ever present “them.” As constructed here, the theory is a little much for me. But if you remove the notion some specific person is running the internet’s grand fakery to some specific end, the overall thrust of the argument is reasonable.

Are most of our online interactions fake? Probably not. But are a non-trivially large number of our interactions fake? Are social movements guided algorithmically, sometimes by accident and sometimes on purpose? Are small groups of highly dedicated individuals gaming our platforms to fake a sense of social consensus on what would otherwise be a niche issue? How many media cycles are kickstarted by an entity paying a group of paid influencers? Are foreign governments participating? Is our own? And is artificial intelligence accelerating all of these trends? Come on. These are not even hypotheticals. We know this is a problem.

Just in terms of recent foreign ops alone, we could list stories all day. In 2022, Meta shut down massive campaigns run out of Russia and China. In 2024, we saw a wide range of such attempts from China, spanning the evocative “Spamouflage” campaign to a whole ass bot army targeting China-critical Republicans running for office. This in turn spurred action from Texas Governor Greg Abbot. But that was only a bot in the bucket. At the same time, both Russia and Iran were running similar campaigns — a long journey from the Internet Research Agency, Russia’s infamous “troll factory.” Today, foreign agents are employing artificial intelligence in the information war, and an endless mass of fake accounts increasingly difficult to tell from the average normie poster.

But have you ever seen a Chinese poster talk about Taiwan? These are innately cringe people, and they aren’t adept at navigating our culture. While much is often made of their impact, my sense is foreign bots are most successful at amplifying native content generated by earnestly stupid (or evil) American influencers.

Small groups of highly-dedicated Westerners working less sophisticated campaigns are probably much more successful, for example on Reddit, where Pirate Wires just uncovered a network of leftists gaming the platform’s content distribution mechanics and laundering terrorist propaganda. We already know political operatives are paying charismatic Western influencers to generate fake grass roots movements. How many of them are coordinating in private chats to boost content with engagement brigading? How many of them are savvy enough to amplify their impact 1000x with bots?

As we circle the question, it’s worth speculating on the size of X’s bot population. When Elon Musk took Twitter private, he publicly doubted the company’s estimate that fewer than 5% of accounts were fake, and commissioned several studies to crack the real number. All of them returned figures higher than 5%, with one reporting an astonishing 13.7% of Twitter accounts were bots. That was three years ago this April. Think of how far artificial intelligence has come in that time. The internet, once shaped by people talking, is now largely governed by a host of private algorithms, the nature of which we know little about. They can’t tell the difference between a human and an even decently-constructed AI. Increasingly, neither can we.

Consider the recent flood of AI-generated content across our social platforms, from evocative, shockingly constructed homes to cozy-maxxed cabin interiors and totally invented species of cuddly animal (really, an entire genre of baby lambs and pigs and puppies slightly tweaked into surreally adorable creatures). Dumb, worthless stuff. Sticky stuff. This is what we call slop: a kind of content that only exists to get your attention, preys on your assumption of reality to grab it, and gets better every day at convincing you it’s real.

While typically understood to be AI-generated, there have been versions of slop for as long as people have been willing to pay for attention on the internet. Back when the last generation of “new media” companies were coming of age, Clickhole was a commentary on the form — which it was also paid to produce. And there has been slop ever since. There has been slop in such quantities entire careers have come up in the form. There are slop influencers, in fact, who don’t even realize they have now themselves become a kind of sentient slop.

In 2017, did anyone really believe it was transphobic for a man to not want to date a “woman with a penis?” Did anyone really believe “everything is racist” (not hyperbole, please take a moment to google literally any word + the word “racist,” and behold what we once were)? Or was the purpose of such content simply to make a lot of people angry, and to drive a lot of attention to the offending content? In any case, regardless of intention, the slop began to shape society, and certainly that first wave of anti-woke YouTubers (“antis,” they briefly called themselves) seemed earnest enough in their initial critique — and outrage. But how quickly did anti-wokeness itself become a kind of slop?

In an information landscape governed by algorithms that prize a creator’s ability to grab attention over everything else, what is really the difference between an AI-generated birthday cake with teeth and LibsOfTikTok? If there was ever any value in the first viral clip of a mentally ill girl talking about her frog/frogs pronouns, do any of us believe there’s still value in the fifth? The fifteenth? The fiftieth? How long will this game of woke whack-a-mole continue?

Slop behavior.

Clearly, the real danger inherent of slop is not that our brain-dead neighbor will believe something crazy. In any given struggle, we tend to become what we fight. And in a cartoon fight, governed by cartoon physics, the only way to survive is to become a cartoon. We fought the slop, and the slop became us. Or, some of us. But the slop isn’t going anywhere, and the question now is how to learn about the world online without becoming a conduit for slop. A slop creator. A slop enjoyer. Is it even possible? It would be swell if one of our platforms were focused on this problem, rather than working in the opposite direction.

In the early days of Reddit, the company deployed an army of fake, founder-controlled users to bring the place to life. Today, companies like Facebook and TikTok are actually assisting users as they set about creating artificial influencers. I spoke with Jack Dorsey right after he left Bluesky’s board, and he made the case for an algorithm store. The algorithms that sift through content for us exist because the average person online is overwhelmed by the amount of information available, and doesn’t know how to find the content they actually want. Tools that help us navigate this stuff will always exist. But what if you had more control over the tool? Yes, you will be manipulated. But you will choose how.

It’s a nice idea. It’s also not clear how it will benefit the companies deploying these tools, and as Jack is to date the only social media founder I’ve ever seen attempt something important that could theoretically hurt his bottom line — the creation of Bluesky, which was initially launched as a solution, if you can believe it, to the censorship regime that sprung up at Twitter — I don’t see it happening.

Today, our internet is a fragmented media ecosystem occupied by relatively honest information brokers totally at odds with one other, battling openly for the “truth,” while information ops are successfully run by every American rival, native terrorist, and probably our government — with multiple actors within our government potentially fighting one another. Content is slopified for maximum spread. AI is employed to run bots by the thousands, amplifying slopified content. Worse, the algorithms that govern AI also govern our accounts, which means any human sufficiently engaged online is gradually trained to behave like slop — is slowly transformed into slop. The result is our present, miserable chaos, a daily slop war, to which we are also addicted.

So what do we do? How do we live and learn (and love) in the Slop World?

What you’re really asking now is for a way to turn down the innate human interest in our neighbors, which is the only thing capable of limiting our interest in what they’re sharing online. That is unfortunately impossible, and the social internet is unfortunately designed to do the opposite. But there might be a few tricks we can employ to sort of shake us out of our daily slop hypnosis. Not permanently, but every now and then. And even brief moments of clarity might be enough to help us course correct.

I recently stumbled back on an old clip of Peter Thiel’s, in which he tried to answer the question “what is worth building?” What is worth working on?

Find something important only you can build, he said, which would not exist without you. It’s inspiring enough advice on its face, which is how it’s generally taken, and how I always took it myself. But the advice requires an important, rarely considered step: introspection.

Do you even know what you’re good at? Do you even know what you believe about the world? These questions, if earnestly asked and regularly considered, act like a kind of mnemonic device in our hyper memetic world. Mental floss. Last week, as I watched the clip and self-reflected, I felt a little calmer. I thought about my life and my work. I remembered why Pirate Wires existed. I remembered who I was.

What do you really think — about the world, your company, yourself? The question seems trivial. But have you tried to answer it lately? It’s an easy enough puzzle, give it a crack. You might be surprised by what you learn, or what you can’t.

Look at the slop, by all means. Let the slop wash over you. Every now and then, when nobody is looking, it’s probably even fine to savor the slop — just a little, if you can handle it. This is the age of the internet, it will be a war of dumb ass blowhards and AI-generated bullshit for the rest of our lives. But remember: if you struggle with it, if you fight it, you become it. And I don’t think you have to become it. Let the slop wash over you, and keep walking forward.

What is something only you can do? What is something only you believe?

Try to remember.

-SOLANA