36 min read

Context Engineering: Making Sense of How the Epstein Class Weaponized the Internet

Art of a family in front of a screen containing the chatgpt logo. Scam Altman is wearing a yellow shirt. "Behavior power" is written with power highlighted in the same shade of yellow.
Bart Fish & Power Tools of AI / https://betterimagesofai.org / https://creativecommons.org/licenses/by/4.0/

I love museums, but I discovered my least favorite museum in Oxford. I don't remember it's name, but I do remember its interior. If you've ever been to the college town of Oxford and ran it's small circuit of museums, maybe you know the one I'm talking about. It was basically a massive square room filled with large rectangular pedestals covered in glass. Encased in them were all these things from the past, and almost none of them from the UK. Some of them were labeled, but unlike many museums I've been to, a lot of them were actually just left to their respective and arbitrarily decided categories like which part of the world they were from. I could see so many different kinds of things, and for many of them I had no context. I didn't know who these items belonged to, or what purpose they actually served, nor how beloved or useful these objects were, and certainly not the process through which they came in being. I had no idea how the museum had acquired them, although among those objects that did include labels, some did say they were looted and some even said from which battle. Most of the labels only really said which other part of the world they were extracted from. That's actually how the rows of rectangular pedestals were sorted; by which other parts of the world these decontextualized objects had come from. I may be misremembering the experience, as is common, but I will never forget the utter disgust I felt leaving this particular museum. This was largely because I had never before or since left a museum having learned nothing.

Many of our curated experiences online sum to the acquisition of content like treasures stolen from one's tribe and placed in a museum without any context. You are rarely the archivist and either everyone is a docent or no one is. I hope my storytelling here provides some context around the avalanche of recent events. This comes from someone who has unfortunately been participating since they were a child.

Jeffery Epstein, seated across from Steve Bannon in an a heinous indictment of our entire political apparatus.
Jeffery Epstein, seated across from Steve Bannon in a heinous indictment of our entire political apparatus.

The story of how we got here starts well before the internet, but let's skip past those bits for some very needed brevity. After all, we're trying to put how Jeffrey Epstein used his vast wealth to build a global network of human traffickers, human rights abusers, and angry horny incels on the internet into context. I will do so by providing labels to decontextualized objects.

The Internet, Freedom, and Sex

Let us begin in 1996 when Maria Farmer lodged a complaint with the FBI about Jeffrey Epstein and his abuse against her sister. The investigation was stamped out while Bill Clinton was president [0]. That same year, congress was also having a heated debate over the future of the internet. We passed Section 230 as part of the Digital Communications act. There were major questions over how to handle freedom on the internet. Questions like "who is responsible for illegal content, the host or the user who uploaded it?" Also bangers like: "how do we moderate hate speech, and who even gets to define it?"

Section 230 contains two key answers:

Section 230(c)(1) — “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.”

This means platforms like Facebook, Reddit, YouTube, or even comment sections on news sites are not legally responsible for what users post — whether it’s defamation, harassment, misinformation, or illegal content — as long as they didn’t create or develop the content themselves.

Section 230(c)(2) — “No provider or user shall be held liable… for any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable”

This gives platforms broad discretion to moderate content — including removing hate speech, spam, or violent material — without losing their legal immunity. I have to cite Speech Police by David Kaye here, that's an excellent book some people can zip through in a single sitting but others might need a lot more time to absorb [1].

Several artistic paintings of women in various states of undress.
Dominika Čupková & Archival Images of AI + AIxDESIGN / https://betterimagesofai.org / https://creativecommons.org/licenses/by/4.0/

To some, at first glance, the questions in the great debate of the internet were settled, just have the platform set a code of conduct and moderate hate speech themselves. I mean after all, the whole world has to go by these rules America decided on right? Moving right on, the biggest problem with content on the internet was the same as it is with most technologies at their inception. Misogyny. Charlie Warzel has a lovely interview with Sophie Gilbert where she lays all this out much better than I can [2]. From Pamela Anderson's stolen sex tape being sold, and shared without her consent, to Zuckerberg creating Facemash to rate the relative hotness of women at Harvard, to the rise of amateur pornography and deepfakes. The trend never really stops of course, years before Chatgpt even launched there was already a chatbot app on the store called Replika, where men were abusing their "AI girlfriends" within weeks [3]. Noelle Perdue, has an excellent substack about these topics and the history of porn, and she also has this great and relevant piece in 404 media on the way AI and isolation are distorting our how we see sex and fantasy [4, 5].

Artist image with harmful, violent, extreme content coming out of a burning pc. A young man is demoralized sitting in front of the pc.
Janet Turra / https://betterimagesofai.org / https://creativecommons.org/licenses/by/4.0/

I think this bit of understanding of internet history is incredibly important for understanding everything that happened next. You cannot make sense of the threat of Epstein's coordination of the global far right without understanding how he weaponized sex and the internet to do so.

Enter Social Media

I don't need to remind people about Myspace, instant messengers, slow-loading chatrooms, and how a bunch of millennials got groomed on them. They were barely before my time, but the story never really changed, it just moved platforms. Now we were post-Facemash, and the time of Facebook had come. Social Media was great, and Facebook was showing how tech could connect people all over the world in ways email couldn't.

An artistic rendering of a woman in a hijab throwing her phone with anger emojis forming a pattern. There are facebook screens in the background.
Image from cover of the Amnesty international report "THE SOCIAL ATROCITY: META AND THE RIGHT TO REMEDY FOR THE ROHINGYA"

Social media allowed trans people to start finding each other. It allowed black and indigenous people of color (BIPOC) to better document and share images of their own oppression. Disabled groups too. I myself found significant community in the Autistic Women and Non-binary self-advocacy networks, and was first published as part their anthology on Autism and Race.

As a teen I had unfettered access to the internet, and nothing had hindered the internet's access to me. While I was watching the US government kill US citizens who look like me, like Trayvon Martin, and Tamir Rice. I also was on every dating app that would allow me, which as it would turn out was all of them from tindr to linkedin. The free and open internet obviously has it's ups and downs, many of us were finally gaining visibility and greater autonomy in society. In middle school, I was convinced the future would be bright, because how could it be anything else when all the danger and history was laid bare before our eyes. By the end of high school, I knew I'd have to advocate for myself for the rest of my life, because few others would.

Social media and the internet didn't just enable progress by illuminating the margins of society. It also provided community for the dregs. On the dark web, anything, no matter how grim, was purchasable, not just the substances along the silk road, but also violence and cruelty itself [6]. When people want for nothing, they seem to eventually desire those wares, though it's hard for me to believe anyone is born with them. However, people with those desires were also forming communities online.

There's so many different iterations and versions of these communities on the free and open internet; and they serve different purposes in the radicalization of men. It is impossible to cover all of them, but lets cover some bases which seem critical to me.

Watching Stormclouds Form

Let's start with Stormfront, cause that's how I got started. It had the motto "The truth is "hate" to those who hate the truth!" and I had stumbled upon the site after being arrested and straddled by police officers at the tender age of 11. It wasn't my first encounter with racism of course, because I had mostly grown up in one of the whitest cities of America. No, literally [7]. I had high school teachers who still referred to Livonia by its nickname 'Sundowners', which was well earned as it was one of the last sundown towns to remain active in the US, if you ignore the present ones [8, 9].

I discovered the website after finding the Southern Poverty Law Center (SPLC) through basic googling about the racism I was actively experiencing in real life as a child. They had a hate group tracker, and I wanted to track the people who were continually hurting or dehumanizing me [10]. I became even more obsessed after discovering white supremacist forums repeatedly mentioning my hometown. Some of these guys were making posts about excitedly starting at my soon to be high school.

A close friend dated one of them for most of high school. He had a Nazi flag in his bedroom, which was unusual, as the confederate flag was a far safer choice at the time. His mom was also a cop, so he got away with a lot of the torment he seemed to delight in. I am over my friend dating him of course, I mean, we were kids. It helped that, before I had even met her, I had come across an old Klan book in a family friend's attic. His grandfather wasn't a wizard anymore and didn't seem to care that I was black at all. I never saw him leave his chair but the experience taught me people were more complex than categories and labels.

You know what was really inspiring though, and kept me engaged with politics in my childhood and kept me from despairing over being surrounded by either people who think demons are real – and that I'm one of them – and people who couldn't believe that these 'upstanding' white men would hate anyone? Presidential candidate Barack Obama. No man had a greater impact on my belief that we could do better than him.

While social media made it near impossible to avoid the topic of black lives being snuffed out without any regard to the law, we still managed to get representation in the highest office. Nearly every other black kid I had known in the Detroit area growing up, had a man in their family in prison. Mostly because of Clinton/Biden's 1994 crime bill, whose three strikes law put many (including innocent) black men in prison [11]. This bill focused on punishing criminals, even for non-violent crimes, as a deterrent rather than alleviating or addressing the systemic issues that lead to more crime in Detroit. To me Obama was promising hope for those kids whose fathers and brothers would spend the rest of their life in prison, a promise I was utterly disappointed by his inability to fill.

The backlash to that promise was immense and largely invisible to those not paying attention. The SPLC hate map was tracking the reaction to President Obama [12, 13]. Membership on these sites was rising rapidly as the representation of marginalized people in government, in schools, in hospitals, in courts, in journalism, in publishing, in most workplaces was increasing, just barely. White supremacists were very predictably mobilized by Obama's presidency. I was hyper-aware many doctors, lawyers, teachers, clergy, journalists, and politicians were white supremacists who did not want to share their profession with the people they hate. I was optimistic their children, born in age of the internet could not plausibly be the same despite all evidence I already had contrary. The internet had a way of fostering that specific cognitive dissonance.

Chans, Anonymity, and Pipelines

A lot of the the o.g internet memes that are ubiquitous with internet history like lolcats, Doge, etc. came from a site called 4chan.org. 4chan became a gathering place for online rightist movements like GamerGate and Qanon [14, 15, 16].

For those who had the distinct pleasure of not encountering or remembering gamergate, it was a movement against women and other marginalized groups, that operated under the guise of "ethics in games journalism." As a gamer of the time, my first encounter was when a friend had told me about a journalist, Anita Sarkeesian, and why she was ruining gaming. I didn't believe him of course, especially once I watched one of her accurate videos pointing out how many games excluded female characters entirely or at the very least excluded their agency. We are no longer friends, but I was keyed in then, as I am now. My google searches were recorded and the algorithms were recommending me ever more content about the brutal and disgusting harassment campaign against this woman and others. Which is somehow still ongoing today.

Three videos from a search engine style grid of influencers trying to farm rage and engagement by attacking a woman.
Screenshot of a search engine showing multiple recent videos from the far right attacking Sarkeesian even in 2025.

Sarkeesian wasn't the first or the last survivor of this extended harassment campaign. It continued to explode, as a gamer in 2013-2015, gamergate felt like an inescapable part of digital discourse. Gamergate had blossomed into an ugly digital movement as prescient and emblematic of the culture wars as post 9/11 neoconservative Islamaphobia or the #BlackLivesMatter and #MeToo movements [17].

You absolutely could not, and even today still would struggle to, play many online competitive games without being called a slur of some kind. But any time someone would advocate for themself and say, 'hey actually maybe we shouldn't let grown men harass young women and queer kids online' they would be met with indignant rage.

Imagine being told it was actually a white boy's privilege to be able to call someone a slur, to call me a f*****t, a n*****r, or a r*****d, and that it was actually just a natural part of being a boy, and that they 'love being a boy.' Don't ask about that last bit, you don't want the context. The vital unsaid part was it was natural for being a cishet white boy, and natural for people who aren't categorized as such to experience harm. It reinforced a social norm of hierarchy and violence. If you're a woman who has ever picked up a controller and joined an online lobby with voice chat on, you probably don't have to imagine. This aspect of our culture, which we continuously are made to believe is unflappable and unchangeable has had a global impact.

Many men and their friends who engage in this culture, do so from completely different understandings of it. For some, they're just too young, too green, to make sense of it. For others, its just part of fitting in. For some the game was only shared with their closest friends, they may talk like this, but only because that's how people talk. Social learning is contextual, after all. Some will even say that this shared language of objectifying and denigrating other humans – just trying to playing a game – is a key aspect of masculinity. To sum it up, either you thought this was a bit, a joke you were in on, a joke you were the butt of, or a joke that represented a truth about the social space of gaming, that it was only for white men. But it's never just a joke.

Gamers were a powerful force to mobilize, as gaming was one of the fastest growing and lucrative industries in America at the time. 4chan was where they were mobilizing.

Email released within the Epstein Files

Epstein had used his wealth and influence to bring Christopher Poole, otherwise known as Moot into his fold. Moot created one 4chan imageboard called /pol/ in 2011 in order to "contain" the rampant fascist hate speech on 4chan to that board. What very predictably happened instead, was that it became a gathering place for pedophiles, meninists, and white supremacists covered by anonymity. What followed was an ample construction of a network of fake news, and influencers in gaming, pornography, and fitness [18]. Teenagers were increasingly thrust into a world of extreme content and violence. Increasingly, loneliness and escapism were simultaneously amplified by big tech platforms, which were used to pull entitled men into these communities by convincing them they were disenfranchised.

A collection of a emails from Jeffrey Epstein illuminating the relationship between Epstein and Moot.

Radicalization wasn't just 4chan, and this like most things, is not a binary either. It was never a given that a boy who was ostracized at school for inappropriately interacting with a classmate would find support and community online with grown white supremacists groomers, even though there are no shortage of incel stories online that read like this, bar for bar. These stories, as unfortunate and numerous as they are, do not negate the fact parents were not helpless. Parents and community could and often did prevent or reverse this. The idea these boys and the people around them have no autonomy at all but all the entitlement is also a comforting lie.

It is not comforting to those of us harmed by these alt-right pipelines continually creating mass shooters like Eliot Roger or Dylan Roof [19, 20]. But it's really important to remember that not every white boy calling strangers the n-word in voice chat ended up on white supremacist or pick up artist forums saying that 'foids' – one of way too many words they invented to dehumanize and classify others – shouldn't exist. For many it really was just about being part of the in-group, just jokes, or just a meme. But if you have ever had an acquaintance who spent almost all their time on 4chan, you come to understand that they typically believe nothing matters. That any joke where cruelty is the point, is not a joke at all.

Memes and Nihilism, the Truth about Pepe the Frog.

Donald Trump utilizing Pepe the Frog as campaign imagery

Me to a friend sending me a funny meme in 2015: "Pepe the frog is a white nationalist symbol." I sounded crazy. That was the point. Matt Furie created pepe in Microsoft Paint as part of his "boys club" zine in 2005 with the catchphrase "feels good man." By 2008, Pepe made it to 4chan. "Feels good man" quickly started to feel pretty bad, but there were many other new variants as well as the meme proliferated rapidly on /b/ and other 4chan imageboards. In 2015, Donald Trump and the far right had fully leaned into their campaign to make Pepe a symbol of hate. By 2017, a pepe variant became the symbol of the Groyper movement [21]. Specifically, this version.

A poorly drawn picture of a toad. That is all it is without context.
A poorly drawn picture of a toad. That is all it is without context.

"Groypers" refers to a subsection of the far right nationalists and conservatives led by political commentator Nicholas J. Fuentes. Fuentes hates women, and himself, and has convinced a lot of men our age to feel the same. The Fuente's line to best understand the Groyper role in the radicalization of young white chronically online men is, "If you get redpilled, don't tell your mom. If you get redpilled, don't tell your mom. If you get redpilled, don't tell your parents, alright? […] DON'T TELL MOM AND DAD! […] If you get redpilled, play it close to the chest, don't tell anybody, keep it to yourself, keep it to you and your online friends, alright?" [22]. I recommend reading his wiki page for any other info, because I just don't care about this gusano enough to expand on his role much further. It's worth the read though for those who only know this incel through his "Your body, My choice" video.

For those who have seen the word incel in this article three times now and do not have the context. Incel is short for 'involuntarily celibate,' the meaning contained in 'involuntarily celibate' differs from the meaning many of us have constructed in the shorthand version. Involuntary celibates are strictly those who are celibate despite wanting to have sex. Incels are an entire internet sub-culture that began on social media sites like reddit and a long list of pick up artist forums dating back to the start of the internet. Most incels today, like Fuentes, wear the label with pride online in spite of its well-known associations with misogyny and violence.

However, I do want to talk about this ideology a bit, if it can be called that. The Groypers are not loyal to far right ideology, as much as they are to Fuentes [23]. Though the movement gained traction around 2019 when it started targeting far right conversative group, Turning Point, led by Charlie Kirk, it was more aligned with the group than not, and honestly looked to be more of an accelerationist strategy to push already highly conservative groups further right.

They were deeply anti-semitic, homophobic, and misogynistic and their strategy was to disrupt college events by asking provocative questions about immigration, Israel, LBGT rights, and bodily autonomy, in an attempt to challenge mainstream conservative figures like Kirk, and Ben Shapiro as "Conservative Inc." Of course, figures on the right like Shapiro and Kirk absolutely loved the engagement.

Simultaneously, this group of young men, who knew how to hide, infiltrated more and more of the GOP. This article from Media Matters, provides a really good insight into how that strategy is going pretty well [24]. Despite the memes these guys would constantly share of "oh no we're fucked if the group chat leaks" which spread all over Instagram alongside Pepe/Wojak memes. The group chat did leak, turns out these guys are just everywhere hiding in plain sight, with some estimates saying they make up to 4 in 10 republicans [25]. More importantly it turns out few actually cared about their hate speech and there were also few consequences besides some staffers getting fired. Our next in line for the gilded crown, JD Vance, said it was just boys being boys.

The thing to understand about the Groypers, is that what looks like politics – the words and images that look like they carry some meaning – are often just stochastically strung together to give the illusion of coherence. Some Groypers are grifters like Candace Owens, successfully convincing isolated people on the far right that Charlie Kirk was a time-traveler. Some Groypers are teenage boys brought into the fold by Fuente's America First podcast, usually via some other fitness or gaming manosphere influencer. Most though just don't think anything matters. Most are just blackpilled nihilists.

Cy Canterel delivers a crash course in meme extremism. [26]

The video above shows very well how the belief that nothing matters is at the center of these online alt-right pipelines. These hopeless ideas have now even bled into the mainstream reinforced by social media algorithms. Social media algorithms push pro-suicide content to suicidal teens, they push misogynistic content to men, highly sexualized and pornographic content to children, and extremist content to everyone [27, 28, 29, 30]. Doom scrolling fostered – in a large portion of an entire generation of men – this idea that nothing matters, that people did not matter, and that reform or improvement of individuals and society was impossible, especially for their own.

In their death march, the Groypers made a targeted online campaign to erode truth itself, and 4chan memes like Pepe and Wojak were just one tool in their black box, not even uniquely theirs. Honestly, a great elaboration of how and why the far right attempts to erode meaning using memes is in this well-researched video on the new aesthetics of fascism [31]. Ben Hoerman gives an excellent explanation of how fascists understood the power and ability of visual imagery to mold our psychology. The most essential take away is that the Groyper will usually provide some idealized version of the past, but never of a better future.

Ben Hoerman's Video on the Aesthetics of Fascism

Back to sex and objectification

Sorry to bring this back to sexist origins of the internet. But something we don't talk about are these #gooning and #edging memes that have been pervasive on social media platforms for young teens. We don't talk about how on tech platforms, content with this tag might be a meme, or it might be actual literal pornography. We do not discuss that every teenager with Tiktok, Twitter, or Instagram, is familiar with this meme or that 4 in 10 teens watch porn during school [32].

This is an uncomfortable topic, but it's hard for someone keyed in on internet culture for so long not see a powerful pipeline there. Charlie Warzel touched on this with his interview with Sophie Gilbert, but it seems like meme culture was weaponized to get young men addicted to pornography at a young age and pull them into a false reality where women are objects.

I think this is not just about young men either. Pornhub, and other pornsites have a dark and massive surveillance apparatus, influenced by both policy and malicious actors [33, 34]. However, it suffers from one of the same problems of social media, any user can upload anything and it's hard to moderate what users upload. Moreover, different demographics are pushed different content. Worse, as we see, dangerous communities form of shameless individuals in a positive feedback loop.

I am not anti-pornography, or anti-sex work, and am fully aware of the harm that anti-porn orgs have done to sex workers. However, I do think Professor Clare Mcglynn makes an excellent case that this is a topic we are going to have reconcile with [35]. If porn is having an effect on our society, on how we view each other and ourselves, then it's worth interrogating our access and regulation of it. Regardless, it's very clear, that porn has been one tool of far too many to amplify the alt-rights recruitment of young men, whether it's just desensitizing them to violent or non-consensual acts, or directly connecting them with men like Epstein.

Enter LLM Slop and Context Windows

There are too many problems with AI to list in this section or even this very long article. It's why I keep reading so much about them, and why after years of this, I have not run out of things to read [36, 37]. But the one I wanna speak about here is the erosion of truth and knowledge; and we can start by talking about the algorithmic boundary line between hate speech and protected speech.

Before that though here is a is quick table for conceptual clarity.

I created this table because it's really frustrating to see the general public and even scientists from other fields toss all of AI in the landfill. There is a strong case to cease all production of LLMs today, but a very weak one to do so for all ML models. We've had ML models for decades and they are used in all sorts of critical systems, although their use in those systems alone still justifies interrogation, but not necessarily removal. LLMs are the products that have flooded the general public with the "AI generated slop" with which we are all familiar. LLMs are exacerbating our race through the climate crisis on a magnitude we cannot even understand because some potentially pro-extinction billionaires do not care to provide full transparency on their emissions [39, 40]. ML does actually have positive impacts on climate technologies and has been used already for that end, at least by the countries who need it least. LLMs do not; though you would not know that listening to the CEOs of tech companies trying to sell you them as a product [41]. Actually, I started this paragraph with a lie. Conceptual clarity is important for it's own sake [42].

My first peer-reviewed publication in AI was during my undergrad, and before Chatgpt had ever existed. It was in a sub-field of machine learning called adversarial machine learning. The entire motivation behind this field was 'adversarial robustness' or in other words making AI models better defended against attacks from bad actors. My first paper was pretty much entirely about how Nazis could evade algorithmic hate speech detection, in the hopes that the people building these models would use that research to make their models better able to stop hate speech from going undetected [43]. A weakly robust model might be able to identify "I hate immigrants" as hate speech, but might miss something like "I h@te imm1grants." Humans often miss that these misspellings can be intentional and not by humans at all. The issue of adversarial robustness while easily addressable is simultaneously impossible to address, h@te can turn into h8 or heit or h4te, and the model has no way of remembering what it feels like to hate something.

AI models can't hate anymore than they can feel, which is not at all. They certainly can't remember the feeling of hate anymore than they can remember fear. Hate and fear are as computationally intractable as the pain that causes them. You can't remember something you can't experience. I don't cite Cognition and Intractability here because I'm still working my way through it. I could tell you about a time when I was able to compute hate and cite that instead, but I have no memory of a time like that.

The way LLMs 'remember' can be thought of in two ways: parametric-knowledge, and non-parametric knowledge. Parametric knowledge is basically the data it was trained on. A recent study shows that LLMs can reliably reproduce any complete work in their parametric knowledge from a NYT article to Harry Potter [38]. We are told these near-complete reproductions are the result of learning and not the process of copying.

Whereas non-parametric knowledge is, well... context. This includes everything from the prompt of the user, to some subset of the 'knowledge base' it crawls during your interaction with it, to the data in your current browsers cookies. However, all of these models are limited in how much of this 'context' they can store at once. This might be called a working memory if it worked or had the critical characteristic of memory that is experience. Instead it is called a context widow [39]. Though Manufacturing Consent author, Noam Chomsky, might call it an overton window if he wasn't pals with a nonce. I'm sorry, do I need to cite uncommon unknowledge?

Like the museum in Oxford, one's experience with the content produced by LLM models is marred by a lack of context. Without that context there can be no learning. Without memory, the true identity of an object can never be known. Without memory, a building block of reasoning, a human could never tell the difference between hate speech and h@te speech.

What is the utility of LLM products if they require so much input human labor to create and maintain, and they still can't tell the difference between hate and h4te better than a human can? When I think about how one labels hate speech, and I think how I would go about it, I realize there are no correct answers that will satisfy everyone. Yet, people over the past few years have overwhelming grown to trust these interfaces as objective arbiters of truth. We all see in them our own biases, confirmed and validated. That is the truth.

For those with enough context, it is hard not to see chatbot interfaces as a product whose strongest utility is distorting truth, if not shaping it. If you have an interest in politics this utility is so valuable you might actually spend trillions of dollars on it. Billionaires know that enough money can mobilize entire movements, and halt others, what could a new cultural technology like LLM chatbots do? How could it shift the overton by changing how we see the world and the contexts under which we communicate with each other?

Tea and Activism

Sorry I'm moving around so much, life can be so chaotic, and I'm just trying to give context about how Epstein was able to weaponize the internet against democracy. While social media was allowing communities and progress to form, a movement that would completely alter our political landscape was also occurring. To be fair, we could talk about multiple different movements from 2000-2010 here, like the anti-war movement against Iraq, the drive for digitization, or even the Arab Spring. However, I'm actually going to talk about the Tea Party movement. The Tea Party movement was a conservative political movement that emerged in the United States around 2009, largely in response to the election of President Barack Obama.

It was fueled by 'grassroots' activism, conservative media, and organizations like FreedomWorks and Americans for Prosperity [44, 45]. The movement emphasized fiscal responsibility, limited government, and constitutional originalism, and played a significant role in shifting the Republican Party further to the right, influencing elections and policy debates throughout the 2010s, by moving "the range of subjects and arguments politically acceptable to the mainstream population at a given time" [46].

While the Tea Party was not uniformly aligned with white supremacist groups, some factions within the Tea Party overlapped ideologically or rhetorically with far right elements, particularly as Obama’s presidency became a focal point for racialized political backlash. They pushed the overton window on immigration and bodily autonomy. While social media was working to polarize the general public online, the tea party was working on the ground to radicalize the GOP. Pushing the overton would stop people from arguing over whether an action or statement was bigoted, or whether it was harmful, and instead arguing over whether or not it was justified.

Trans people may or may not love tea parties, but the Tea Party hated trans people and acted like it [47, 48]. These guys like Todd Kincannon were saying on Twitter that trans people should be put in concentration camps in 2013 while he was mocking U.S citizens murdered by police [49]. This was protected speech apparently, but now while U.S citizens continue to be murdered by the state today, there are also trans people in the concentration camps that had already existed in 2013. While trans people were just finding each other on the internet, and I was protesting Obama putting kids in cages, the Tea Party was mobilized and united in their fervor to oppress us.

It's important to understand that most Americans and most people globally simply did not care about trans people, and I actually mean that in a somewhat good way. On the one hand, most people at the time didn't even know what transgender meant. It's shocking how many still don't. On the other hand, even if you did know, so what? This is the common response, other people exist and there are other ways of being, if you assumed (wrongly) that trans people were on equal footing in terms of rights, then without an external influence, why would you care?

The Tea Party base made it their full-time day job to make people care. They called their representatives, EVERY DAY. They organized highly visible protests, carried by tailwinds blown by mainstream media. While 90% of America did not care about trans people, as it was simply not a salient issue, the tea party – who most sane people wrongfully mocked as insane – was putting LGBT rights front and center alongside immigration and abortion. The biggest downside of low saliency was that trans people also wouldn't get support, resources, or even just the civil rights the community needs. The Tea Party showed up at every town hall, and wrote and called their reps every day, while more and more American's were growing convinced that political engagement was pointless. MAGA coming to power in 2016 was one of direct results of the tea party 'grassroots' activism.

As trans people reached the first trans tipping point in 2014, the visible progress of the community started to aggressively clash with the unrelenting vitriol of the evangelical right [50]. Trans issues were now salient – if not to the general populace – then certainly to politicians across the country. In fairness, its not as if the left was silent either, but it was certainly frustrating to organize any sort of political action back then, and find that thousands of people might show to a protest, but only 50 of those people would actually mobilize and contact their representative. From my perspective the right doesn't understand many difficult things, but they understand many easy ones, and civic engagement is as simple as making tea.

How to Influence an Election

This section isn't a guide but it is another part of this tragically short and simply tragic history lesson. This section is all about Cambridge Analytica (CA) [51]. Skip if you're hyperaware, as this one got a lot more coverage from my perspective; yet seems far less prevalent in our shared public memory.

Cambridge Analytica was a British company founded as a subsidiary to a right-wing British think tank. It was founded by Alexander Nix, and was more a collection of firms and companies aimed at providing data analytics to organizations to augment political strategy for campaigns of various kinds. They ran the 2016 election campaigns for both Donald Trump and Ted Cruz.

They succeeded because they harvested the Facebook interests, likes, and other data to predict behavioral patterns of over 50 million Americans [52]. Those harmless online quizzes weren't so harmless. They allowed CA to distribute targeted disinformation and fake news to people who could potentially be most vulnerable to it. This alone was a global scandal, and led to many countries demanding transparency and accountability from Facebook and CA. Mark Zuckerberg called it an "issue", a "mistake" and a "breach of trust" [53].

It was an even bigger scandal when an investigative reporter from Channel 4 news filmed a confession from Nix that he had used young Ukrainian sex workers to honeypot politicians across the US and UK [54]. Guess who was the Vice President of Cambridge-Analytica? No, it wasn't Epstein, but very close guess. It was Steve Bannon – founder of far right wing disinformation outlet Breitbart – who would be appointed as Donald Trump's White House Chief Strategist. One of the major victories the left did have during Trump's first term, was getting Bannon's resignation after he spurred on the Unite the Right movement in Charlottesville [55]. You remember, the angry incels with tiki torches? Yeah, that was looking like Epstein's influence too. Epstein regularly was meeting Bannon and these figures on the alt-right, globally [56, 57].

I hate to encourage apophenia to be honest, which is distinctly hard to do here, but take a second to think about this. We didn't have much insight into Nix's confession before, but now with the small portion of Epstein files we do have, we might be seeing that Cambridge-Analytica was almost like a test of how you could weaponize wealth and the internet against Democracy. It's not hard to deduce where Bannon and Nix where trafficking these girls from, not to say that information is useful. We really should not be forgetting about Cambridge-Analytica though, or any of Trump's first term, when reacting to current events or trying to undo Brexit.

Rage, Entertainment, and Engines of Extremism

While the British had Peter Mandelson architecting their new labour party, Boris Johnson was apparently meeting with Cambridge Analytica much to the ire of his colleagues. Ire being my more pleasant way of saying anger – and delight is such a lucrative thing. The thing Stephen Miller loves and probably the only talent he actually has, is inciting rage. From his high school election to the White House podium and the algorithmic underbelly of Twitter, he weaponizes outrage not as collateral damage, but as design. He architected the Trump regime's policy of family separations to provoke, polarize, and terrorize [58, 59]. There's a reason Trump promised to "shock and awe" on day 1. The more the public screams, the more his base rallies, and the more the algorithms amplify the engagement. Miller's modus operandi is almost like an evolution of Chomsky's work of 'Manufacturing Consent.' In light of Chomsky's ties to Epstein and the global far right, this honestly is no longer that surprising. Instead of waiting for the media filters to limit debate, marginalize dissent, and frame issues to fit dominant – or at this point, nonsensical – narratives, Miller creates the outrage himself. Then he simply lets the media and the algorithm amplify it.

This content and influence first strategy that helped Trump win is readily apparent in how this regime is obsessed with generating content for social media. Content, influence, and engagement seem to be the tactical pillars for exacerbating rage and core to whatever semblance of strategy the White House has. In the feedback loop of social media where nuance dies and fury thrives; as long as the platforms reward the loudest, the angriest, and the most extreme, the Trump regime wins.

Social media obviously didn’t invent human anger, but it has transformed it into a kind of currency. When we design algorithms that reward outrage, platforms like Facebook, Twitter, and YouTube become optimized to serve you outrage. That rage becomes polarization, not because we’re inherently divided, but because division is profitable. The more you rage, the longer you scroll, the more ads you click, the more valuable data you leak. And the more you’re fed the next outrage, the next conspiracy, the next meme that tells you the world is collapsing and that you’re one of the few who see it.

This really isn’t just about the content pipeline. From incel forums to the Groyper group chats, from Gamergate to the Tea Party’s Twitter mobs, rage is an adhesive. It binds the lonely, the alienated, the disillusioned to the illusion of power in being loud. The proliferation of hate speech and violence on Facebook lead to a brutal genocide of the Rohyingya people in 2017 for which Meta faced a $150 Billion lawsuit [60]. Zuckerberg said in 2018 due to external pressure that they would and could do better (ignoring the fact this is literally profitable). Meta enabled genocide again in Ethiopia just a couple years later [61]. While nothing makes them more money than rage, it's important to understand the nihilists thrive on any reaction whatsoever. Nuance is weakness on the internet and context is even worse because context is capable of fostering empathy and indifference in equal measure.

Decontextualized Noise

The reason why I wrote this article is because our information environment is full of decontextualized objects, like a free, public, museum we are paying a tax to keep open. We see only a fragment of the truth of each object and a small abstraction of reality – the embodiment of one piece of someone else's lived experience and perspective. A decontextualized object is almost like a stereotype, in that it is a small window we use to see into an outside world. We are expected to be in the know, to understand this obscure reference, or trending hashtag. We are expected to have witnessed all of the same events, and to have the same understanding of them as the 'OP' or original poster.

If I made a Bluesky post saying "we need to get money out of politics," this means so much less to someone unfamiliar with Citizen's United than it does to someone not only keyed in on what that decision meant, but also who regularly would visit opensecrets.org before its recent changes [62, 63]. Many somehow simultaneously assume a person in the know could only be fully aligned, as that is the stance aligned with the values we attribute to others. However, there are so many people who have no idea what Citizen's United was or how it completely corrupted American politics, which had already been pretty corrupt.

One might suggest just asking, but people are often shamed just for asking for context, because we're primed for the specific type of engagement that is defensive moral outrage rather than deliberation and education. This must end because we all have different understandings and subjective experiences of the world. We all have different values and beliefs. No two humans are identical in their thoughts and ideas, especially not your digital twin. Our diversity is profound and should be celebrated rather than forgotten or suppressed.

I wrote this piece to give some people – especially those people who are just starting to pay attention – some understanding of my perspective on this fight. Our fight against the 1%, the pedophile class, that has oozed it's way into US and UK politics. It is a suggestion of how we can rail against this stupid idea that keeps rearing it's ugly head, and now poses us all existential risk. I could not even touch on half the facets or depths of what appears to be a global conspiracy to empower fascism, at which Epstein was the center, but not the only one complicit in human trafficking [64]. Though it's important for conceptual clarity to say, that none of this was a conspiracy, but rather just how the world we collectively construct operates. For clarity of action, it's important to understand the boundary lines between 'hate' and 'love' are not a binary classification problem.

The context window we view the world through is shaped just as much by the data points we don't see. I did not cover the history of fascism, which is fascinating and goes back much further than the Nazis, who actually were heavily plagiarizing ideas from Jim Crow, Slavery, the Trail of Tears, and a slew of prominent American eugenicists. I did not touch much on Q-Anon, which like Gamergate also started on the same 4chan board, nor covid and how pandemics and polarization enable fascism. I didn't even talk much about the other tech shaping our current political nightmare, like the move to the cloud and our dependence on private energy, they will continue to weaponize against us. I don't talk about the surveillance state we've allowed these tech companies to create either. I don't go as far as to write out quietly how we need to have a global and free internet but haven't figured out how to do so without helping billionaires traffic children. I don't talk about how Obama made fun of Trump at a statehouse dinner and in response he started a birther movement and ran for President or even Kristi Lynn Arnold Noem's initials. Even I find the humor in that one, though I'm not sure if I should. I also don't talk much about many of the figures in the Epstein files.

It is impossible for any individual to track all of it. Our academics and journalists will spend decades discovering just how much violence these men have inflicted on and hidden from the rest of us. They will never have a complete answer. That's okay though, not everything has or needs answers, and we should make peace with the uncomputable. However, we also should see the release of these files for what they are, decontextualized objects which can only be given context collectively. They need to be investigated just as much as we need to eat food to survive or reverse climate change, but complexity is a time-suck best tackled by collaboration. I hate the othering that I'm doing here, but in my own weakness I know we will win because we will choose it, because we can deliberate and decide upon the meaning of words too. It's why I'm sure of it. The Epstein class realized long ago that they can shape reality through framing and filtering information. They use coercion to help the rest of us shape the context we live in to their own design specification.

How to Create Context - Beginner's Guide

In the previous section, I lied, because I'm imperfect, and imperfect people lie sometimes. The real reason I wrote this is because the fascists are still in power but they have removed their masks. I wrote this because I am painfully aware it's going to be a global effort. I have been trying to fight this deadly idea since I was a child. Those of you who still have empathy need to understand what we are up against.

Fascism is not an ideology of white supremacy. It's an ideology of death and power. It is not about inequality alone. It does not care for whatever colors or gender you are currently wearing. It may use labels to divide and conquer, but it only cares that you wake up and draw breath. The only thing it loves more than death is power. Fear is just one tool in their arsenal to amplify their power over death. It is always fostered by men who believe nothing matters. Those men are wrong and it's not a necessity to engage with their enmity, it's a choice. Every choice matters and it's good to care about each other and the world. It feels great to breathe air and eat food and touch grass. Every body matters, every single one of us, no matter how different or flawed.

These billionaires believe they can use the algorithms we have integrated into every aspect of our lives to force us into submission, but we cannot submit to death. It is impossible to automate ambiguity and we should say it [65]! Fascist systems cannot see the 'map for the territory', nor the person from their classification. The abstractions they deliberated and decided on behind closed doors are why some of the fascists think they have won, but it is actually why they have already lost.

You cannot predict or label a human in their totality anymore than you can impose order unto chaos. Algorithms are often wrongly assumed to be neutral. That perceived neutrality is precisely why we turn toward them to solve complex problems which we believe require objectivity like moderating hate speech online [66]. However, human bias is an inalienable aspect of being. Human bias will be present in any algorithms we create, whether through the data they’re trained on, the design choices we make, or the contexts in which they’re applied. Without careful intention in their design, they amplify those biases, often beyond the context of the system itself. Moreover, the ones designing them are – at the least – aligned with fascism. This would remain a major obstruction to fairness even if we could universally agree on an abstraction of what fairness means [67].

Democracy and deliberation are key parts of how one can make sense of the world, they're my best guess at what fairness is. Participation is my best guess at how each of us can resist being stalled by decontextualized objects simply by seeking context. Read books. Read independent news. Ask experts. Talk to your elected officials and tell them you think immigration is good, and accessibility matters to you. If you would like to live under tyranny of the majority rather than the Epstein class, than you should actually just directly tell your representatives and communities what your needs are and advocate for yourself. Emailing your representative is cheaper than the therapy that should be free and will need to be. You can always ask academics for guidance, a lot of great ideas are already being deliberated, but for the worlds sake, learn to recognize grift. It's critical to think about how you know you know something – as we're all learning. It is not too late.

Speak up from where you are and whatever your role. We can and need to practice better. We need to stop competing and start collaborating if we are to survive on this rock and to survive each other. We need to mobilize peacefully, today and every day. It's either we summon the courage to talk to each other openly and gently now or suffer more when the billionaires run out of lies. I don't know yet if "openly and gently" are the right words, but I do know that during his trial for war crimes, Adolph Eichmann, excused his actions by saying the language of the Nazis made it easy to send thousands of people to their death. That language was 'Amtssprache' - the language that denies choice [How to do non-violent communication]. It takes courage to seek out and use the right words. Courage is the antidote to terror and it's going to take immense courage to achieve a world of consent and consequence rather than power and punishment. Like interrogating incarceration [68].

It will also require us to stop simply critiquing systems we can no longer accept as unchangeable. Once we see the world for what it is by asking what else is true – and name reality – we need to do the hard and laborious task of imagining the world we actually want to live in. Then we need to work towards those futures by practicing deliberation as a collective and daily exercise.

I hope the undergrads studying International Relations today think about what consent and sovereignty really mean by reflecting on Mark Carney's speech carefully, and brainstorming how we actually make these ideas a reality for everyone. [69]

Please log off and talk to each other as much as you love each other. Please join your local unions and your organized political parties, invite your neighbors too. Please invite academics and policymakers to focus more on how we achieve better futures and educating the public on why we need a better present. Remind them the very concept of the university will not survive fascism, and remind each other it's imperative to voice your needs. Practice establishing boundaries so that we may share the time needed to imagine and advocate for a better today. Everyone has a different role to play and capacity for labor, but none of us can afford to opt out. Resisting fascism will require 100x the organizing fervor of the Tea Party, and it will include many of them as well.

In the end I know this article will disappoint many, because it doesn't provide enough context on its own. Some will even liken it to anti-research, just words in the form of noise without the practices of accountability behind. I will say this is not research done on behalf of my institutions, but just a way of expressing myself. As none of us are removed from the context and conditions we live in, neither is this journal article entirely disconnected from my research, my community, or my values.

Lastly and again, we create meaning collectively and I've got to get back to my 'real' thesis which may or may not require me to accept perfection as the 'enemy' of the 'good'. I'm happy to get back to it though because I'm confident time will reveal that love is just as infinite – and that our love for our communities can never be decontextualized into slop. That's how we win, we do "the work we need to do to survive," while we act upon words and create new meaning together.