Mindf*ck: Cambridge Analytica and the Plot to Break America

by Christopher Wylie

Although each like taken in isolation was almost always too weak to predict anything on its own, when those likes were combined with hundreds of other likes, as well as other voter and cons...

We all went into the boardroom again, with the giant screen at the front of the room. Jucikas made a brief presentation before turning to Bannon. “Give me a name.” Bannon looked bemused and gave a name. “Okay. Now give me a state.” “I don’t know,” he said. “Nebraska.” Jucikas typed in a query, and a list of links popped up. He clicked on one of the many people who went by that name in Nebraska—and there was everything about her, right up on the screen. Here’s her photo, here’s where she works, here’s her house. Here are her kids, this is where they go to school, this is the car she drives. She voted for Mitt Romney in 2012, she loves Katy Perry, she drives an Audi, she’s a bit basic…and on and on and on. We knew everything about her—and for many records, the information was updated in real time, so if she posted to Facebook, we could see it happening. And not only did we have all her Facebook data, but we were merging it with all the commercial and state bureau data we’d bought as well. And imputations made from the U.S. Census. We had data about her mortgage applications, we knew how much money she made, whether she owned a gun. We had information from her airline mileage programs, so we knew how often she flew. We could see if she was married (she wasn’t). We had a sense of her physical health. And we had a satellite photo of her house, easily obtained from Google Earth. We had re-created her life in our computer. She had no idea.

“Wait,” he said, his eyes widening behind his black-rimmed glasses. “How many of these do we have?” “What the fuck?” Bannon interjected with a look of annoyance at Nix’s disengagement with the project. “We’re in the tens of millions now,” said Jucikas. “At this pace, we could get to 200 million by the end of the year with enough funding.” “And we know literally everything about these people?” asked Nix. “Yes,” I told him. “That’s the whole point.”

CA had noticed was how many Americans felt closeted—and not just gay people. This first came up in focus groups and later was confirmed in quantitative research done via online panels. Straight white men, particularly ones who were older, had grown up with a value set that granted them certain social privileges. Straight white men did not have to moderate their speech around women or people of color, because casual racism and misogyny were normalized behaviors. As social norms in America evolved, these privileges began to erode and many of these men were experiencing challenges to their behavior for the first time. At the workplace, “casual flirting” with female secretaries now imperiled your job, and talking about the “thugs” in the African American part of town could get you shunned by peers. These encounters were often uncomfortable and threatening to their identity as “regular men.”

Men who were not used to moderating their impulses, body language, and speech began to resent what they saw as the unfair mental and emotional labor it took to change and constantly correct how they presented in public. What I found interesting was how similar the discourse that emerged from these groups of angry straight men was to liberation discourse from gay communities. These men began to experience the burden of the closet, and they did not like the feeling of having to change who they felt they were in order to “pass” in society. Although there were very different reasons for the closeting of gays and the closeting of racists and misogynists, these straight white men nonetheless felt a subjective experience of oppression in their own minds. And they were ready to emerge from the closet and return to a time when America was great—for them.

“Think about it,” I said to Bannon. “The message at a Tea Party rally is the same as at a Gay Pride parade: Don’t tread on me! Let me be who I am!” Embittered conservatives felt like they couldn’t be “real men” anymore, because women wouldn’t date men who behaved the way men had behaved for millennia. They had to hide their true selves to please society...

The incel community, just coming to the fore when Cambridge Analytica was being established, was the kind of group he had in mind. Incels, or “involuntary celibates,” were men who felt ignored and chastised by a society—particularly women—that did not value average men anymore. An offshoot of the Men’s Rights Movement, the incel community was in part propelled by the increasing economic inequality depriving young millennial men from accessing the same kinds of well-paying jobs their fathers had. This economic deprivation was coupled with increasingly unattainable body image standards for men in conventional and social media (without the same public recognition of male body issues or gendered pressures as for women) and the growing importance placed on physical looks in a dating scene increasingly defined by swiping left or right on a split-second glance at a photo. And as women had become more economically independent, they could afford to be more selective about their partners. Deprived of good looks and a respectable paycheck, “average men” faced a hard reality of constant romantic rejection.

Ongoing jokes and memes would be shared about resisting their life sentences and waging a Beta Rebellion or Beta Uprising to fight for the redistribution of sex for the betas. But lurking behind the strange humor was the rage of a life of rejection. In scrolling through these narratives of victimhood, my mind turned back to the narratives of extreme jihadist recruitment media, with the same naïve romanticism of oppressed men breaking the shackles of a vapid society to transform themselves into glorified heroes of rebellion. Likewise, these incels were perversely attracted to society’s “winners,” like Donald Trump and Milo Yiannopoulos, who in their warped view represented the epitome of the same hypercompetitive alphas who brutalized them, to lead the charge. Many of these seething young men were ready to burn society to the ground. Bannon sought to give them an outlet via Breitbart, but his ambition didn’t stop there. He saw these young men as the early recruits in his future insurgency.

First we used focus groups and qualitative observation to unpack the perceptions of a given population and learn what people cared about—term limits, the deep state, draining the swamp, guns, and the concept of walls to keep out immigrants were all explored in 2014, several years before the Trump campaign.

Cambridge Analytica did this because of a specific feature of Facebook’s algorithm at the time. When someone follows pages of generic brands like Walmart or some prime-time sitcom, nothing much changes in his newsfeed. But liking an extreme group, such as the Proud Boys or the Incel Liberation Army, marks the user as distinct from others in such a way that a recommendation engine will prioritize these topics for personalization. Which means the site’s algorithm will start to funnel the user similar stories and pages—all to increase engagement. For Facebook, rising engagement is the only metric that matters, as more engagement means more screen time to be exposed to advertisements.

This is the darker side of Silicon Valley’s much celebrated metric of “user engagement.” By focusing so heavily on greater engagement, social media tends to parasitize our brain’s adaptive mechanisms. As it happens, the most engaging content on social media is often horrible or enraging.

IN THE SUMMER OF 2014, Cambridge Analytica began developing fake pages on Facebook and other platforms that looked like real forums, groups, and news sources. This was an extremely common tactic that Cambridge Analytica’s parent firm SCL had used throughout its counterinsurgency operations in other parts of the world.

The firm did this at the local level, creating right-wing pages with vague names like Smith County Patriots or I Love My Country. Because of the way Facebook’s recommendation algorithm worked, these pages would pop up in the feeds of people who had already liked similar content. When users joined CA’s fake groups, it would post videos and articles that would further provoke and inflame them.

Now CA had users who (1) self-identified as part of an extreme group, (2) were a captive audience, and (3) could be manipulated with data.

Once a group reached a certain number of members, CA would set up a physical event. CA teams would choose small venues—a coffee shop or bar—to make the crowd feel larger. Let’s say you have a thousand people in a group, which is modest in Facebook terms. Even if only a small fraction shows up, that’s still a few dozen people. A group of forty makes for a huge crowd in the local coffee shop. People would show up and find a fellowship of anger and paranoia. This naturally led them to feel like they were part of a giant movement, and it allowed them to further feed off one another’s paranoia and fears of conspiracy. Sometimes a Cambridge Analytica staffer would act as a “confederate”—a tactic commonly used by militaries to stir up anxieties in target groups. But most of the time, these situations unfolded organically.

That was when you introduced the next tier: people whose initial resistance to rumors had started to weaken. And this is how you gradually destabilize an organization from the inside. CA wanted to do the same to America, using social media as the spearhead. Once a county-based group begins self-organizing, you introduce them to a similar group in the next county over. Then you do it again. In time, you’ve created a statewide movement of neurotic, conspiratorial citizens. The alt-right.

Internal tests also showed that the digital and social ad content being piloted by CA was effective at garnering online engagement. Those being targeted online with test advertisements had their social profiles matched to their voting records, so the firm knew their names and “real world” identities.

CA estimated that if only 25 percent of the infrequent voters who began clicking on this new CA content eventually turned out to vote, they could increase statewide turnout for the Republicans in several key states by around 1 percent, which is often the margin of victory in tight races.

Steve Bannon loved this. But he wanted CA to go further—and darker. He wanted to test the malleability of the American psyche. He urged us to include what were in effect racially biased questions in our research, to see just how far we could push people. The firm started testing questions about black people—whether they were capable of succeeding in America without the help of whites, for example, or whether they were genetically predetermined to fail. Bannon believed that the civil rights movement had limited “free thinking” in America. He was determined to liberate people by revealing what he saw as the forbidden truths about race.

What he meant was that white Democrats revealed their prejudices against minorities without realizing it. He posited that although these Democrats think that they like African Americans, they do not respect African Americans, and that many Democratic policies stemmed from an implicit acknowledgment that those people cannot help themselves.

When we think about racism, we often think of overt hatred. But racism can persist in different ways. Racism can be aversive, where a person consciously or subconsciously avoids a racial group (e.g., gated communities, sexual and romantic avoidance, etc.), and racism can be symbolic, where a person holds negative evaluations of a racial group (e.g., stereotypes, double standards, etc.). However, because the label “racism” can hold such social stigma in modern America, we found that white people often ignore or discount their internalized prejudices and react strongly to any inference that they hold such beliefs.

Bannon’s first request of our team was to study who felt oppressed by political correctness. Cambridge Analytica found that, because people often overestimate how much others notice them, spotlighting socially uncomfortable situations was an effective prime for eliciting bias in target cohorts, such as when you get in trouble for mispronouncing a foreign-sounding name. One of the most effective messages the firm tested was getting subjects to “imagine an America where you can’t pronounce anyone’s name.” Subjects would be shown a series of uncommon names and then asked, “How hard is it to pronounce this name? Can you recall a time where people were laughing at someone who messed up an ethnic name? Do some people use political correctness to make others feel dumb or to get ahead?”

This framing of political correctness as an identity threat catalyzed a “boomerang” effect in people where counternarratives would actually strengthen, not weaken, the prior bias or belief. This means that when targets would see clips containing criticism of racist statements by candidates or celebrities, this exposure would have the effect of further entrenching the target’s racialized views, rather than causing them to question those beliefs. In this way, if you could frame racialized views through the lens of identity prior to exposure to a counternarrative, that counternarrative would be interpreted as an attack on identity instead. What was so useful for Bannon was that it in effect inoculated target groups from counternarratives criticizing ethno-nationalism. It created a wicked reinforcement cycle in which the cohort would strengthen their racialized views when they were exposed to criticism. This may be in part because the area of the brain that is most highly activated when we process strongly held beliefs is the same area that is involved when we think about who we are and our identity. Later, when Donald Trump was aggressively criticized in the media for racist or misogynist statements, these critiques likely created a similar effect, where the criticism of Trump strengthened the resolve of supporters who would internalize the critique as a threat to their very identity.

What CA observed was that when respondents were angry, their need for complete and rational explanations was also significantly reduced. In particular, anger put people in a frame of mind in which they were more indiscriminately punitive, particularly to out-groups. They would also underestimate the risk of negative outcomes.

also identified that there were relationships between target attitudes and a psychological effect called the just-world hypothesis (JWH). This is a cognitive bias where some people rely on a presumption of a fair world: The world is a fair place where bad things “happen for a reason” or will be offset by some sort of “moral balancing” in the universe. We found that people who displayed the JWH bias were, for example, more prone to victim-blaming in hypothetical scenarios of sexual assault. If the world is fair, then random bad things should not happen to innocent people, and therefore there must have been a fault in the victim’s behavior. Finding ways to blame victims is psychologically prophylactic for some people because it helps them cope with anxiety induced by uncontrollable environmental threats while maintaining a comforting view that the world will still be fair to them.

CA then discovered that for those with evangelical worldviews in particular, a “just world” exists because God rewards people with success if they follow his rules. In other words, people who live good lives won’t get preexisting conditions, and they will succeed in life, even if they are black. Cambridge Analytica began feeding these cohorts narratives with an expanded religious valence. “God is fair and just, right? Wealthy people are blessed by God for a reason, right? Because He is fair. If minorities complain about receiving less, perhaps there is a reason—because He is fair. Or are you daring to question God?”

in some cases, the stronger the refugee claim, the harsher the responses. The targets were less and less concerned with hypothetical refugees and more concerned with maintaining the consistency of their worldview. If you are strongly invested in the idea that the world is just, evidence to the contrary can feel deeply threatening.

there were new characters arriving at Cambridge Analytica’s London office almost daily. The firm became a revolving door of foreign politicians, fixers, security agencies, and businessmen with their scantily clad private secretaries in tow. It was obvious that many of these men were associates of Russian oligarchs who wanted to influence a foreign government,

technical and analytical experience,” the memo read. “Our experience shows that in many cases utilizing social media or ‘foreign’ publications to ‘expose’ an opponent is often more effectual than using potentially biased local media channels.” The

Cambridge Analytica had connected with a man named Sam Patten, who had lived a colorful life as a political operative for hire all over the world. In the 1990s, Patten worked in the oil sector in Kazakhstan before moving into Eastern European politics. When CA hired him, he had just finished a project for pro-Russian political parties in Ukraine. At the time, he was working with a man named Konstantin Kilimnik, a former officer of Russia’s Main Intelligence Directorate (the GRU). Although Patten denies that he gave his Russian partner any data, it was later revealed that Paul Manafort, who was for several months Donald Trump’s campaign manager, did pass along voter polling data to Kilimnik in a separate instance. Patten and Kilimnik had met in Moscow in the early 2000s and later worked in Ukraine for Paul Manafort’s consultancy. The two became formal business partners soon after Patten was brought onto CA.

The memo outlined the Foreign Agents Registration Act and was emphatically clear: Foreign nationals are strictly prohibited from managing or influencing an American campaign or PAC at the local, state, or federal level.

But one of the psychologists on the team started coming to me to show me some of the new race projects. He showed me the master document of research questions that were being fielded in America, and my stomach dropped when I started reading. We were testing how to use cognitive biases as a gateway to move people’s perceptions of racial out-groups. We were using questions and images clearly designed to elicit racism in our subjects. As I watched a video of a man who was a participant in one of the field experiments, who’d been provoked by a CA researcher’s guided questioning into spasms of rage, racist insults flying from his mouth, I started to confront what I was helping to build.

the Russian Federal Security Service (FSB)—the successor to the Soviet KGB.

It’s eye-opening to summarize what was going on over those final months of my tenure. Our research was being seeded with questions about Putin and Russia. The head psychologist who had access to Facebook data was also working for a Russian-funded project in St. Petersburg, giving presentations in Russian and describing Cambridge Analytica’s efforts to build a psychological profiling database of American voters. We had Palantir executives coming in and out of the office. We had a major Russian company with ties to the FSB probing for information about our American data assets. We had Nix giving the Russians a presentation about how good we were at spreading fake news and rumors. And then there were the internal memos outlining how Cambridge Analytica was developing new hacking capacity in concert with former Russian intelligence officers.

In the year after Steve Bannon became vice president, Cambridge Analytica started deploying tactics that eerily foreshadowed what was still to come in the 2016 American presidential election. To get access to their opponent’s emails, Cambridge Analytica made use of hackers, some of whom may have been Russian, according to internal documents. The hacked emails CA procured were then used to undermine their opponent, including a concerted effort to leak rumors about the opposing candidate’s health. And this stolen kompromat was then combined with widespread online disinformation targeting social media networks. The overlap in events could be entirely coincidental, but many of the personnel who worked on Nigeria also worked on CA’s American operations.

The Russian government has its own domestic propaganda channels, but one of its global strategies is to cultivate pro-Russia assets in other countries. If you’re interested in spreading your narratives digitally, it’s helpful to have a roster of people to target who are more likely to support your country’s worldview. Using the Internet to cultivate local populations with Russian propaganda was an elegant way to bypass all Western notions of “national security.” In most Western countries, citizens have free speech rights—including the right to agree with a hostile nation’s propaganda.

To Moscow, civil rights and the First Amendment are the American political system’s most glaring vulnerabilities. And so the Russian state sought to exploit this vulnerability—to hack American democracy. It would work, they decided, because American democracy is an inherently flawed system. The Russians created their self-fulfilling prophecy of social chaos by targeting and domesticating their propaganda to American citizens of similar worldviews, who would then click, like, and share. These narratives spread through a system of constitutionally protected free speech, and the U.S. government did nothing to stop them. Neither did Facebook.

Nigel Farage, a former commodities trader and founding UKIP member, ousted the leader of the party. Farage became leader in 2006, and under his leadership, UKIP began stoking virulent anti-immigration sentiment among working-class whites and tapping into nostalgia for Britain’s imperial past in wealthy white communities. The world had been transformed by the September 11 attacks, the rise of Islamophobia, and the conflicts of the Bush and Blair years. As the fate of black and brown refugees developed into a European crisis, Cameron moved to appease nationalist sentiment to retain right-wing voters. The Conservative Party drew up a plan for a referendum, to be held before the end of 2017. The date was set for June 23, 2016.

College-educated city dwellers, accustomed to living among immigrants and working in businesses that benefit from their skilled labor, rejected right-wing fearmongering and generally supported Remain. Lower-income Britons and those who lived in rural areas or old industrial heartlands were much more likely to support Leave. National sovereignty has always been a core part of British identity, and the Leave campaign argued that EU membership was undermining that sovereignty. Remain supporters countered by pointing to economic, trade, and national security benefits in the status quo.

AIQ’s Trinidad contract with SCL included building infrastructure for Facebook data harvesting, clickstream data, ISP logs, and the reconciliation of IPs and user agents to home addresses, which would help de-anonymize Internet browsing data.

I realized something deeply sinister was happening in Britain. Even so, 72 percent of voters cast ballots. For hours, the vote was too close to call, but in the end, Leave emerged victorious with 51.89 percent of the vote.

After the referendum, Borwick revealed that Vote Leave and AIQ had together disseminated more than a hundred different ads with 1,433 different messages to their target voters in the weeks leading up to the referendum. Cummings later revealed that these ads were viewed more than 169 million times, but only targeted at a narrow segment of a few million voters, which resulted in their newsfeeds being dominated by Vote Leave messaging. The people of the United Kingdom were the targets of a scaled information operation deployed by AIQ, and the problem with Remain was that they completely failed to understand what they were up against. As Cambridge Analytica identified, provoking anger and indignation reduced the need for full rational explanations and would put voters into a more indiscriminately punitive mindset. CA found that not only did this anger immunize target voters to the notion that the economy would suffer, but some people would support the economy suffering if it meant that out-groups like metropolitan liberals or immigrants would suffer in the process—that, in effect, their vote would be used as a form of punishment.

Remain’s focus on the economy also neglected to stop and ask people what they thought the economy was in the first place. Cambridge Analytica identified that many people in non-urban regions or in lower socioeconomic strata often externalized the notion of “the economy” to something that only the wealthy and metropolitan participated in. “The economy” was not their job in a local store; it was something that bankers did. This is also what made certain groups comfortable with economic risks and even trade wars, since, in their minds, that chaos would be unleashed upon the people who worked in “the economy.” And the more forceful the economic argument they heard, the more confident they would become that what they were “actually” hearing was the fears of a cowering elite worried about losing its wealth. This made them feel powerful, and it would become a power they wanted to wield.

The world did not know it yet, but Brexit was a crime scene. Britain was the first victim of an operation Bannon had set in motion years before. The so-called “patriots” of the Brexit movement, with their loud calls to rescue British law and sovereignty from the grips of the faceless European Union, decided to win a vote by mocking those very laws. And to do so, they deployed a web of companies associated with Cambridge Analytica in foreign jurisdictions, away from the scrutiny of the agencies charged with protecting the integrity of our democracies. Foreshadowing what was to come in America, a clear pattern emerged during the Brexit debacle, where previously unknown foreign entities began exerting influence on domestic elections by deploying large data sets of unexplained origins. And with social media companies not performing any checks on the advertising campaigns spreading throughout their platforms, there was no one standing guard to stop hostile entities seeking to sow chaos and disrupt our democracies.

As Trump continued to gain ground, their curiosity grew. I described Cambridge Analytica’s tactics of voter manipulation—how the firm identified and targeted people with neurotic or conspiratorial predispositions, then disseminated propaganda designed to deepen and accentuate those traits. I explained how, after obtaining people’s data from Facebook, Cambridge Analytica could in some cases predict their behavior better than their own spouses could, and how the firm was using that information to, in effect, radicalize people inside the Republican Party.

A bunch of the people they were targeting were those who typically didn’t vote Republican or didn’t vote at all. They were trying to expand the electorate through this while at the same time they were committed to voter suppression. In particular, they focused on disengaging African Americans and other minority communities. One of the ways they did this was to peddle left-wing social justice rhetoric to depict Hillary Clinton as a propagator of white supremacy—while themselves working for a white supremacist. The aim was to move people from all demographics of a more left-wing ideology to vote for a third-party candidate, like Jill Stein.

Even the Facebook vice president seemed mostly unfazed. If I had a problem with Cambridge Analytica, he said, then I should create a rival firm—respond to the Uber of propaganda by developing the Lyft. This suggestion struck me as perverse—not to mention irresponsible—coming from an executive at a company well positioned to take meaningful action. But that was how Silicon Valley operated, I soon realized. The reaction to any problem, even one as serious as a threat to the integrity of our elections, is not “How can we fix it?” Rather, it’s “How can we monetize it?”

The Germans have an expression called Mauer im Kopf, which translates roughly as “the wall in the mind.” After the East and the West reunified in 1990, the legal border between the two Germanys was dissolved. The checkpoints disappeared, the barbed wire was ripped away, and the Berlin Wall was finally taken down. But even fifteen years after reunification, many Germans still overestimated the distances between cities in the East and the West. There was, it seems, a lingering psychological distance that betrayed the nation’s geography, creating a virtual border in people’s minds. Although that wall of concrete and steel had long since crumbled, its shadow lived on, etched into the psyches of the German people.

Just as worrisome, Bannon was in a position to help arrange for Cambridge Analytica to get contracts with the U.S. government. Cambridge Analytica’s parent company, SCL Group, was already working on projects for the U.S. State Department. This meant that CA could access the U.S. government’s data, and vice versa. To my horror, I realized that Bannon could be creating his own private intelligence apparatus. And he was doing it for an administration that didn’t trust the CIA, the FBI, or the NSA. It felt like I was living in a nightmare. Worse, it felt like Richard Nixon’s wet dream. Just imagine if Nixon had had access to that kind of intimate, granular data on every single American citizen. He wouldn’t have just fucked rats, he would have fucked the whole Constitution.

Cadwalladr listened as the source told her some wild stories about what SCL had done in Africa, Asia, and the Caribbean—honey traps, bribes, espionage, hackers, strange deaths in hotel rooms.