“You know DARPA funds some of their work,” Brent Clickard told me on a train ride from London to Cambridge. “If you want to expand your team, these are the ones you want.” As one of SCL’s psychologists, he floated between the company and his ongoing academic work inside one of the psychology labs at the University of Cambridge. Like me, Clickard was becoming enamored with the possibilities of what our research could yield, which is why he was so willing to provide the firm access to the world’s leading research psychologists. The psychology department at Cambridge had spearheaded several breakthroughs in using social-media data for psychological profiling, which in turn prompted interest from government research agencies. What Cambridge Analytica eventually became depended in large part on the academic research published at the university it was named after.
Cambridge Analytica was a company that took large amounts of data and used it to design and deliver targeted content capable of moving public opinion at scale. None of this is possible, though, without access to the psychological profiles of the target population—and this, it turned out, was surprisingly easy to acquire through Facebook, with Facebook’s loosely supervised permissioning procedures. The story of how this came to be started in my early days at SCL, before they created Cambridge Analytica as a spin-off American brand. Brent Clickard had shown me around the Psychometrics Centre at Cambridge. Having read through many of his papers, and those of his colleagues at the Centre, I was intrigued by their novel tactic of integrating machine learning with psychometric testing. It seemed like they were working on almost the same research questions we were at SCL, albeit with a slightly different purpose—or so I thought.
Research into using social data to infer the psychological disposition of individuals was published in some of psychology’s top academic journals, such as the Proceedings of the National Academy of Sciences (PNAS), Psychological Science, and the Journal of Personality and Social Psychology, among many others. The evidence was clear: The patterns of a social media user’s likes, status updates, groups, follows, and clicks all serve as discrete clues that could accurately reveal a person’s personality profile when compiled together. Facebook was frequently a supporter of this psychological research into its users and provided academic researchers with privileged access to its users’ private data. In 2012, Facebook filed for a U.S. patent for “Determining user personality characteristics from social networking system communications and characteristics.” Facebook’s patent application explained that its interest in psychological profiling was because “inferred personality characteristics are stored in connection with the user’s profile, and may be used for targeting, ranking, selecting versions of products, and various other purposes.” So while DARPA was interested in psychological profiling for military information operations, Facebook was interested in using it for increased sales of online advertising.
As we approached the Downing Site building, I spotted a small plaque that read PSYCHOLOGICAL LABORATORY. Inside, the air was stale, and the décor hadn’t been updated since at least the 1970s. We walked up a few flights of stairs and then to the last office at the end of a narrow corridor, where Clickard introduced me to Dr. Aleksandr Kogan, a professor at the University of Cambridge who specialized in computational modeling of psychological traits. Kogan was a boyish-looking man and dressed as awkwardly as his manner. He stood with a simpering grin in the middle of the room, which was filled with stacks of papers and random decorations from his time studying in Hong Kong.
At first, I had no idea about Kogan’s background, as he spoke English with a perfect American accent, albeit with an exaggerated prosody. I later learned he was born in the Moldavian SSR during the final years of the Soviet Union and spent part of his childhood in Moscow. Not long after the Soviet Union collapsed, in 1991, his family emigrated to the United States, where he studied at the University of California, Berkeley, before completing his Ph.D. in psychology in Hong Kong and joining the faculty at the University of Cambridge.
Clickard had introduced me to Kogan, as he knew the work he was doing in his lab at Cambridge could be extremely useful for SCL. But, knowing Nix’s preferred style of venue, Clickard decided that the introduction should happen over canapés and wine. Nix was fickle, and he could completely write people off because he didn’t like their tie or choice of restaurant. So we all met at a table booked by Clickard at an upstairs bar inside the Great Northern Hotel, beside Kings Cross train station. Kogan was visiting London for the day and had made time to tell us about his work before heading back to Cambridge. It was common enough for Nix to drink too much wine on a night out, but I’d never seen him intoxicated by a voice other than his own. The topic was social media.
“Facebook knows more about you than any other person in your life, even your wife,” Kogan told us.
Nix snapped out of his trance, reverting to his usual embarrassing self. “Sometimes it’s best wives don’t know certain details,” he quipped, sipping his wine. “Why would I ever need or want a computer to remind me—or her?”
“You might not want it,” the professor answered, “but advertisers do.”
“He’s interesting, but he doesn’t sound like a Cambridge man to me,” mumbled Nix, drinking more wine while Kogan was in the restroom.
“Because he’s not from Cambridge, Alexander. Jesus…He just teaches there!”
Clickard rolled his eyes. Nix was a distraction from more pressing concerns. After the firm looked at Kogan’s research, Nix was eager to put him to work. SCL had just secured the financing from Mercer and was in the process of setting up a new American entity. But before Nix was to let Kogan near his new prize project in America, he would have to prove himself in the Caribbean first. At the time, in early 2014, Kogan was working with researchers based at St. Petersburg State University on a psychological profiling project funded by the Russian state through a public research grant. Kogan advised a team in St. Petersburg that was pulling swaths of social media profile data and using it to analyze online trolling behavior. Given that this Russian social media research focused on maladaptive and antisocial traits, SCL thought it could be applied to the Trinidad project, as Ministry of National Security staff there were interested in experimenting with predictive modeling of Trinidadian citizens’ propensity to commit crimes.
In an email to Trinidad’s security ministry and its National Security Council about “criminal psychographic profiling via [data] intercepts,” one SCL staffer said that “we may want to either loop in or find out a bit more about the interesting work Alex Kogan has been doing for the Russians and see how / if it applies.”
Kogan eventually signed up to assist SCL on the Trinidad project, where he offered advice on how to model a set of psychological constructs that past research had identified as related to antisocial or deviant behavior. Kogan wanted data in exchange for helping to plan the project, and he started discussions with SCL about accessing its data set of 1.3 million Trinidadians for his own research. What I liked about Kogan was that he wanted to work fast and to get stuff done, which was not common for professors accustomed to the glacial pace of academic life. And he came across as honest, ambitious, and upfront, if a little bit naïve, in his excitement for ideas and intellectual ambition.
I got along quite well with Kogan in the beginning. He shared my interest in the emerging fields of computational psychology and computational sociology. We would talk for hours about the promise of behavioral simulation, and when we discussed SCL, he was palpably excited. At the same time, Kogan was slightly odd, and I noticed that his colleagues would make snide remarks about him when he wasn’t around. But it wasn’t as if this bothered me. If anything, it made me relate to him more—after all, I’d been on the receiving end of plenty of snide remarks myself. Besides, you had to be a bit weird to work at SCL.
When Kogan joined the Trinidad initiative in January 2014, we were just launching the early trial phases of the America project with Bannon. Based on our qualitative studies, we had some theories we wanted to test, but the available data was insufficient for psychological profiling. Consumer information—from sources like airline memberships, media companies, and big-box stores—didn’t produce a strong enough signal to predict the psychological attributes we were exploring. This wasn’t surprising, because shopping at Walmart, for example, doesn’t define who you are as a person. We could infer demographic or financial attributes, but not personality—extroverts and introverts both shop at Walmart, for example. We needed data sets that didn’t just cover a large percentage of the American population but also contained data that was significantly related to psychological attributes. We suspected we needed the kind of social data we had used on other projects in other parts of the world, such as clickstreams or the types of variables observed in a census record, which Kogan had picked up on.
Kogan started on Trinidad, but he was far more intrigued by SCL’s work in the United States. He told me that if he was brought on to the American job, we could work with his team at the Psychometrics Centre to fill gaps in the variables and data categories in order to create more reliable models. He started asking to access some of our data sets to see what might be missing in the training set, which is the sample data sets one uses to “train” a model to identify patterns. But that wasn’t quite the problem. Clickard told him that we’d done preliminary modeling and had training sets but that we needed data at scale. We couldn’t find data sets that contained variables that we knew helped predict for psychological traits and covered a wide population. It was becoming a major stumbling block. Kogan said that he could solve the problem for us—as long as he could use the data for his research too. When he said that if he was brought onto the America project, we could set up the first global institute for computational social psychology at the University of Cambridge, I was instantly on board. One of the challenges for social sciences like psychology, anthropology, and sociology is a relative lack of numerical data, since it’s extremely hard to measure and quantify the abstract cultural or social dynamics of an entire society. That is, unless you can throw a virtual clone of everyone into a computer and observe their dynamics. It felt like we were holding the keys to unlock a new way of studying society. How could I say no to that?
In the spring of 2014, Kogan introduced me to a couple of other professors at the Psychometrics Centre. Dr. David Stillwell and Dr. Michal Kosinski were working with a massive data set they’d harvested legally from Facebook. They were pioneers in social-media-enabled psychological profiling. In 2007, Stillwell set up an application called myPersonality, which offered users a personality profile for joining the app. After giving the user a result, the app would harvest the profile and store it for use in research.
The professors’ first paper on Facebook was published in 2012, and it quickly caught the attention of academics. After Kogan connected us, Kosinski and Stillwell told me about the huge Facebook data sets they’d acquired in their years of research. The U.S. military’s research agency, DARPA, was one of the funders of their research, they said, making them well suited to work with a military contractor. Stillwell was typically muted in our interactions, but Kosinski was clearly ambitious and tended to push Stillwell into keeping the conversation going. Kosinski knew this data could be extremely valuable, but he needed Stillwell to agree to any data transfers.
“How did you get it?” I asked.
They told me, essentially, that Facebook simply let them take it, through apps the professors had created. Facebook wants people to do research on its platform. The more it learns about its users, the more it can monetize them. It became clear when they explained how they collected data that Facebook’s permissions and controls were incredibly lax. When a person used their app, Stillwell and Kosinski could receive not only that person’s Facebook data, but the data of all of their friends as well. Facebook did not require express consent for apps to collect data from an app user’s friends, as it viewed being a user of Facebook as enough consent to take their data—even if the friends had no idea the app was harvesting their private data. The average Facebook user has somewhere between 150 and 300 friends. My mind turned to Bannon and Mercer, as I knew they would love this idea—and Nix would simply love that they loved it.
“Let me get this straight,” I said. “If I create a Facebook app, and a thousand people use it, I’ll get…like 150,000 profiles? Really? Facebook actually lets you do that?”
That’s right, they said. And if a couple million people downloaded the app, then we’d get 300 million profiles, minus the overlapping mutual friends. This would be an astonishingly huge data set. Up to that point, the largest data set I had worked on was Trinidad, which I thought was quite large, with profiles of one million people. But this set was on an entirely different level. In other countries, we had to get special access to data or spend months scraping and harvesting for populations several orders of magnitude smaller.
“So how do you get people to download this app?” I asked.
“We just pay them.”
“How much?”
“A dollar. Sometimes two.”
Now, remember, I’ve got a potential $20 million burning a hole in our firm’s pocket. And these profs have just told me that I can get tens of millions of Facebook profiles for…a million dollars, give or take. This was a no-brainer.
I asked Stillwell if I could run some tests on their data. I wanted to see if we could replicate our results from Trinidad, where we had access to similar types of Internet browsing data. If the Facebook profiles proved as valuable as I hoped, we would not only be able to fulfill Robert Mercer’s desire to create a powerful tool—what was even cooler was that we could mainstream a whole new field of academia: computational psychology. We were standing at the frontier of a new science of behavioral simulation and I was bursting with excitement at the prospect.
FACEBOOK LAUNCHED IN 2004 as a platform to connect students and peers in college. In a few years, the site grew to become the largest social network in the world—a place where almost everyone, even your parents, shared photos, posted innocuous status updates, and organized parties. On Facebook, you could “like” things—pages of brands or topics, along with the posts of friends. The purpose of liking was to allow users a chance to curate their personas and follow updates from their favorite brands, bands, or celebrities. Facebook considers this phenomenon of liking and sharing the basis of what it calls a “community.” Of course, it also considers this the basis of its revenue model, where advertisers can optimize their targeting using Facebook data. The site also launched an API (application programming interface) to allow users to join apps on Facebook, which would then ingest their profile data for a “better user experience.”
In the early 2010s, researchers quickly caught on that entire populations were organizing data about themselves in one place. A Facebook page contains data on “natural” behavior in the home environment, minus the fingerprints of a researcher. Every scroll is tracked, every movement is tracked, every like is tracked. It’s all there—nuance, interests, dislikes—and it’s all quantifiable. This means the data from Facebook has increasingly more ecological validity, in that it is not prompted by a researcher’s questions, which inevitably inject some kind of bias. In other words, many of the benefits of the passive qualitative observation traditionally used in anthropology or sociology could be maintained, but as many social and cultural interactions were now captured in digital data, we could add the benefits of generalizability one achieves in quantitative research. Previously, the only way one could have acquired such data would have been from your bank or phone company, which are strictly regulated to prevent access to that sort of private information. But unlike a bank or telecom company, social media operated with virtually no laws governing its access to extremely granular personal data.
Although many users tend to distinguish between what happens online from what happens IRL (in real life), the data that is generated from their use of social media—from posting reactions to the season finale of a show to liking photos from Saturday night out—is generated from life outside the Internet. In other words, Facebook data is IRL data. And it is only increasing as people live their lives more and more on their phones and on the Internet. This means that, for an analyst, there’s often no need to ask questions: You simply create algorithms that find discrete patterns in a user’s naturally occurring data. And once you do that, the system itself can reveal patterns in the data that you otherwise would have never noticed.
Facebook users curate themselves all in one place in a single data form. We don’t need to connect a million data sets; we don’t have to do complicated math to fill in missing data. The information is already in place, because everyone serves up their real-time autobiography, right there on the site. If you were creating a system from scratch to watch and study people, you couldn’t do much better than Facebook.
In fact, a 2015 study by Youyou, Kosinski, and Stillwell showed that, using Facebook likes, a computer model reigned supreme in predicting human behavior. With ten likes, the model predicted a person’s behavior more accurately than one of their co-workers. With 150 likes, better than a family member. And with 300 likes, the model knew the person better than their own spouse. This is in part because friends, colleagues, spouses, and parents typically see only part of your life, where your behavior is moderated by the context of that relationship. Your parents may never see how wild you can get at a 3 A.M. rave after dropping two hits of MDMA, and your friends may never see how reserved and deferential you are in the office with your boss. They all have slightly different impressions of who you are. But Facebook peers into your relationships, follows you around in your phone, and tracks what you click and buy on the Internet. This is how data from the site becomes more reflective of who you “really are” than the judgments of friends or family. In some respects, a computer model can know a person’s habits better than they even know themselves—a finding that compelled the researchers to add a warning. “Computers outpacing humans in personality judgment,” they wrote, “presents significant opportunities and challenges in the areas of psychological assessment, marketing, and privacy.”
With access to enough Facebook data, it would finally be possible to take the first stab at simulating society in silico. The implications were astonishing: You could, in theory, simulate a future society to create problems like ethnic tension or wealth disparity and watch how they play out. You could then backtrack and change inputs, to figure out how to mitigate those problems. In other words, you could actually start to model solutions to real-world issues, but inside a computer. For me, this whole idea of society as a game was super epic. I was obsessed with the idea of the institute that Kogan suggested to me, and became extremely eager to somehow make it happen. And it wasn’t just our pet obsession; professors all over were getting just as enthused. After meetings at Harvard, Kogan emailed me about their feedback, saying, “The operative term is game changing and revolutionizing social science.” And at first, it seemed like Stillwell and Kosinski were excited, too. Then Kogan let slip to them that CA had a budget of $20 million. And all the academic camaraderie ground immediately to a halt.
Kosinski sent Kogan an email saying they wanted half a million dollars up front, plus 50 percent of all “royalties” for the use of their Facebook data. We had not even proven this could work at scale in a field trial yet, and they were already demanding huge amounts of money. Nix told me to refuse, and this made Kogan panic that the project was going to fall apart before it even began. So the day after we rejected Kosinski’s demand for cash, Kogan said he could do it on his own, on his original terms—he would help us get the data, CA would pay for it at cost, and he would get to use it for his research. Kogan said he had access to more apps that had the same friends-collection permission from Facebook and that he could use those apps. I was immediately wary, thinking that Kogan might have just been planning to use Stillwell and Kosinski’s app under the radar. But Kogan insisted to me that he’d built his own. “Okay,” I said. “Prove it. Give me a dump of data.” To make sure these were not just pulled from the other app, we gave Kogan $10,000 to pilot his new app with a new data set. He agreed and did not ask for any money for himself, so long as he could keep a copy of the data.
Although he never told me this at the time, Kosinski has since said that he intended to give the money from the licensing of the Facebook data to the University of Cambridge. However, the University of Cambridge also strongly denies that it was involved with any Facebook data projects, so it is unclear that the university was aware of this potential financial arrangement, or would have accepted the funds if offered.
The following week, Kogan sent SCL tens of thousands of Facebook profiles, and we did some tests to make sure the data was as valuable as we’d hoped. It was even better. It contained complete profiles of tens of thousands of users—name, gender, age, location, status updates, likes, friends—everything. Kogan said his Facebook app could even pull private messages. “Okay,” I told him. “Let’s go.”
WHEN I STARTED WORKING with Kogan, we were eager to set up an institute that would warehouse the Facebook, clickstream, and consumer data we were collecting for use by psychologists, anthropologists, sociologists, data scientists—any academics who were interested. Much to the delight of my fashion professors at UAL, Kogan even let me add several clothing-style and aesthetic items that I could test for my Ph.D. research. We planned to go to different universities around the world, continuing to build up the data set so we could then start modeling things in the social sciences. After some professors at Harvard Medical School suggested we could access millions of their patient genetic profiles, even I was surprised at how this idea was evolving. Imagine the power, Kogan told me, of a database that linked up a person’s live digital behavior with a database of their genes. Kogan was excited—with genetic data, we could run powerful experiments unpacking the nature-vs.-nurture debate. We knew we were on the cusp of something big.
We got our first batch of data through a micro-task site called Amazon MTurk. Originally, Amazon built MTurk as an internal tool to support an image-recognition project. Because the company needed to train algorithms to recognize photographs, the first step was to have humans label them manually, so the AI would have a set of correctly identified photos to learn from. Amazon offered to pay a penny for each label, and thousands of people signed up to do the work.
Seeing a business opportunity, Amazon spun out MTurk as a product in 2005, calling it “artificial artificial intelligence.” Now other companies could pay to access people who, in their spare time, were willing to do micro-tasks—such as typing in scans of receipts or identifying photographs—for small amounts of money. It was humans doing the work of machines, and even the name MTurk played on this. MTurk was short for “Mechanical Turk,” an eighteenth-century chess-playing “machine” that had amazed crowds but was actually a small man hiding in a box, manipulating the chess pieces through specially constructed levers.
Psychologists and university researchers soon discovered that MTurk was a great way to leverage large numbers of people to fill out personality tests. Rather than have to scrounge for undergraduates willing to take surveys, which never gave a truly representative sample anyway, researchers could draw all kinds of people from all over the world. They would invite MTurk members to take a one-minute test, paying them a small fee to do so. At the end of the session, there would be a payment code, which the person could input on their Amazon page, and Amazon would transfer payment into the person’s account.
Kogan’s app worked in concert with MTurk: A person would agree to take a test in exchange for a small payment. But in order to get paid, they would have to download Kogan’s app on Facebook and input a special code. The app, in turn, would take all the responses from the survey and put those into one table. It would then pull all of the user’s Facebook data and put it into a second table. And then it would pull all the data for all the person’s Facebook friends and put that into another table.
Users would fill out a wide battery of psychometric inventories, but it always started with a peer-reviewed and internationally validated personality measure called the IPIP NEO-PI, which presented hundreds of items, like “I keep others at a distance,” “I enjoy hearing new ideas,” and “I act without thinking.” When these responses were combined with Facebook likes, reliable inferences could then be made. For example, extroverts were more likely to like electronic music and people scoring higher in openness were more likely to like fantasy films, whereas more neurotic people would like pages such as “I hate it when my parents look at my phone.” But it wasn’t simply personality traits we could infer. Perhaps not surprisingly, American men on Facebook who liked Britney Spears, MAC Cosmetics, or Lady Gaga were slightly more likely to be gay. 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 consumer data, then powerful predictions could be made. Once the profiling algorithm was trained and validated, it would then be turned onto the database of Facebook friends. Although we did not have surveys for the friend profiles, we had access to their likes page, which meant that the algorithm could ingest the data and infer how they likely would have responded to each question if they had taken a survey.
As the project grew over the summer, more constructs were explored, and Kogan’s suggestions began to match exactly what Bannon wanted. Kogan outlined that we should begin examining people’s life satisfaction, fair-mindedness (fair or suspicious of others), and a construct called “sensational and extreme interests,” which has been used increasingly in forensic psychology to understand deviant behavior. This included “militarism” (guns and shooting, martial arts, crossbows, knives), “violent occultism” (drugs, black magic, paganism), “intellectual activities” (singing and making music, foreign travel, the environment), “occult credulousness” (the paranormal, flying saucers), and “wholesome interests” (camping, gardening, hiking). My personal favorite was a five-point scale for “belief in star signs,” which several of the gays in the office joked we should spin off into an “astrological compatibility” feature and link it to the gay dating app Grindr.
Using Kogan’s app, we would not only get a training set that gave us the ability to create a really good algorithm—because the data was so rich, dense, and meaningful—but we also got the extra benefit of hundreds of additional friend profiles. All for $1 to $2 per app install. We finished the first round of harvesting with money left over. In management, they always say there is a golden rule for running any project: You can get a project done cheap, fast, or well. But the catch is you can choose only two, because you’ll never get all three. For the first time in my life, I saw that rule totally broken—because the Facebook app Kogan created was faster, better, and cheaper than anything I could have imagined.
THE LAUNCH WAS PLANNED for June 2014. I remember it was hot: Even though the summer was coming, Nix kept the air-conditioning off to lower the office bills. We had spent several weeks calibrating everything, making sure the app worked, that it would pull in the right data, and that everything matched when it injected the data into the internal databases. One person’s response would, on average, produce the records of three hundred other people. Each of those people would have, say, a couple hundred likes that we could analyze. We needed to organize and track all of those likes. How many possible items, photos, links, and pages are there to like across all of Facebook? Trillions. A Facebook page for some random band in Oklahoma, for example, might have twenty-eight likes in the whole country, but it still counts as its own like in the feature set. A lot of things can go wrong with a project of such size and complexity, so we spent a lot of time testing the best way to process the data set for when it scaled. Once we were confident that everything worked, it was time to launch the project. We put $100,000 into the account to start recruiting people via MTurk, then waited.
We were standing by the computer, and Kogan was in Cambridge. Kogan launched the app, and someone said, “Yay.” With that, we were live.
At first, it was the most anticlimactic project launch in history. Nothing happened. Five, ten, fifteen minutes went by, and people started shuffling around in anticipation. “What the fuck is this?” Nix barked. “Why are we standing here?” But I knew that it would take a bit of time for people to see the survey on MTurk, fill it out, then install the app to get paid. Not long after Nix started complaining, we saw our first hit.
Then the flood came. We got our first record, then two, then twenty, then a hundred, then a thousand—all within seconds. Jucikas added a random beeping sound to a record counter, mostly because he knew Nix had a thing for stupid sound effects, and he found it amusing how easy it was to impress Nix with gimmicky tech clichés. Jucikas’s computer started going boop-boop-boop as the numbers went insane. The increments of zeroes just kept building, growing the tables at exponential rates as friend profiles were added to the database. This was exciting for everyone, but for the data scientists among us, it was like an injection of pure adrenaline.
Jucikas, our suave chief technology officer, grabbed a bottle of champagne. He was always full of bonhomie, the life of the party, and he made sure we had a case of champagne in the office at all times for just such occasions. He had grown up extremely poor on a farm in the waning days of the Lithuanian SSR, and over the years he had remade himself into a Cambridge elite, a dandy whose motto seemed to be live it up today, because tomorrow you might die. With Jucikas, everything was extreme and over the top. That’s why he’d bought for the office an antique saber from the Napoleonic Wars, which he now intended to use. Why open champagne the normal way when you can use a saber?
He grabbed a bottle of Perrier-Jouët Belle Epoque (his favorite), loosened the cage holding the cork, held the bottle at an angle, and elegantly swiped the saber down the side. The entire top snapped clean off, and champagne gushed out. We filled the flutes and toasted our success, enjoying the first of many bottles we would drink that night. Jucikas went on to explain that sabering champagne is not about brute force; it’s about studying the bottle and hitting the weakest spot with graceful precision. Done correctly, this requires very little pressure—you essentially let the bottle break itself. You hack the bottle’s design flaw.
WHEN MERCER FIRST MADE the investment, we assumed we had a couple of years to get the project fully running. But Bannon shot that notion down right away. “Have it ready by September,” he said. When I suggested that was too quick, he said, “I don’t care. We just gave you millions, and that’s your deadline. Figure it out.” The 2014 midterms were coming, and he wanted what he now started referring to as Project Ripon—named after the small town in Wisconsin where the Republican Party was formed—to be up and running. Many of us rolled our eyes at Bannon, who started to get weirder and weirder after the investment. But we thought we just had to placate his niche political obsessions to achieve our potential at creating something revolutionary in science. The ends would justify the means, we kept telling ourselves.
He started traveling to London more frequently, to check on our progress. One of those visits happened to be not long after we launched the app. 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.
“Give me another,” said Jucikas. And he did it again. And again. And by the third profile, Nix—who’d hardly been paying attention at all—suddenly sat up very straight.
“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.”
The light went on: This was the first time Nix truly understood what we were doing. He could not have been less interested in things like “data” and “algorithms,” but seeing actual people onscreen, knowing everything about them, had seized his imagination.
“Do we have their phone numbers?” Nix asked. I told him we did. And then, in one of those moments of weird brilliance he occasionally had, he reached for the speakerphone and asked for the number. As Jucikas relayed it to him, he punched in the number.
After a couple of rings, someone picked up. We heard a woman say “Hello?” and Nix, in his most posh accent, said, “Hello, ma’am. I’m terribly sorry to bother you, but I’m calling from the University of Cambridge. We are conducting a survey. Might I speak with Ms. Jenny Smith, please?” The woman confirmed that she was Jenny, and Nix started asking her questions based on what we knew from her data.
“Ms. Smith, I’d like to know, what is your opinion of the television show Game of Thrones?” Jenny raved about it—just as she had on Facebook. “Did you vote for Mitt Romney in the last election?” Jenny confirmed that she had. Nix asked whether her kids went to such-and-such elementary school, and Jenny confirmed that, too. When I looked over at Bannon, he had a huge grin on his face.
After Nix hung up with Jenny, Bannon said, “Let me do one!” We went around the room, all of us taking a turn. It was surreal to think that these people were sitting in their kitchen in Iowa or Oklahoma or Indiana, talking to a bunch of guys in London who were looking at satellite pictures of where they lived, family photos, all of their personal information. Looking back, it’s crazy to think that Bannon—who then was a total unknown, still more than a year away from gaining infamy as an adviser to Donald Trump—sat in our office calling random Americans to ask them personal questions. And people were more than happy to answer him.
We had done it. We had reconstructed tens of millions of Americans in silico, with potentially hundreds of millions more to come. This was an epic moment. I was proud that we had created something so powerful. I felt sure it was something that people would be talking about for decades.