Data mining on social media platforms is hardly anything new. For the better part of a decade, it happened with reckless abandon and nary a soul made a peep — because it paid well, of course, and this is America. But this time…..this time it helped Trump. And that, my friends, makes it a scandal.
Facing growing death threats from pitchfork-wielding coochie-cappers, Facebook CEO Mark Zuckerberg finally took a break from banning my account to address the escalating controversy over how his social media giant handles people’s private information.
Zuckerberg took responsibility for what he called “a breach of trust between Facebook and the people who share their data with us and expect us to protect it.”
Translation: We betrayed the Democrats; it won’t happen again.
The Facebook nerd (or intellectual property thief, to be more accurate) outlined a three-point plan: investigate all third-party apps with access to large amounts of information, further restrict third-party access to prevent future abuses and make sure users understand which apps they’ve given access to their information.
“I’m serious about doing what it takes to protect our community,” he said in a Facebook post.
He stopped short of apologizing and didn’t answer some of the questions raised since the news broke last Friday that Cambridge Analytica pilfered sensitive information from tens of millions of Facebook users. He didn’t answer why Facebook didn’t reveal the breach when it first learned about it in 2015.
So who is Cambridge Analytica?
I’ll try to make this as concise as possible; a tall feat for tech explanations.
Before there was Cambridge Analytica, there was the Strategic Communication Laboratories Group — SCL Group, for short. Founded in 1993 by a British ad man named Nigel Oakes, it’s basically a messaging and PR firm for gubmints. They also do work for politicians and militaries around the world. Its clients included governments and politicians in Indonesia, Thailand, Kenya, the UK, and elsewhere.
SCL tends to describe its capabilities in grandiose language, touting its expertise at ”psychological warfare” and “influence operations.” It’s long claimed that its sophisticated understanding of human psychology helps it target and persuade people of its clients’ preferred message. Lately, its preferred buzzwords have focused on “big data” and “psychographic profiling.”
Color me skeptical.
Around 2013, word of SCL’s work reached the ears of one Sloppy Steve Bannon — and he liked what he heard. Bannon was then running Breitbart News, but perhaps more importantly, he was a trusted political adviser to Robert and Rebekah Mercer, the father-daughter pair of Republican mega-donors. Bannon connected SCL with the Mercers, and the British firm pitched a major expansion of its operation into US politics.
SCL claimed to be able to do a new type of political targeting that would be based on modeling individuals’ personalities. This was groundbreaking because political firms traditionally relied on simple demographic traits like age, race, or gender. It would do so based on survey questions that measure the Big Five or “OCEAN” personality traits commonly used in psychology research — Openness, Conscientiousness, Extroversion, Agreeableness, and Neuroticism. (You can take the test here, if you’re into that kind of thing.)
It would also use modeling to extrapolate the personalities of people it didn’t directly survey. Then, SCL claimed, it could tailor messaging to better target and persuade voters based on their particular personalities. They pitched this personality-based model as the future of politics.
The ever-generous Mercers agreed to provide $15 million for the project. To do so, they set up a new company mostly owned by Robert Mercer, with SCL Group executive Alexander Nix as the CEO. It would come to be known as Cambridge Analytica (a name reportedly provided by Bannon, who would also take a stake in the company and serve as a vice president).
At the core of this new effort would be a vast amount of data that an outside researcher harvested from tens of millions of people’s Facebook profiles — an effort Cambridge Analytica reportedly funded to the tune of $7 million. The data was matched with other records to construct personality profiles on millions of American voters, in hopes that Cambridge could better target them with customized political messaging.
After working on various low-profile GOP political efforts in 2014, Cambridge landed its first major gig: a Mercer-funded Super PAC supporting the presidential candidacy of Complyin’ Ted Cruz. But the gig, like most everything related to the Cruz campaign, went poorly. The Cruz team claimed Cambridge’s data was worthless and fought with them over money before ultimately severing ties.
Eventually, though, #TheDonald would emerge as the GOP nominee-in-waiting, and Cambridge would encounter a more forward-thinking client as a result. Though Bannon was still a few months away from officially joining the Trump campaign, he made the introduction between Cambridge and the Trump team. I’m sure it helped that Bannon stood to profit handsomely from a Trump/Cambridge partnership, #ButTheseAreMinorDetails.
That’s nice, Trey, but how did Cambridge get my friggin data already?
Through apps, of course. And the generosity of the American public.
A Cambridge University academic named Aleksandr Kogan got permission from Facebook to pull data via an app he created — but he reportedly claimed he’d use this data only for academic purposes, not commercial ones.
The app, titled “thisismydigitallife,” was a personality quiz Facebook users could take. However, to take the quiz, users had to consent to give the app access to their and their friends’ Facebook profiles.
More than 270,000 people used the app and took the quiz. However, because they consented to give the app access to their friends’ profiles, too, Kogan ended up collecting data from far more — 50 million raw profiles, of which about 30 million could be matched with other records that helped identify people. It’s pretty simple. You get a few people’s info, then their friends’ info, then their friends’ friends’ info, and pretty soon you’ve got a mountain of data from which to tailor political ads.
Facebook says it learned that Kogan had violated its rules by passing on the information to third parties in 2015 and that it subsequently got assurances from Kogan and Cambridge that they deleted the data they’d collected. However, a former Cambridge employee says that “he had recently seen hundreds of gigabytes” of that data still on the company’s servers.
Even though Facebook claims that the data was not to be used for commercial purposes, it’s not clear whether any laws were broken here. It appeared to be little more than a gentleman’s agreement, since the terms of the app were ultimately agreed to by the downloader. Nevertheless, regulators in the US and the UK both say they’re looking into the matter.
For their part, the Trump campaign says their reliance on Cambridge Analytica data is simply a myth, asserting that the bulk of their political data came courtesy of the RNC. This is a more than plausible claim, since the RNC, fresh off a data drubbing by the Obama campaign 4 years earlier, invested heavily in voter data for the 2016 election and beyond. It wouldn’t make much sense for the Trump camp to rely on outside data help when they could get the same information from the RNC for free.
Despite their MK Ultra-like claims, Cambridge’s work for Trump resembled less sexy, more traditional political consulting. More specifically:
- Oversaw a $5 million placement of TV ads — though it’s reported that the Trump campaign was unhappy with the results, since Cambridge ended up paying for ads on cable channels in Washington, DC. Very inefficient spending.
- Worked on targeting digital ads and online fundraising.
- Did polling of swing states.
For all the outlandish language surrounding Cambridge Analytica, the truth seems to be much more…..meh.
What (most of) the media won’t tell you:
As noted earlier, the use of social media data for political purposes is hardly anything new. In fact, our buddy Barry Soetoro led the way in this field, which is what compelled the RNC to get in on the fun.
n 2012, Team Barry encouraged supporters to download an Obama 2012 Facebook app that, when activated, let the campaign collect Facebook data both on users and their friends. Gee, that sounds EXACTLY like the “scandal” we’re discussing today!
When you installed the app, it said it would grab information about friends: their birth dates, locations, and ‘likes.’ The campaign boasted that more than a million people downloaded the app, which, given an average friend-list size of 190, means that as many as 190 million people had at least some of their Facebook data vacuumed up by the Obama campaign — without their knowledge or consent.
And he didn’t even need a FISA warrant!
Good ol Zuckerberg didn’t make a peep about that, though. If anything, Facebook made it easy for Obama. A former campaign director, Carol Davidsen, tweeted that “Facebook was surprised we were able to suck out the whole social graph, but they didn’t stop us once they realized that was what we were doing.”
Of course they didn’t.
This Facebook treasure trove gave Obama an unprecedented ability to reach out to nonsupporters. More important, the campaign could deliver carefully targeted campaign messages disguised as messages from friends to millions of Facebook users. The campaign readily admitted that this subtle deception was key to their Facebook strategy.
“People don’t trust campaigns. They don’t even trust media organizations,” Teddy Goff, the Obama campaign’s digital director, said at the time. “Who do they trust? Their friends.”
According to a Time magazine account just after Obama won re-election, “the team blitzed the supporters who had signed up for the app with requests to share specific online content with specific friends simply by clicking a button.”
The effort was called a “game-changer” in the 2012 election, and the Obama campaign boasted that it was “the most groundbreaking piece of technology developed for the campaign.”
While you draw your comparisons between PDT and Barry, keep in mind that it wasn’t the Trump campaign that solicited the collection of the data. And, as I said, it didn’t use the data in the general election campaign.
Obama, in contrast, was collecting live data on active users right up until Election Day, and at a scale that dwarfed anything the Trump campaign could access. The Obama data-mining campaign was Cambridge Analytica on steroids.
More important, the vast majority of people involved in these data-mining operations had no idea they were participating. And in the case of Obama, they had no way of knowing that the Obama campaign material cluttering their feed wasn’t really just political urgings from their friends.
But when Obama did it in 2012, he was wildly celebrated and worshiped by fawning reporters in news stories for his mastery of Big Data, and his genius at mining it to get out the vote.
We were told then about how the campaign “won the race for voter data,” and how it “connected with young voters.” His data analytics gurus were treated as heroes.
When Obama was exploiting Facebook users to help win re-election, it was an act of political brilliance. When Trump attempted something similar, with unclear results, it’s an assault on democracy and further evidence that somehow he stole the election.
The mainstream media are either schizophrenic to the point of oblivion regarding their own hypocrisy, or simply shameless liars.
I tend to believe both are amply true.
Hi everyone, if you enjoyed this article and feel that I’ve earned a tip, I would greatly appreciate any help you can give. If you would like to give more than $3, simply change the number in the box to multiply the donation. If not, I still love you and keep up the good fight!