Skip to main content

Four ways the internet changed the midterm elections

Four ways the internet changed the midterm elections


Social networks continue to transform campaigns

Share this story

If you buy something from a Verge link, Vox Media may earn a commission. See our ethics statement.

Voters Across The Country Head To The Polls For The Midterm Elections
A couple walks into a polling center to vote in the midterm elections on November 6, 2018 in Provo, Utah. 
Photo by George Frey/Getty Images

This morning’s edition of The Interface was sent out to subscribers on Tuesday afternoon. As such, certain items included here will have been overtaken by events. We’ll have a more complete look at the midterm elections in the Wednesday edition of The Interface, which you can sign up to receive for free here.

A little over a year ago, I found myself growingly concerned with the unintended consequences of our biggest social platforms, particularly with respect to how they could be contributing to a global decline in democracy. I started writing about the subject every day in hopes that some good old-fashioned blogging could help bring shape to that discussion as we learned more. More than anything, I wondered what we would see in the months leading up to our US midterm elections. How would platforms transform in the wake of the 2016 disaster? And how would their adversaries adapt?

We won’t have a definitive answer to that question today. But what happens will go a long way in shaping the discussion around these platforms in the next two years. If Democrats manage to take back at least one house of Congress, I expect that concerns about the effectiveness about misinformation and online voter suppression efforts may relax somewhat as reporters turn their attention to the change in power. Facebook may face less scrutiny over the viral spread of fake news over the next two years — but more scrutiny from newly empowered Democrats, of the regulatory variety.

If on the other hand the Democrats fail, and the political status quo is maintained, journalists will have ample time to investigate everything that went wrong in the 2018 election, both online and off. As voters head the the polls today — and if you are a US citizen and haven’t voted, please proceed directly to your polling place — here are the biggest storylines I’m watching — in roughly descending order of how important they are.

Voter suppression. Bad people are trying to intimidate or trick people into not voting. Tony Romm has an overview of platforms’ efforts to thwart voter suppression in the Washington Post; Reuters reported that Twitter has deleted more than 10,000 accounts attempting to mislead people about what day the election is on. NBC’s Ben Collins finagled his way into a far-right Discord chat and found that trolls were complaining about Twitter deleting their accounts urging people to vote on the wrong day — but also found that they were growing more sophisticated in creating fake accounts in which they masquerade as middle-aged women by adding Snapchat filters to other people’s photos that they find online.

Most of the voter suppression efforts reported over the past week have targeted Democrats in an effort to persuade them not to vote. But in North Dakota, Jane Lytvynenko found that Democrats ran a misleading Facebook ad which targeted hunters in an apparent effort to suppress Republican turnout.

Misinformation and disinformation. Deceptive information about issues and candidates continued to thrive on Facebook, Twitter, and elsewhere. The New York Times’ Kevin Roose has been regularly tweeting the most viral stories and memes of the day, many of which are deceptive or outright hoaxes. In an overview of the paper’s findings, he writes that readers submitted more than 4,000 examples of misinformation over the course of the campaign.

Sarah Frier finds that Facebook has had some limited success in suppressing the reach of fake news, but that hyper-partisan content — which may prove to be even worse for the country in the long run — continues to thrive. And it’s not just Facebook, of course: trolls have lately come to favor LinkedIn for its more permissive rules and lax approach to enforcement, Craig Silverman reports.

Jonathan Albright finds that Facebook’s system for managing public pages allows foreign users to take over domestic pages and run ads. Arguing that the situation is ripe for election interference, Albright calls on Facebook to expand its reporting about managers and their geographic locations.

Infrastructure hacking. Efforts to disrupt voting machines and election websites are fully underway. In the Boston Globe, Jana Winter reported on what the Department of Homeland Security has detected so far. There have been more than 160 reports of hacking to date, with much of it believed to originate from Russia:

The hackers have targeted voter registration databases, election officials, and networks across the country, from counties in the Southwest to a city government in the Midwest, according to Department of Homeland Security election threat reports reviewed by the Globe. The agency says publicly all the recent attempts have been prevented or mitigated, but internal documents show hackers have had “limited success.”

The recent incidents, ranging from injections of malicious computer code to a massive number of bogus requests for voter registration forms, have not been publicly disclosed until now.

Meanwhile, late Monday, Facebook disclosed a potential new effort to disrupt the election. The company said it had taken down 30 Facebook accounts and 85 Instagram accounts taking part in “coordinated inauthentic behavior.” It’s not clear whether they were Russian in origin, but Facebook’s blog post took the unusual step of saying it would update its post if “these accounts are linked to the Russia-based Internet Research Agency or other foreign entities.”

Depending on what happens Tuesday, the United States could retaliate against Russia and has been preparing a digital counterattack, Zachary Fryer-Biggs reported.

Ad warfare. It wouldn’t be an election if there weren’t a lot of terrible ads. The 2018 election was only novel insofar is was the first time in modern history that major networks and Facebook had to pull an ad from a sitting president because it was too racist.

Facebook’s dark money issues persisted. An unknown group used Bernie Sanders’ image to urge Democrats to vote for the Green Party this year. Authentic sentiment or attempted voter suppression? You tell me. Meanwhile, the liberal activist organization MoveOn developed hundreds of ads featuring real voters and is blasting them at Facebook users with targeting techniques reminiscent of Cambridge Analytica. (The main difference being that they are doing this in accordance with Facebook’s rules and describing their methods publicly to Wired.)

But hey, don’t let the massive uncoordinated attack on democracy get you down too much. According to early Facebook backer Yuri Milner, it’s all much ado about nothing:

”I don’t agree that it’s an existential crisis, I think it’s a little bump in the road,” Milner said during an interview with Bloomberg TV. Facebook is an evolving system continuously responding to challenges and there will “always be bad actors who are testing different products,” he added.

I hope you’re right, Yuri. And if not, can I live in your underground panic bunker?

As for me, I’m spending Election Day on a (delayed!) flight to New York for an election event with Tech:NYC — say hi if you’re there. Meanwhile, Bloomberg is live-blogging the cyber-security aspects of the election — follow along, if you dare.


Facebook referred to EU watchdog over targeting, fake ads

Britain’s information watchdog asked Facebook’s lead European regulator to investigate how the company targets, monitors and shows ads to users, Kate Holton writes:

“We have uncovered a disturbing disregard for voters’ personal privacy,” Denham said. “Social media platforms, political parties, data brokers and credit reference agencies have started to question their own processes – sending ripples through the big data ecosystem.”

Facebook admits failings over incitement to violence in Myanmar

In an extremely sketchy move, Facebook used the eve of the election to dump a report it commissioned about its potential contributions to genocide in Myanmar. I haven’t yet read it — lot going on over here! — but this recap in the Guardian says it tells us largely what we heard from the United Nations earlier this year:

Facebook has admitted it did not do enough to prevent the incitement of violence and hate speech in Myanmar, after a report it commissioned concluded that it had become a platform for harmful and racially-inflammatory content.

The report by San Francisco-based nonprofit Business for Social Responsibility (BSR) found that, in Myanmar, “Facebook has become a means for those seeking to spread hate and cause harm, and posts have been linked to offline violence.”

Amazon, Facebook and Google are all being looked at for antitrust violations, Trump says

One of the president’s defining features, to my mind, is that in public remarks he is always saying that he is “looking at” every solution to every problem, simultaneously. As such, I don’t make too much of this latest suggestion that big tech platforms could face antitrust regulation.

Podcasts help extremists get their message out

Here’s a subject that hasn’t gotten nearly enough attention to date: how podcasts amplify extremism. A timely report from Sara Ashley O’Brien and Kaya Yurieff:

Extremists, spouting racist, anti-Semitic, homophobic, and anti-immigrant beliefs, have followed the trend and are broadcasting their messages as established social media outlets crack down on similar content.

Take LoveStreet, previously known as HateHouse. An episode labeled “HateHouse EP 18: Black people are disabled,” was posted on YouTube nine months ago. Just a few minutes in, one of the speakers says: “Black people look very suspect and they fit the profile for a lot of crimes committed” and “Police know that n****s are violent all the time.” It was removed on Friday after CNN Business flagged the video.

Weibo Gives Media, Government Power to Quash ‘Rumors’

Weibo now has a “rumor-busting” feature, which one imagines will also be used as a truth-busting feature by the Chinese censors it is empowering here:

Weibo is giving media outlets and the Chinese government the authority to directly flag online rumors, an account affiliated with the microblogging platform announcedFriday.

According to the statement posted by the Sina Administrative Microblog Academy — operated by Weibo’s parent company — verified accounts of government departments and “credible media” will be able to mark posts disseminating fake news with a customizable notice identifying them as untrue or misleading. The notice will then appear immediately below the post, with Weibo playing no role in screening or approving it.


Alex Jones banned from Facebook? His videos are still there — and so are his followers

It’s harder to de-platform someone than it looks, as Facebook is learning with Alex Jones and his fan. Here’s Craig Timberg:

In the three months since Facebook removed four of Jones’s pages over allegations of hate speech, the NewsWars page has remained intact and surged in posts and page views. The NewsWars Facebook page identifies, which Jones said his company operates, as the website associated with the page and lists it under “Contact Info.” Jones said he doesn’t run the Facebook page.

Videos hosted by the NewsWars Facebook page have totaled 3.9 million views since August, nearly reaching the monthly viewership of Jones’s videos on Infowars and other pages he controlled before they were shut down.

How Right-Wing Social Media Site Gab Got Back Online | WIRED

Speaking of the de-platformed being re-platformed, Gab found a new host:

The extremist-friendly social media site’s reappearance was made possible by two companies and the men behind them: digital security company Cloudflare, helmed by self-described “free speech absolutist” Matthew Prince, and domain registrar Epik, led by Rob Monster.

Hate Speech on Live ‘Super Chats’ Tests YouTube

YouTube is taking a cut of profits from some extremely problematic paid comments on right-wing live streams, Yoree Koh reports:

Mr. Ralph, whose channel had 22,500 subscribers, is one of several far-right YouTube celebrities who have used the Super Chat function to make money. Topics among such users can be wide-ranging, from events like the tragedy in Pittsburgh and the confirmation of Supreme Court Justice Brett Kavanaugh to critiques of the media and internal debates among members of the far-right online communities.

Most Super Chats generate a few hundred dollars in revenue, according to an analysis conducted for The Wall Street Journal, with YouTube typically collecting 30%, people familiar with the matter said.

HQ Trivia was a blockbuster hit — but internal turmoil and a shrinking audience have pushed it to the brink

Here’s a great Kurt Wagner story about how HQ replaced its CEO amid worries he wasn’t moving fast enough — only to have its next CEO the subject of a harassment complaint. (The company says the complaint is unfounded.)

How pro-Trump Facebook pages make a living.

April Glaser investigates the economics behind MAGA meme pages on Facebook. It’s another piece of evidence for the idea that while Facebook may have ratched down some of the misinformation in the News Feed, hyper-partisan content — which often is deceptive, if not downright false — continues to be profitable.

None of this will give comfort to anyone worried about the synergistic relationship between political polarization and Facebook engagement. Pages like the ones Ferretti, Keirns, and Murphy run appear to break none of the rules that govern the Facebook news feed, but their entire business is premised on the insight that partisan self-victimization and fear of immigrants and elites will lead to clicks, and therefore cash. All of this was still the case for this cycle’s election, and unless something about the incentives that drive Facebook engagement change dramatically, it will probably be the case for the next one too. 

Sen. Mark Warner on breaking up Facebook and Congress’s plan to regulate tech

Makena Kelly asks the Virginia senator whether he wants to make Facebook spin subsidiaries including Instagram and WhatsApp:

WARNER: I see breakup as more of a last resort. Also, we could look at companies more on a domestic scene. The caution is, if we were to start with the breakup of Facebook and Google, what would we do with Alibaba, Baidu, and Tencent? In a world-based economy, you can’t look at these only on a national basis. My fear is that the Chinese companies, which are frankly on a growth rate even faster than Facebook, Google, and Twitter’s growth, don’t have any of the constraints of the other companies.

Deepfake-busting apps can spot even a single pixel out of place

Here’s a reason to be optimistic that deepfakes will not completely dissolve the boundaries between truth and fiction, from Karen Hao:

There are two main ways to deal with the challenge of verifying images, explains Farid. The first is to look for modifications in an image. Image forensics experts use computational techniques to pick out whether any pixels or metadata seem altered. They can look for shadows or reflections that don’t follow the laws of physics, for example, or check how many times an image file has been compressed to determine whether it has been saved multiple times.

The second and newer method is to verify an image’s integrity the moment it is taken. This involves performing dozens of checks to make sure the photographer isn’t trying to spoof the device’s location data and time stamp. Do the camera’s coordinates, time zone, and altitude and nearby Wi-Fi networks all corroborate each other? Does the light in the image refract as it would for a three-dimensional scene? Or is someone taking a picture of another two-dimensional photo?

Facebook tops 3,000 Seattle-area employees as major hiring spree continues

Facebook is growing fast in Seattle, with headcount there increasing by 50 percent since March, Nat Levy reports.

’Shows for nobody’: Facebook Watch has moved away from early short-form video formats

My main takeaway from this Sahil Patel story about Facebook Watch is that the company still has basically no idea what kinds of video programming are going to thrive there over the long term.

TikTok surpassed Facebook, Instagram, Snapchat & YouTube in downloads last month

Facebook, Twitter, and Snapchat are no longer growing in the United States. But you know what is? China’s TikTok. My long-term goal is to get kicked off TikTok for making viral memes promoting democratic reforms in China.

Is Celery Juice Good for You?

A bunch of quacks with big Instagram followings began promoting the health benefits of “celery juice,” Amanda Mull reports.

A recent study out of the Cleveland Clinic bears out celery’s blood-pressure benefits, but the researchers recommend consuming full stalks instead of extracts in order to get maximum benefits. Young, too, mentioned that maybe celery might be better intact, as a snack or as an addition to a meal rather than as a medicine. “You don’t have to drink it; you can also chew it,” she says. “Whatever happened to chewing?”


Facebook launches retail pop-up stores ahead of holiday season

Facebook is dipping a toe into retail with pop-up shops in Macy’s promoting allegedly beloved brands. As Julia Alexander notes here, it’s most interesting as a first step toward maybe wanting some permanent locations to sell hardware.

The company announced today that it will open pop-ups with more than 100 brands in nine different Macy’s locations. The 100 brands’ products will be available for purchase through the pop-up shops, and they are some of the “most-loved brands” on Instagram and Facebook, according to a blog post on the company’s website. Some of those brands include Love Your Melon, an apparel line that helps raise awareness and fight against pediatric cancers, and Two Blind Brothers, a nonprofit clothing line that donates all proceeds to blindness research.

Instagram prototypes bully-proof moderated School Stories

Jane Manchun Wong found code suggesting Instagram plans to have school-specific public stories, a la Snapchat. The code suggests that posts will be “manually reviewed,” but the headline’s suggestion here that this will make them “bully-proof” strikes me as optimistic in the extreme.


Facebook Didn’t Sway the 2018 Midterms

Before any votes could be counted, Alexis Madrigal said Facebook was not a determining factor in the midterms. I’d argue that (1) it’s strange to argue that before we know the outcome; and (2) a list of ways Facebook appears to be influencing the election can be found in the lead item above.

Don’t Give Up on Snap ($)

Jessica Lessin says Snap is due for a comeback — eventually:

Twitter has 2.5 times Snap’s revenue and investors are currently valuing it at three times the market cap. Snap grew revenues 43% percent in the last quarter. Twitter grew 29% percent.

Of course, Twitter and Snap users are doing different things. Snap’s users are heavily into messaging while Twitter’s users are browsing tweets, articles and videos. The latter is much more appealing to advertisers—which partially explains why Twitter’s revenues are larger. But I believe there will be new ways to make money through messaging—ways Snap invents or can copy from others like Facebook over time.

And finally ...

Scientists say mysterious ‘Oumuamua’ object could be an alien spacecraft

Is it possible that aliens are coming to earth on Election Day to end our present simulation and restore us to our original reality? Well, here are a couple of Harvard researchers who are not ruling it out:

Scientists have been puzzling over Oumuamua ever since the mysterious space object was observed tumbling past the sun in late 2017. Given its high speed and its unusual trajectory, the reddish, stadium-sized whatever-it-is had clearly come from outside our solar system. But its flattened, elongated shape and the way it accelerated on its way through the solar system set it apart from conventional asteroids and comets.

Now a pair of Harvard researchers are raising the possibility that Oumuamua is an alien spacecraft. As they say in a paper to be published Nov. 12 in The Astrophysical Journal Letters, the object “may be a fully operational probe sent intentionally to Earth vicinity by an alien civilization.”

I, for one, welcome our new alien overlords, and remind them that I can be of use in drafting newsletters about their activities here on earth.

Talk to me

Send me tips, comments, questions, and proof that you voted: