Can anyone stop Facebook from poisoning Democracy?
Freedom (to lie) isn’t free. For the campaign to re-elect Donald Trump it costs $1.5 million a week. And that’s just last week’s tab.
In exchange, Trump’s team gets to spread false and misleading political ads across Facebook’s social network. This includes a 30-second ad that makes the unsubstantiated claim that former Vice President Biden used his influence to block an investigation of a Ukrainian energy company with ties to his son. CNN rejected the ad, noting it contained inaccuracies. But it passed muster with Facebook, Google, Twitter and YouTube, where it’s been viewed hundreds of thousands of times.
The 2020 election cycle is shaping up to be very lucrative for all the leading tech platforms. Digital political-ad spending is expected to top $2.7 billion before Election Day, according to industry forecasts. Presidential candidates have already spent more than $70 million on Facebook and Google placements, according to data firm Acronym.
Facebook flip flop
All of this cash is great news for online platforms, which are going out of their way to accommodate the rush of campaign ads.
Last month, Facebook changed its policies against misleading ads to give politicians and their campaigns free rein to lie in ads posted on the massive social network. Why bother with fact-checking when there’s easy money in deceptive attack ads?
Under most circumstances, the FCC requires television and radio broadcasters to air political ads by federal candidates; they can’t make decisions based on the content of those candidate ads, whether they’re truthful or not.
Like TV and radio stations, online platforms have no obligation to truth-test political advertising. Yet Facebook once promised to do more than that, pledging that politicians must abide by the same standards as every other advertiser on its site, including the rule that “prohibits ads that include claims debunked by third-party fact checkers.”
That’s changed, and Facebook is painting itself as a champion of First-Amendment freedoms to justify the switch.
“Our approach is grounded in Facebook’s fundamental belief in free expression, respect for the democratic process, and the belief that, in mature democracies with a free press, political speech is already arguably the most scrutinized speech there is,” Facebook’s Katie Harbath explained to the Biden campaign after the company refused to take down the misleading Trump 2020 ad featuring Biden’s son.
But Facebook’s laissez-faire attitude toward political speech has more to do with politics and profits than principle. And it differs dramatically from an earlier commitment by founder Mark Zuckerberg, who told the Senate last year that Facebook wasn't doing enough to prevent its technology “from being used for harm.”
“It's not enough to just give people a voice,” he said. “We need to make sure that people aren't using it to harm other people or to spread misinformation.”
To that end Facebook has deployed A.I. tools that Zuckerberg claims do a better job of identifying accounts that may be trying to interfere in elections. He also pledged to put people first: “My top priority has always been our social mission of connecting people, building community and bringing the world closer together. Advertisers and developers will never take priority over that.”
Until they do, that is.
Facebook’s anti-democracy bias
Zuck’s words aren’t enough for Sen. Elizabeth Warren, who has some of her own: “Facebook has incredible power to affect elections and our national debate,” the presidential candidate tweeted last Monday, noting that the company is taking “deliberate steps to help one candidate intentionally mislead the American people.”
Facebook’s policy change followed a White House meeting between Zuckerberg and Trump in late September. “What did they talk about?” Warren asks. Zuckerberg’s press office won’t say; neither will the White House.
In the meantime, that new and improved A.I. is blocking legitimate news.
Investigative journalist Judd Legum, who first uncovered Facebook’s policy reversal on political ads, last Tuesday reported that Facebook had buried newsworthy posts from a public-radio station in Juneau, Alaska. Facebook claimed that the headline to a post titled “Confused about Juneau's municipal ballot measures? You're not alone,” was clickbait, which it forbids.
Last year, Facebook blocked news reports about a hack of its users’ data. Many users found that they couldn’t share stories from legitimate news outlets including the Guardian, Sacramento Bee, and Associated Press.
Despite the chorus from the right claiming anti-conservative bias at the social network, Facebook has a disturbing habit of content-moderation decisions that disproportionately go against social-justice advocates and vulnerable populations. Last year, it blocked an event page for organizers planning a counter-march against a white-supremacist rally that took place in Washington, D.C. The company also removed a post about racism from a prominent Black Lives Matter activist and journalist, though Facebook later apologized and rescinded the block.
The platform has also helped advertisers discriminate. A 2016 ProPublica investigation found that a Facebook ad-targeting tool allowed realtors to restrict who could see housing-related ads based on their “Ethnic Affinities.”
ProPublica later reported that advertisers could use Facebook’s tools to target ads based on people’s self-reported jobs, even if the job was “Jew hater.”
Facebook claims to have subsequently tweaked its tech to prevent advertisers from targeting groups or individuals in ways that could be considered discriminatory. Still, it has no problem with political-advertising content that targets vulnerable communities.
According to research by Media Matters, the company allowed the Trump campaign’s Facebook page to run more than 2,000 ads describing immigrants of color — particularly people from Central and South America — as an “invasion,” a term that white-supremacist groups in the United States frequently use in this context.
Poisoning the well
If all of this is leaving you angry and confused, you’re not alone.
Facebook’s promises to do better typically amount to little in practice. If things continue at this rate, one thing is certain: Facebook will play a significant role in poisoning the well of our democracy… again.
Sen. Warren has called for a new round of congressional hearings and an investigation so Facebook executives can explain what they’re doing to prevent a repeat of 2016 election interference and disinformation. But it’s difficult to see how this might result in new guidelines to curtail deceptive ads in time for it to matter. And existing legislative proposals, like Sen. Amy Klobuchar’s worthy Honest Ads Act, focus on disclosure of funders and preventing foreign meddling in U.S. elections. They would do little to stem the flood of misleading political ads coming from known, homegrown candidates themselves.
To date, Congress has advanced no privacy legislation that protects voters from having their online data harvested without their consent for targeting by dishonest political campaigns.
Unless Warren or others plan to introduce new legislation, containing provisions that no one has articulated clearly to this point, there seems very little of legal substance for Congress to discuss at an upcoming hearing with Facebook — certainly nothing that that would effectively bar politicians from peddling lies online.
And while Facebook wraps itself in the banner of free expression and political speech, it serves more as a scrim to hide its continued spread of political disinformation to tens of millions of voters in advance of the 2020 elections.
Those naive enough to expect Facebook to do the right thing are ignoring the company’s long history of doing the exact opposite. Facebook now boasts nearly 2.5 billion active users; these users generate four new petabytes of data every day and click “like” 4 million times every minute. Facebook has become too massive to govern itself, and politicians seem too overwhelmed to step into the maelstrom.
But whether and how lawmakers even can step in without running afoul of other cherished principles and legal protections is debatable.
Beyond fixing Facebook
The only real solution to Facebook is one that billions of don’t seem ready to commit to: leaving the social platform altogether.
In the interim, we must consider ways to help create its alternative: content-sharing platforms that don’t hammer at the foundations of democracy.
Earlier this year, Craig Aaron and I proposed a tax on the targeted-advertising revenues that are the economic engine fueling the most dominant online platforms. Proceeds from this tax, estimated at $2 billion a year, would go to a “public-interest media endowment” to support local, independent and noncommercial journalism. Additional funds would help incubate new ideas for gathering and distributing news content, preferably those that don’t rely on the clickbait economy and political-advertising budgets to survive.
All of us must think beyond fixing Facebook to a social-media solution of, by and for the people — not one that protects the powerful and spreads their lies. What we accomplish by the end of 2020 will help determine whether the future of our democracy brightens or dims, and whether our media system becomes an agent of civic engagement or alienation.
To do this requires a dramatic change in the way online platforms work today. Silicon Valley is incapable of cleaning up its toxic mess. More democratic media must help improve political participation, deepen discourse and increase voter turnout. Transparency about political ads is a key element of this shift, but so are efforts to return working journalists to civic-minded beats and limit the damaging effects of disinformation.
These changes must be an intrinsic part of the broader movement to restore democratic norms. A part of that is changing business as usual at platforms that profit from deception.