(ANDREW CABALLERO-REYNOLDS / Getty Contributor)

On Oct. 23, Facebook CEO Mark Zuckerberg was summoned to Capitol Hill, ostensibly to give details about the company’s Libra cryptocurrency. Instead, questions from Congress ran the gamut and included a headline-grabbing exchange with Democratic Rep. Alexandria Ocasio-Cortez about whether Facebook fact-checks its political ads.

Ocasio-Cortez asked whether she could pay to target predominantly black areas, using location and demographic information held by Facebook, to advertise the wrong election date. Zuckerberg said such an ad would be taken down.

Ocasio-Cortez then asked whether she could theoretically run ads falsely claiming that certain Republicans had voted for the Green New Deal. “Probably,” Zuckerberg responded.

“In most cases, in a democracy, I believe that people should be able to see for themselves what politicians that they may or may not vote for are saying and judge their character for themselves,” Zuckerberg said. In essence, Facebook will remove ads that might incite violence or supress turnout, but will not fact-check ads from politicians.

“So, you won’t take down lies or you will take down lies? I think that’s just a pretty simple yes or no.”

Complete exchange between @RepAOC @AOC and Mark Zuckerberg at today’s House Financial Services Cmte hearing.

Full video here: https://t.co/heT7Psnlp1 pic.twitter.com/0iiWtfU5gQ

— CSPAN (@cspan) October 23, 2019

This issue came up because the Trump campaign ran a digital ad last month, which falsely stated that “Joe Biden promised Ukraine $1 billion dollars if they fired the prosecutor investigating his son’s company.” Both Factcheck and Politifact had already said the claim was false, but because political ads are held to different standards, the ad remained on Facebook.

The criticisms with this policy are obvious. The electorate’s ability to fact-check information is a privilege, not a duty. Most people simply do not have the time. As we have seen before on Facebook, Instagram, and WhatsApp—from anti-vaccination content to dangerous misinformation—people can be easily taken in by fake news when they have more pressing daily responsibilities.

Since 2016, Facebook has collaborated with fact-checking organizations around the globe to flag and identify content that contains falsehoods. It does not use this service for political ads because it’s not an “appropriate role for us to referee political debates and prevent a politician’s speech from reaching its audience and being subject to public debate and scrutiny,” according to Nick Clegg, Facebook’s VP of Global Affairs and Communications.

Even if it did mark ads as false, it’s unclear whether that would have a huge impact. In 2017, the company began putting a large, red warning label underneath content it had deemed fake news. However, Facebook found that “putting a strong image, like a red flag, next to an article may actually entrench deeply held beliefs” rather than inform or change any minds.

And fact-checking really only scratches the surface of what’s posted to Facebook, which has 1.63 billion daily active users. UK fact-checking partner Full Fact said this summer that Facebook “need[s] to scale up the volume of content and speed of response.” Though Facebook is working on AI tools that could identify fake news at scale, Full Fact is doubtful that the data and rating systems required to train a machine learning algorithm would produce high-quality outputs.

Just Ban Everything?

Facebook could ban all political ads, as Twitter has done, but Facebook argues that such a move would “tilt the scales in favor of incumbent politicians and candidates with deep pockets.” In fact, it tried this in Washington state to avoid a campaign finance legal battle, but it “resulted in a tangle of uneven enforcement and confusing rules, making it a cautionary tale for what a poorly implemented ad ban might mean for the 2020 campaigns,” The Verge reports.

Facebook could ban microtargetting of political ads, and its former head of the elections integrity team has suggested that custom targeting tools allow for an unequal playing field in exchange for cash.

That was my reasoning (the “not”) when I took the job to head FB’s elections integrity team for political ads. But custom targeting tools mean there is no level playing field. Profiting off of amplifying blatant lies, which are targeted to custom audiences, endangers democracy.

— Yael Eisenstat (@YaelEisenstat) October 30, 2019

Zuckerberg says “ads from politicians will be less than 0.5 percent of our revenue next year.”

No Power of Censorship

These challenges in the social media world are symptoms of a wider problem: political messages are not subject to fact checking at all.

Political advertisements are protected under the Communications Act, which states that broadcasters running political ads “shall have no power of censorship over the material broadcast.” This is underpinned by the First Amendment and the very notion of self-governance on which America is built. In theory, voters—given enough information—should be able to figure out who is most qualified to serve. As such, politicians are somewhat paradoxically allowed to lie to voters.

In 2014, Ohio passed a law making it illegal to broadcast “a false statement concerning the voting record” of a candidate. Nevertheless, when an anti-abortion group tried to put up billboards falsely claiming a congressman had voted for abortion funding by voting for President Obama’s Affordable Care Act, a federal District Court judge ruled against the law, reasoning that that “We do not want the government deciding what is political truth—for fear that the government might persecute those who criticize it. Instead, in a democracy, the voters should decide.”

Facebook point to the First Amendment too, and it’s what Zuckerberg leaned on when he said “we should err on the side of greater expression.” It’s also why he met with prominent conservatives like Tucker Carlson and Lindsey Graham; “hearing from a wide range of viewpoints is part of learning. If you haven’t tried it, I suggest you do!” he wrote. When asked whether Zuckerberg had met with any left-leaning personalities, Facebook did not say.

This issue is not unique to the US. Despite not having freedom of speech enshrined in the same way, political advertisements in the United Kingdom are also not subject to any standard of truth. Since 1999, the Advertising Standards Authority (ASA), the independent regulatory body for the advertising industry, has not been responsible for political ads because of concerns that the “independence of the system could be damaged by rulings for or against political parties.”

The UK, though, does not broadcast political ads on television; instead, parties make their case via party political broadcasts. But on social media, politicians on both sides of the Atlantic are playing the same game.

The Conservative Party, for example, put out an ad on Facebook claiming, inaccurately, that they were allocating double the amount of money to schools than they actually were. Following an investigation by Full Fact, Facebook pulled the ad—not because it was inaccurate, but because it violated Facebook’s Pages policy. At that point, it had been seen between 222,000 and 510,000 times.

It is obvious why political ads on Facebook should be held to a different standard than ads through traditional media, such as newspapers and broadcasts. Facebook content has the potential to reach millions of people in seconds, and can be tailored specifically to the person receiving it based on their location, interests, age, or gender. Moreover, we’ve already seen the impact of how the vast quantity of data collected by the social media site can be weaponized to influence elections. Facebook built the tools, and it should take some responsibility for the behavior of malicious individuals using them.

But ultimately, Facebook’s confusing arguments dig deep into the contradiction of free speech: it is there so that we may judge our politicians, but how beneficial is that to a democracy when politicians can use it as a shield for dishonesty? Facebook is not the problem when it is mirroring the philosophy of political ads, and political ads are not the problem when they are built upon this paradox. Yet it is a paradox that needs addressing—for the good of Facebook, politicians, and of democracy.