Skip to main content

It turns out there really is an American social network censoring political speech

It turns out there really is an American social network censoring political speech


Et tu, TikTok?

Share this story

Illustration by Alex Castro / The Verge

For the past couple of years we have endured a lengthy, one-sided, and largely fruitless debate over “censorship” on our US social networks. The issue reached a low point in May, when the Trump White House put out a call for all Americans to report perceived incidents of censorship on Facebook, Twitter, Instagram, and YouTube. Nothing came of the exercise but a campaign rally disguised as “a social media summit,” along with a volley of red-meat headlines designed to comfort the comfortable. As right-wing meme lords fulminated in the White House about censorship, the Trump administration shut off comments on the live video stream and required participants to submit all questions in advance so they could be moderated.

Of course, American companies do remove content from social networks, including some political speech, when it violates their guidelines around hate speech, violence, or nudity. For the most part, though, they permit the maximum range of free expression. As businesses that only make money when we are paying attention to them, they are financially incentivized to include as many viewpoints on their platforms as possible, and to treat them all with relative equality. (It turns out that when you do this, partisan posts outperform centrist ones, and conservative content outperforms liberal content.)

But what if censorship of political speech on an American social network was real? What if you were prohibited from discussing the Trump impeachment, say, or the 2016 election? Well, let’s take a look at how America’s latest social networking sensation — TikTok, the product of Chinese company ByteDance — has handled politically sensitive content. From Alex Hern’s very good story in The Guardian:

TikTok, the popular Chinese-owned social network, instructs its moderators to censor videos that mention Tiananmen Square, Tibetan independence, or the banned religious group Falun Gong, according to leaked documents detailing the site’s moderation guidelines.

The documents, revealed by the Guardian for the first time, lay out how ByteDance, the Beijing-headquartered technology company that owns TikTok, is advancing Chinese foreign policy aims abroad through the app.

As Hern notes, suspicions about TikTok’s censorship are on the rise. Earlier this month, as protests raged, the Washington Post reported that a search for #hongkong turned up “playful selfies, food photos and singalongs, with barely a hint of unrest in sight.” In August, an Australian think tank called for regulators to look into the app amid evidence it was quashing videos about Hong Kong protests.

On the one hand, it’s no surprise that TikTok is censoring political speech. Censorship is a mandate for any Chinese internet company, and ByteDance has had multiple run-ins with the Communist party already. In one case, Chinese regulators ordered its news app Toutiao to shut down for 24 hours after discovering unspecified “inappropriate content.” In another case, they forced ByteDance to shutter a social app called Neihan Duanzi, which let people share jokes and videos. In the aftermath, the company’s founder apologized profusely — and pledged to hire 4,000 new censors, bringing the total to 10,000.

“Our product took the wrong path, and content appeared that was incommensurate with socialist core values,” Bytedance CEO and founder Zhang Yiming wrote on his official WeChat account.

”I am personally responsible for the punishments we have received,” he added.

For its part, TikTok told the Guardian that its story was based on now-outdated guidelines that are no longer enforced:

“The old guidelines in question are outdated and no longer in use. Today we take localised approaches, including local moderators, local content and moderation policies, local refinement of global policies, and more. We also consult with a number of independent local committees and are working to scale this at a global level, including forming an independent committee of leading industry organisations and experts to continually assess these policies.

As it happens, TikTok does have a publicly posted set of guidelines, which I read for the first time today. Notably absent is any policy on how to handle posts about politics. It remains unclear to me whether applying a more localized set of approaches to content moderation means that American users will be able to post content that Chinese users cannot on their own version of the app. (Maybe they can?) Here’s what TikTok had to say when I asked:

“We know users gravitate to TikTok because it provides a positive, joyful app experience that fosters their creativity. Fun, entertaining short-form videos are what users overwhelmingly upload and engage with, and that’s what we tend to provide the most support for through things like partnerships or creative filters. While political content is fine, it’s not our focus and it’s not what users are generally looking for.”

As the 2020 election campaign — or, uh, impeachment proceedings — ramp up, I can see that changing in a hurry. If TikTok’s ascent continues, politics will naturally appear there, just as it has everywhere else. With US-China tensions running high, the issue feels quite sensitive. We saw how aggressively our politicians responded to disingenuous complaints of censorship. Imagine what they might do when the censorship is real.

The Ratio

🔼 Trending up: Facebook showed off some legitimately innovative technology at the Oculus developer conference, including a VR headset that you can use without hand controllers.

🔼 Trending up: Google announced its releasing a database of deepfakes to help researchers develop AI detection techniques. Tech companies have been ramping up deepfake detection efforts since 2016, but researchers warn AI might not be the best solution. (Kyle Wiggers / VentureBeat)

🔽Trending down: Even if they’re outdated, TikTok’s content moderation guidelines confirmed a lot of fears about the company’s approach to free expression.

🔽Trending down: The U.S. Equal Employment Opportunity Commission is investigating seven companies — including Capital One and Enterprise Rent-A-Car — for allegedly restricting job postings on Facebook to people of certain ages or genders. (Josh Eidelson, Paige Smith and Jaclyn Diaz / Bloomberg)


The activist behind the California Consumer Privacy Act is working on a new ballot initiative for the 2020 election that would give California consumers more control over their personal data. Tony Romm at The Washington Post said it’s particularly focused on location data, health records and financial information:

Consumers would have to give their permission before such data could be sold, and they would gain the ability to block companies from monetizing those sensitive insights through targeted ads.

Mactaggart’s proposal also includes the creation of an agency in California to enforce privacy protections, along with tougher penalties for mishaps involving children under age 16. And it would require companies to demystify their secret algorithms when such software is used to profile a person, such as determining their employment prospects or their ability to obtain housing, credit cards, loans or other key services.

The power of Facebook’s Oversight Board will largely depend on its jurisdiction, membership, and the process people have to go through to submit cases for consideration, this researcher argues. (Evelyn Douek / Lawfare)

UK prime minister Boris Johnson warned that technology companies were building an unregulated surveillance state. He called on the UN to work together on a common set of global principles for emerging technologies. (Tom Warren / The Verge)

Russia’s state-sponsored hacks don’t typically share code with one another, indicating less coordination between Kremlin hackers than you might expect. When they do, it’s usually within groups managed by the same intelligence service, according to a report from Check Point and Intezer Labs. (Catalin Cimpanu / ZDNET)

The Ukrainians behind the “I Love America” Facebook page —a pro-Trump propaganda site with 1.1 million followers — revealed why they do what they do. (Spoiler: it’s for the money.) (Judd Legum / Popular Information)


Oculus Connect 6 kicked off in San Jose today with announcements of a cable that delivers Oculus Rift content from the PC to the Quest headset; coming support for hand-tracking control of the Quest, and a new virtual world to explore what the company calls Horizon. It also confirmed that it is working on augmented reality glasses and that it plans to build a live, real-time view of the entire world. Fun! Whatever you think of Facebook’s hardware efforts, it seems to me that the company is advancing faster — and talking about it more publicly — than any of its high-profile rivals in the space. (Where the hell has Magic Leap been this year?)

Ahead of the event, CNET talked to Mark Zuckerberg, who tried to sell them on the social graph as VR’s killer app.

“The thing that we care about is delivering human connection and helping people come together,” Zuckerberg said, leaning towards the theme of presence, and away from a model “that’s more just around, here’s your app, here’s your content, I’m gonna pull it from a store.”

Zuckerberg’s not-so-veiled knock toward competitors with competing headsets and app stores include may sound like typical Silicon Valley infighting. The competitors are many: Google with its Daydream headset projectValve’s Index headset, Microsoft’s Mixed Reality and even Apple’s nascent VR and AR efforts. But it’s also a sign of how serious Facebook is about backing VR and, soon, AR too.

David Marcus’ effort to make Libra happen has now entered its “writing a Medium post about it” phase.

Google said it won’t pay publishers in France to display snippets of their news stories. Instead, the company will change the way articles appear in search results. The EU’s new copyright regime allows publishers to request money from platforms like Google and Facebook. But, obviously, the platforms don’t have to comply. (Laura Kayali / Politico)

Twitter is letting users add up to five lists as alternative timelines in the main Twitter app, allowing people to swipe between different groups of accounts directly from their home screen. (Chaim Gartenberg / The Verge)

Snapchat is letting marketers make video ads a bit longer — three minutes as opposed to 10-seconds. The ads will still be mostly skippable. (Garett Sloane / AdAge)

And finally...

Hard day at the White House:

Our thanks, as always, to the Facebook ad archive.

Talk to us

Send us tips, comments, questions, and censored TikTok videos: and