Brian Contreras Los Angeles Times
WWR Article Summary (tl;dr) As Brian Contreras reports, "With a few quick taps, TikTokers can flag videos as falling into specific categories of prohibited content — misleading information, hate speech, pornography — and send them to the company for review. Given the immense scale of content that gets posted to the app, this crowdsourcing is an important weapon in TikTok's content moderation arsenal."
One hundred and forty-seven dollar signs fill the opening lines of the computer program. Rendered in an icy blue against a matte black background, each "$" has been carefully placed so that, all together, they spell out a name: "H4xton."
It's a signature of sorts, and not a subtle one. Actual code doesn't show up until a third of the way down the screen.
The purpose of that code: to send a surge of content violation reports to the moderators of the wildly popular short-form video app TikTok, with the intent of getting videos removed and their creators banned.
It's a practice called "mass reporting," and for would-be TikTok celebrities, it's the sort of thing that keeps you up at night.
As with many social media platforms, TikTok relies on users to report content they think violates the platform's rules. With a few quick taps, TikTokers can flag videos as falling into specific categories of prohibited content — misleading information, hate speech, pornography — and send them to the company for review. Given the immense scale of content that gets posted to the app, this crowdsourcing is an important weapon in TikTok's content moderation arsenal.
Mass reporting simply scales that process up. Rather than one person reporting a post to TikTok, multiple people all report it in concert or — as programs such as H4xton's purport to do — a single person uses automated scripts to send multiple reports.
H4xton, who described himself as a 14-year-old from Denmark, said he saw his "TikTok Reportation Bot" as a force for good. "I want to eliminate those who spread false information or … made fun of others," he said, citing QAnon and anti-vax conspiracy theories. (He declined to share his real name, saying he was concerned about being doxxed, or having personal information spread online; The Times was unable to independently confirm his identity.)
But the practice has become something of a boogeyman on TikTok, where having a video removed can mean losing a chance to go viral, build a brand or catch the eye of corporate sponsors. It's an especially frightening prospect because many TikTokers believe that mass reporting is effective even against posts that don't actually break the rules. If a video gets too many reports, they worry, TikTok will remove it, regardless of whether those reports were fair.
It's a very 2021 thing to fear. The policing of user-generated internet content has emerged as a hot-button issue in the age of social-mediated connectivity, pitting free speech proponents against those who seek to protect internet users from digital toxicity. Spurred by concerns about misinformation and extremism — as well as events such as the Jan. 6 insurrection — many Democrats have called for social media companies to moderate user content more aggressively.
Republicans have responded with cries of censorship and threats to punish internet companies that restrict expression.
Mass reporting tools exist for other social media platforms too. But TikTok's popularity and growth rate — it was the most downloaded app in the world last year — raise the stakes of what happens there for influencers and other power-users.
When The Times spoke this summer with a number of Black TikTokers about their struggles on the app, several expressed suspicion that organized mass reporting campaigns had targeted them for their race and political outspokenness, resulting in posts being taken down which didn't seem to violate any site policies. Other users — from transgender and Jewish TikTokers to gossip blogger Perez Hilton and mega-influencer Bella Poarch — have similarly speculated that they've been restricted from using TikTok, or had their content removed from it, after bad actors co-opted the platform's reporting system.
"TikTok has so much traffic, I just wonder if it gets to a certain threshold of people reporting [a video] that they just take it down," said Jacob Coyne, 29, a TikToker focused on making Christian content who's struggled with video takedowns he thinks stem from mass reporting campaigns.
H4xton posted his mass reporting script on GitHub, a popular website for hosting computer code — but that's not the only place such tools can be found. On YouTube, videos set to up-tempo electronica walk curious viewers through where to find and how to run mass reporting software. Hacking and piracy forums with names such as Leak Zone, ELeaks and RaidForums offer similar access. Under download links for mass reporting scripts, anonymous users leave comments including "I need my girlfriend off of TikTok" and "I really want to see my local classmates banned."
The opacity of most social media content moderation makes it hard to know how big of a problem mass reporting actually is.
Sarah Roberts, an assistant professor of information studies at UCLA, said that social media users experience content moderation as a complicated, dynamic, often opaque web of policies that makes it "difficult to understand or accurately assess" what they did wrong.
"Although users have things like Terms of Service and Community Guidelines, how those actually are implemented in their granularity — in an operational setting by content moderators — is often considered proprietary information," Roberts said. "So when [content moderation] happens, in the absence of a clear explanation, a user might feel that there are circumstances conspiring against them."
"The creepiest part," she said, "is that in some cases that might be true."
Such cases include instances of "brigading," or coordinated campaigns of harassment in the form of hostile replies or downvotes. Forums such as the notoriously toxic 8chan have historically served as home bases for such efforts. Prominent politicians including Donald Trump and Ted Cruz have also, without evidence, accused Twitter of "shadowbanning," or suppressing the reach of certain users' accounts without telling them.
TikTok has downplayed the risk that mass reporting poses to users and says it has systems in place to prevent the tactic from succeeding. A statement the company put out in July said that although certain categories of content are moderated by algorithms, human moderators review reported posts. Last year, the company said it had more than 10,000 employees working on trust and safety efforts.
The company has also said that mass reporting "does not lead to an automatic removal or to a greater likelihood of removal" by platform moderators.
Some of the programmers behind automated mass reporting tools affirm this. H4xton — who spoke with The Times over a mix of online messaging apps — said that his Reportation Bot can only get TikToks taken down that legitimately violate the platform's rules. It can speed up a moderation process that might otherwise take days, he said, but "won't work if there is not anything wrong with the video."
Filza Omran, a 22-year-old Saudi coder who identified himself as the author of another mass reporting script posted on GitHub, said that if his tool was used to mass-report a video that didn't break any of TikTok's rules, the most he thinks would happen would be that the reported account would get briefly blocked from posting new videos. Within minutes, Omran said over the messaging app Telegram, TikTok would confirm that the reported video hadn't broken any rules and restore the user's full access.
But other people involved in this shadow economy make more sweeping claims. One of the scripts circulated on hacker forums comes with the description: "Quick little bot I made. Mass reports an account til it gets banned which takes about an hour."