Antisemitism on TikTok spiked 912% says a new study. But is it accurate?
It’s no secret that antisemitism runs rampant on TikTok, the viral video sharing app. And yet when a recently released study from the University of Haifa showed a 912% increase in antisemitism on the app since 2020, the size of the spike seemed shocking.
I’ve reported on TikTok for months, both on its inventive Jewish creators and the app’s struggles to contain antisemitism — such as when, for Jewish Heritage Month, the platform highlighted Jewish creators on its “Discover” page and exposed them to a barrage of antisemitic comments.
Antisemitism and hate speech is very difficult to regulate on social media, and most platforms struggle with identifying and blocking it. Yet a 912% increase is beyond the pale; what does it mean for antisemitism on the app to rise by such an exponential amount?
A better understanding of that alarming spike requires some context about how TikTok evolved during the pandemic, what antisemitism looks like on the app and how exactly the study is measuring the increase. While the findings sound alarming, they don’t necessarily show that TikTok is more of a magnet for antisemitism than it once was.
Antisemitism on TikTok takes myriad forms. These can include:
- Videos featuring Nazi salutes, Holocaust denial or other antisemitic conspiracy theories or allusions.
- Viral trends that are repeated and remade by thousands of users across the platform – combining songs, dances, gestures or filters in a way that mocks Jews. Previous examples include a trend using a filter to distort the user’s face, exaggerating the nose and smile, set to the song “If I Were a Rich Man” from “Fiddler on the Roof.”
- Antisemitic comments, such as posting “How was the Holocaust?” on Jewish videos; pushing Jewish creators to accept Christianity with remarks such as “Jesus is the Messiah”; saying “free Palestine” on non-political videos from non-Israeli Jews; and posting coded antisemitic references such the numbers 109 or 110, alluding to number of countries from which white supremacists claim Jews have been expelled.
- Antisemitic references in usernames.
- Using TikTok’s “duet” feature to place antisemitic content opposite a Jewish creator’s video, such as posting an oven opposite a video about Passover customs.
Breaking down 912% increase
All of the above could be found on TikTok before the pandemic, but the pandemic helped grow the platform into a bigger stage for all kinds of content — amusing and malicious alike.
Tiktok’s short, addictive videos and intelligent algorithm made for a great amusement during quarantine, and downloads of the app shot up. It went from being an app largely for teens and tweens to being one used by people of all ages around the world.
But the shocking 912% increase in antisemitism that headlines the University of Haifa study is only referring to comments; the increase in antisemitic videos was only 41%. In its sample, antisemitic comments rose from 41 to 415, and videos from 43 to 61. Meanwhile, the study notes antisemitic usernames grew by 1,375%, but the numbers used are quite small, showing a growth from just four to 59; TikTok’s overall users likely number in the billions, given that it has now been downloaded 3 billion times.
However, the study does not identify when in 2020 the initial sample was taken; TikTok was downloaded 315 million times just in the first quarter of 2020 according to Sensor Tower, a third-party site. While TikTok has not released 2021 user numbers, its 2020 Transparency Report says there were over 891 billion videos uploaded in the second half of 2020; in comparison, the first half of the year saw around 104 billion, for an increase of 856%.
So though the study does not give the increase in antisemitism in relationship to the app’s overall growth, from their data it looks as though the growth in instances of antisemitism on the app may be relatively proportional to the app’s overall growth.
It can be difficult to paint an accurate picture of activity on TikTok. The app is known for its proprietary algorithm, which feeds videos to each user based on its assessment of their interests, so no single user’s profile can give a representative cross-section of the app. It’s also difficult to search the app, since it’s not structured around keywords and users often purposefully misspell words to avoid filters. However, there is no mention in the press release provided to journalists of how the study selected videos or accounted for the algorithm to get a global view of the app, and I did not receive a response to my inquiry to the study’s authors.
It also appears that the study did not use a large sample size, though the total number of videos viewed by researchers is not mentioned. However, TikTok has trillions of videos and many have thousands of comments; the 415 antisemitic comments referenced in the study is a very small number of the overall comments on the app, and the same can be said about the other numbers used.
Finally, the study used the IHRA definition of antisemitism, which can be read to define criticism of Israel as antisemitism; the results did not break out forms of antisemitism in their data, so it’s impossible to know whether criticism of Israel represented a large portion of the antisemitism they found.
None of this is to say that antisemitism is not a problem on TikTok — every Jewish creator I’ve spoken to says they struggle to police comments on their videos, and find themselves constantly reporting antisemitic videos and users. Meanwhile, they often find their own videos flagged as hate speech by the app’s algorithm or other users.
Antisemitism is an ongoing issue on TikTok, and it certainly hasn’t gotten better. But it also probably hasn’t gotten nine times worse.