This report from the Online Hate Prevention Institute, in partnership with the Executive Council of Australian Jewry, examines online antisemitism in Australia. It is based on a sample of antisemitic content manually collected by experts from Twitter, Facebook, Telegram, TikTok, Gab, YouTube, Bitchute, Instagram, and Reddit between November 3, 2022 and June 2, 2023. The report provides both quantitative and qualitative analysis of this data including an examination of takedown rates by the platforms after content was reported to them. 

The report also categorises content according a taxonomy based on the International Holocaust Remembrance Alliance (IHRA) Working Definition of Antisemitism and IHRA’s Working Definition of Holocaust Denial and Distortion. The taxonomy includes four major categories of antisemitism and twenty-six subcategories.

Impact of this work

ABC 7.30 broadcast a story discussing the new report at 7.30pm on Monday, August 14th, 2023. You can watch it by clicking on the tweet below, or watch / read about it at the ABC website.

As the federal government pursues a ban on Nazi symbols, new research reveals the disturbing levels of antisemitism facing young people, both on campus and online. @begley_patrick #abc730 pic.twitter.com/x1cXh8cgzG

— ABC News (@abcnews) August 14, 2023

We presented some of the key findings from the report in the European Parliament in June, prior to publication. Our testimony was provided to a hearing of the Interparliamentary Task Force To Combat Online Antisemitism, a group of parliamentarians from around the world.

We’ve presented details from the report privately to some of the major technology platforms prior to publication and will be working with those we have relationships with to help them better understand the kinds of antisemitism are currently falling through the cracks in both AI based moderations systems and the systems that handle reports from users. We welcome outreach from additional technology platforms wishing to learn from this work to improve their moderation efforts, please contact us.

We’re looking forward to sharing this work at the IEEE Computer Society’s Digital Platforms and Societal Harms event in Washington DC in October. We’ll also be presenting it at a meeting of the International Holocaust Remembrance Alliance in Croatia in November.

Findings

An initial sample of 432 items, all visible from Australia and likely to be antisemitic were collected. After review by two experts, we retained 370 items. These items form the basis of this report. Many of the excluded items are still of interest as they promoted extremism, glorified Nazism, distorted facts around WWII (but not specifically related to the Holocaust or Jews), or engaged in general racism and xenophobia (but did not specifically target Jews).

An effort was made to seek out content posted by Australian accounts and 165 (38%) of the items covered by this analysis are believed to have been posted by Australians.

As items have been classified into multiple categories, a total of 947 classifications have been made, 428 classifications into four major categories, and a further 519 classifications into their sub-categories. The most common major category is traditional antisemitism which accounts for 40% of the major classifications. This is closely followed by Holocaust related content which makes up 38% of the major classification. The remaining two major categories, Israel related antisemitism and content inciting violence each account for 11% each of the major classifications.

The relative prevalence of each major category of antisemitism (compared to the prevalence of the other major categories) varies significantly by platform. Holocaust related content was the most prevalent major category on Facebook (58%), YouTube (50%), Reddit (46%), and Twitter (39%). On Instagram Holocaust related content and traditional antisemitism were equally prevalent (44% of each). Traditional antisemitism was the most prevalent category on BitChute (61%), Telegram (61%), Gab (54%), and TikTok (43%). To the extent it was possible, all items were reported to their respective platforms and the removal rates monitored. TikTok has the best removal rate at 15%, while Reddit and Bitchute had the worst with no items removed. Other removal rates were: Gab 3%, YouTube 4%, Instagram 13%, Telegram 12%, Twitter 14%, and Facebook 12%.

The report includes a significant number of examples of the antisemitic content and discussion deconstructing them.

Recommendations

The following recommendations appear in the report:

1. Australia needs to start preparing for potential regulation of online hate.
2. The government should close the gap in eSafety’s remit so it can respond to unlawful online abuse of minority communities, including the Jewish community.
3. Government should provide the necessary resources to facilitate the on-going collection, measurement, and regular reports on online antisemitism impacting Australia.
4. Government should provide the necessary resources to facilitate the on-going collection, measurement, and regular reports focused on other forms of online hate impacting Australia.
5. The IHRA Working Definition of Antisemitism needs to be operationalised as part of government efforts to tackle antisemitism, and the IHRA Working Definition of Holocaust Denial and Distortion needs to be both adopted and operationalised by the government.
6. Twitter / X needs to add proactive, ideally automated, removal of antisemitic content. This should be a basic expectation for any large platform. Government should require this as part of its basic online safety expectations.
7. Platforms should provide specific transparency reports on antisemitism, and reports on other specific forms of hate, rather than generic hate speech reports. Governments may need to regulate to require this to ensure it occurs.
8. Platforms should fund audits, like this report, and use them to improve their responses to online antisemitism.
9. Platforms that have not yet banned Holocaust denial should do so as a matter of urgency.
10. Platforms that have banned Holocaust denial need to do more to remove old content that is in violation, and to enforce this policy on new uploads.
11. The Australian government through relevant agencies, departments, and the parliament should engage more deeply with IHRA, Inter-Parliamentary Task Force to Combat Online Antisemitism, and other international efforts to address antisemitism, particularly online antisemitism.

Download the report

The report has been released under a Creative Commons license, you can be download it here.

Support our work

Your support for our work combating antisemitism and Holocaust denial directly helps us identify and a secure the removal of harmful content. We accept donations from all over the world, and donations from Australian tax payers are tax deductible.