Social Media and the Normalisation of Hate: October 7 Two Years On

Antisemitism rose sharply after the Hamas terrorist attack on October 7, 2023. This report from the Online Hate Prevention Institute (OHPI) examines changes to the prevalence and nature of antisemitism across ten social platforms from the year before October 7 through to the second anniversary of the attack. It reflects a sharp rise in the months after October 7, a decline later in 2024, then a renewed rise in antisemitism during 2025. On four platforms the level of antisemitism is now higher than it was in the months immediately after October 7, and on three of them the current level is at an all time high.

Overview

Since October 7, 2023, the level of antisemitism has failed to return to the pre-October 7 level. In 2025 it has infact been climbing once again. Not all platforms are behaving the same way, though the general trend on most if upwards and on some that had already reach record levels. The data is based on almost 11,000 items of online antisemitic content, with over 750 antisemitic items gathered from each of: Instagram, TikTok, Facebook, YouTube, X (Twitter), Reddit, LinkedIn, Gab, Telegram, and BitChute.

The report documents the relative frequency of antisemitic narratives and discusses the top 10 categories which are:

  1. Promoting traditional antisemitism, such as blood libel and claims that Jews killed Jesus (36.2%)
  2. Promoting the idea of a world Jewish conspiracy (26.5%)
  3. Describing Israel or Israelis using antisemitic words or imagery (e.g., claims of Jews killing Jesus or blood libel) (25.1%)
  4. Promoting the idea of Jews controlling government or other societal institutions (16.5%)
  5. Dehumanising Jews (10.3%)
  6. Denying Jewish people self-determination, e.g., by claiming Israel’s existence is racist (9.1%)
  7. Glorifying the Holocaust or suggesting it did not go far enough (8.1%)
  8. Promoting the idea of Jews controlling the media (7.5%)
  9. Comparisons of Israeli policy to Nazism (6.4%)
  10. Calling for, aiding, or justifying the killing or harming of Jews in the name of a radical ideology or an extremist view of religion (6.3%)

It also examines a number of more specific “emerging narratives” over their period. Many of these cross into real world incidents. They are:

  • October 7 denial
  • Glorification of October 7, Hamas and terrorism
  • Denial of antisemitism: The Melbourne synagogue attacks
  • Racist Anti-Zionism
    • A recent case study of Racist Anti-Zionism at Sydney University
  • Nazi glorification is reenergised
  • The Iranian influence
  • Tucker Carlson, Candace Owens, Nick Fuentes & Marjorie Taylor Greene

The report examines the spread of the more frequent antisemitic narratives within each platform at three points in time, before October 7, in the months after October 7, and at they stand at the end of September 2025.

  • Before October 7 the impact of COVID-19 was dominant with denial and distortion of the Holocaust common across many mainstream platforms, and conspiracy theories of Jewish world control common across all platforms.
  • After October 7 traditional antisemitism dominated. It involves claims of blood libel, that Jews killed Jesus, and antisemitic slurs. It was the most dominant narrative, or just slightly less frequent than the most dominant narrative, on every platform. Conspiracy theories about Jews controlling the world still remained, and a newly dominant trend of using the traditional antisemitism but targeted at Zionists, Israelis, or Israel emerged.
  • In September 2025:
    • Traditional antisemitism remain the most frequent form of antisemitism (or very close to it) across all but three platforms. On some platforms the narratives consolidated and a far higher percent of the antisemitic content used the traditional antisemitic narratives (e.g. on X this has risen from 24.4% in the months after October 7 to 45.5% in September 2025).
    • The three platforms where traditional antisemitism was not the most common form of antisemitism were Facebook, LinkedIn, and Reddit. These platforms still saw traditional antisemitism appear very frequently, but their most common form of antisemitism was content that using traditional antisemitic narratives like blood libel and deicide in reference to Zionists, Israelis, or Israel. This form of antisemitism, while not dominant, was also at significant levels on YouTube, Instagram and TikTok.

The report concludes with a section on addressing antisemitism expressed in terms of Israel, Israelis or Zionists in which it discusses the Wertheim v Haddad court case and reflects back on the Sydney University incident. The section gives some important guidance:

  • Perspective: Antisemitic content targeting Zionist Jews must be assessed from the perspective of a Zionist Jew.
  • Targeting based on identity: Harassing someone after identifying them based on them presenting as Jewish is antisemitic. This remains true even if the harassment is then framed as something else e.g. anti-Zionism.
  • The importance of language: Harassing someone presenting themselves as a Zionist can still be antisemitic when the language of traditional antisemitism is used in the harassment.
  • The line between ideas and people: Stating Zionism is an ideology, so its critique is not antisemitic, is an over simplification. Some critiques of beliefs are really attacks on the adherents of those beliefs.
  • Zionism’s connection to Jewish identity: For most Jews, Zionism is a part of their Jewish identity. Jewish in this context means Jewish people and a people is a group with a distinct language, culture, history, and land. To say harassing someone because they are a Zionist is not connected to their racial or ethnic identity is factually wrong. It is like saying harassing an Australian First Nations person because of Native Title is “political” and not about their identity, so not racist.

Additional Resources

  • The report can be accessed here
  • Statements of support are currently being collected. The document will be updated to include these statements before being finalised and submitted to the National Library of Australia.

Media coverage and blogs