Monthly Report February 2025

Data for February was based on 2.505 M messages in 11 languages, across 6 social media platforms: Reddit, X, 4chan, Gab, YouTube, and Facebook.

Content warning: Presented data may contain disturbing language related to online hate speech.

Average toxicity

The average toxicity in February 2025 was 0.2, maintaining a steady trend throughout the month.

Are you unsure about the terminology we use in this report? Visit our FAQ for more details.

Baseline channel analysis

The timeline graph illustrates the evolving toxicity levels concerning our baselines. The data reveals that the highest toxicity was related to antisemitic narratives, with an average of 0.3. LGBTQ+ toxicity remained consistent at 0.28, followed by Islam-related toxic messages at 0.27, which slightly decreased at the end of the month. Gender and refugee-related toxicity remained relatively lower, with minor variations.

Baseline Sexism

Volume: 1,468,726 posts were analysed, of which 307,976 (21%) are toxic.

Platforms: Most posts originated from X (75%), with Reddit (10%) and TikTok (9%). Toxicity: Average score was 0.18. Posts with extreme toxicity (0.8 or higher) totaled 17,748, with the highest proportion found on 4chan.

Themes:

• Sexism (29%) represents the highest category of hate content, followed by Politics (16%) and Violence (12%).

• Amongst toxic messages, Ridicule (38%) follows Sexism (46%).

• Common keywords in toxic messages included gendered slurs: bitch, slut, حد, whore, and freak. 

Key Insight: While the average toxicity score is relatively low (0.18), the intersection of sexism, politics and violence is cause for concern.

Baseline Anti-LGBTQ+

Volume
291,939 posts were analysed and 94,108 (32%) were identified as toxic.

Platforms
Most posts originated from X (84%), with Reddit (9%) and TikTok (3%).

Toxicity
Average score was 0.29. Posts with extreme toxicity (0.8 or higher) totaled 4437, with the highest proportion found on 4chan.

Themes

• Sexism (58%) and Politics (18%) represent the highest categories.

• Common keywords in toxic messages: شاذ (abnormal/deviant), slut, bitch, pqp (gendered curse word in Portuguese), and puto.

Key Insight
High concentration of gender-based hate speech and LGBTQ+ targeted content, with a significant toxicity score (0.29).

Baseline Anti-Muslim

Volume: 411,555 posts were analysed and 129,804 (31%) were identified as toxic.

Platforms: Most posts originated from X (89%), with TikTok (4%) and Reddit (3%). 

Toxicity: Average score was 0.27. Posts with extreme toxicity (0.8 or higher) totaled 10,906, with the highest proportion found on 4chan. 

Themes:

• Religion (66%), Racism (40%) and Politics (28%) represent the most prominent categories.

• When considering toxic messages only, Religion (76%), Racism (52%) and Violence (38%) are most prevalent.

• Common keywords in toxic messages included Spanish/Arabic terms: moros, moro, المشركين (polytheists/disbelievers), الصهاينة (Zionists), and قتل (kill).

Key Insight: With a high average toxicity score (0.27) and the prevalence of racist and violent terms, this channel shows concerning levels of harmful content.

Baseline Anti-Refugee/Migrants

Volume: 191,294 posts were analysed and 129,804 (23%) were identified as toxic. 

Platforms: Most posts originated from X (88%), with Reddit (4%) and Facebook (4%). 

Toxicity: Average score was 0.22. Posts with extreme toxicity (0.8 or higher) totaled 1,497, with the highest proportion found on 4chan. 

Themes:

• Politics (48%), Racism (37%) and Violence (16%) are the most prominent hate categories. This trend holds true when observing toxic messages only.

• Common keywords in toxic messages included multilingual terms: fachkräfte (skilled workers, used ironically), "migrants", immigrati clandestini (illegal immigrants), negro, and vergewaltiger (rapist). 

Key Insight: While this channel has fewer posts than previous datasets, the consistent combination of political content, racism, and violent messaging indicates targeted toxic discourse around immigration.

Baseline Antisemitism

Volume: 388,040 posts were analysed and 156,079 (40%) were identified as toxic.

Platforms: Most posts originated from X (89%), with Reddit (3%) and 4chan (3%). 

Toxicity: Average score was 0.34. Posts with extreme toxicity (0.8 or higher) totaled 10,043, with the highest proportion found on 4chan. 

Themes:

• Racism (75%), Religion (48%) and Politics (41%) are featured at extreme levels in discussions around Jews, with the following category being Violence (21%).

• Common keywords in toxic messages included multilingual extremist terms: الصهاينة (Zionists), kill jews, nazista (Nazi) and racial slurs.

 Key Insight: This channel shows the highest average toxicity score (0.34),with very high levels of racism, religion and politics, suggesting concentrated extremist discourse with antisemitic and racist elements.

Baseline Roma

Volume: 45,712 posts were analysed and 16,913 (37%) were identified as toxic.

Platforms: Most posts originated from X (83%), with Reddit (6%) and YouTube (4%). 

Toxicity: Average score was 0.27. Posts with extreme toxicity (0.8 or higher) totaled 864, with the highest proportion found on Gab. 

Themes:

• 49% of messages feature Racism, followed by Politics (12%).

• In toxic messages, Ridicule (17%) also plays a significant role.

• Common keywords in toxic messages included anti-Roma/anti-Moorish terms: gitanos, gitano (Roma people in Spanish), zigeuner (Roma people in German), moros (Moors/Muslims in Spanish), and ciganos (Roma people in Portuguese). 

Key Insight: While smaller in volume than other datasets, this channel's focus on racism (nearly half of messages) targeting specific ethnic groups (particularly Roma people) across multiple languages suggests coordinated cross-border ethnic targeting. Toxic discussions suggest Roma people are often dehumanized.

This graph illustrates the distribution of toxicity levels—Neutral, Low, Medium, and High—across our monitored baselines. Sexism recorded the highest proportion of neutral messages (59.9%), while antisemitic content had the largest share of medium toxicity messages (37.6%), closely followed by anti-Roma messages (35.1%).

Highly toxic content, characterised by extreme hate speech, was most prevalent in messages targeting Muslim (2.6%), Jewish (2.6%), and Roma (1.9%) communities.

Hate speech by category

This table outlines key themes in hate speech across antisemitism, anti-Muslim sentiment, anti-LGBTQ+, sexism, anti-refugee sentiment, and anti-Roma narratives. Hate speech targeting refugees is the most politicised (48.5%), followed by antisemitism (41.4%). Racism appears most frequently in antisemitic narratives (75.4%) and Roma communities (49.1%), underscoring the deep-rooted racial biases against these groups.

Religious hostility is particularly pronounced in anti-Muslim (66.2%) and antisemitic (48%) discourse, indicating the significant role of religion in targeted hate speech. The LGBTQ+ community faces the highest levels of sexism (58%), followed by sexist hate speech more broadly (28.8%).