Monthly Report November 2025
Data for November 2025 was based on 11.8M messages in 25 languages across 12 social media platforms such as X, YouTube, TikTok, Threads and Instagram.
Content warning: Presented data may contain disturbing language related to online hate speech.
Average toxicity
Toxicity levels in November remained highly stable, holding steady at 0.20 for most of the month. A brief dip to 0.19 occurred between 13–19 November, after which levels returned to 0.20 and closed slightly higher at 0.21 on the final day. Overall, the month showed minimal fluctuation and no notable spikes in toxicity.
Baseline channel analysis
Across November, toxicity levels remained broadly stable across all baselines, with only small day-to-day variations. Antisemitic content showed the highest averages throughout the month, fluctuating narrowly between 0.29 and 0.31. Anti-LGBTQ+ narratives followed a consistent pattern around 0.26–0.27, with no marked shifts. Anti-Muslim content trended slightly downward from 0.25 to around 0.22 toward mid-month before stabilising again. Anti-refugee discourse stayed steady around 0.21–0.23, while anti-Roma content showed minor fluctuations between 0.15 and 0.18. Sexism remained the lowest and most stable category, holding close to 0.16 throughout the entire period.
Baseline Sexism
Volume: 4.3M of 46M posts (9.4%) in this channel were identified as toxic.
Platforms: Most data was observed on X (52%), and YouTube (34%).
Languages: Posts are mostly written in English (51%), and Arabic (11%).
Toxicity Analysis: The average toxicity score was 0.16. Posts with very high toxicity (≥0.8) totaled 185k, and are most prominent on Chirpwire (13%).
Toxicity Categories:
49% of toxic posts were categorised as Sexism.
35% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: women, woman, الله, stupid, girl, femmes, femme, الصهاينة, homme, charlie
Key Findings: The provided social media posts reveal a range of sentiments and accusations related to gender, power, and societal issues. Several posts express strong opinions on justice systems and immigration policies, particularly in Germany, with one user suggesting the deportation of rapists as a solution to overburdened courts. In Spain, there are heated exchanges, with one post referring to a public figure as "cowardly and disgusting scum" and accusing them of misogyny and femicide. There's also a discussion in Dutch about the word "feeks" (witch) and its appropriateness in referring to a politician. English posts touch upon themes of racism, antisemitism, and political figures like Trump. The overall sentiment across many posts is one of anger, frustration, and the use of derogatory and offensive language towards individuals and groups.
Selected News items
• Italia introduce el delito de feminicidio y lo castiga con cadena perpetua | El Correo
• Un clamor contra el machismo recorrerá con varias marchas los municipios de Bizkaia | El Correo
• Presidente mexicana apresenta reforma para criminalizar assédio após ter sido vítima
Baseline Anti-LGBTQ+
Volume: 914.4k of 6.1M posts (15.0%) in this channel were identified as toxic.
Platforms: Most data was observed on X (72%), and YouTube (17%).
Languages: Posts are mostly written in English (53%), and Spanish (10%).
Toxicity Analysis: The average toxicity score was 0.27. Posts with very high toxicity (≥0.8) totaled 34k, and are most prominent on Gab (3%).
Toxicity Categories:
73% of toxic posts were categorized as Sexism.
17% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: homo, women, schwule, lésbica, lesbiana, lesbica, lesben, lesbian, queer, trans
Key Findings: The provided posts contain a significant amount of hateful and derogatory language directed towards individuals based on their perceived sexual orientation and gender identity. Many posts use slurs and express extreme animosity, often associating these identities with negative traits such as criminality, mental illness, and moral corruption. There are recurring themes of violence, with some posts explicitly wishing harm or death upon individuals. Some posts also conflate different identity groups, such as associating "queer" individuals with pedophilia or linking specific political ideologies with negative stereotypes. The language used is consistently aggressive and dehumanising, reflecting a deep-seated prejudice. The posts originate from various European countries, with some referencing specific political figures or events within those nations. The overall sentiment expressed is one of intense hostility and rejection towards the targeted groups.
Selected News items
• Moment street racer smashes into Florida gay bar, killing four
• El Supremo de EEUU avala el veto de Trump a los pasaportes para las personas transgénero
• Luxemburg/Warschau: Urteil: Polen muss gleichgeschlechtliche Ehe anerkennen
Baseline Anti-Muslim
Volume: 1.9M of 11.5M posts (16.8%) in this channel were identified as toxic.
Platforms: Most data was observed on X (72%), and YouTube (20%).
Languages: Posts are mostly written in English (49%), Arabic (17%), and French (10%).
Toxicity Analysis: The average toxicity score was 0.24. Posts with very high toxicity (≥0.8) totaled 142k, and are most prominent on Chirpwire (12%).
Toxicity Categories:
80% of toxic posts were categorised as Religion.
36% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: muslim, muslims, moros, moro, islam, islamic, الله, الإسلام, المسلمين, quran
Key Findings: The social media posts reveal a strong undercurrent of anti-Muslim sentiment, particularly within Spain and the Netherlands. Several posts from Spain use derogatory terms like "gentuza" and "moros" to describe Muslims, linking them to criminality, illegality, and negative societal issues such as debt and political corruption. There's a recurring theme of associating Muslims with terrorism and murder. The sentiment extends to a call for the government to imprison those perceived as supporting or being associated with Muslims. In the Netherlands, a post directly targets Muslims with hateful language, calling them liars, dogs, and murderers, and advocating for their removal. The posts also touch upon broader societal anxieties, linking immigration and the presence of Muslims to a perceived decline in national identity and an increase in crime. Some posts express a sense of being overwhelmed by the perceived influx of Muslims, with one lamenting, "They hate us and want to rob us of the country."
Selected News items
• Trump grozi wojną kolejnemu krajowi. "Rząd Nigerii powinien działać szybko"
• Trump grozi wojną! "Oni zabijają chrześcijan"
• Progon kršćana u Nigeriji: Što se događa u najmnogoljudnijoj afričkoj državi? - narod.hr
• Trump zagrozil Nigeriji z vojaškim posredovanjem zaradi, kot pravi, nasilja nad kristjani
Baseline Anti-Refugee/Migrants
Volume: 384.4k of 3.3M posts (11.5%) in this channel were identified as toxic.
Platforms: Most data was observed on X (73%), and YouTube (15%).
Languages: Posts are mostly written in English (35%), Spanish (16%), and German (13%).
Toxicity Analysis: The average toxicity score was 0.22. Posts with very high toxicity (≥0.8) totaled 15k, and are most prominent on Gab (1%).
Toxicity Categories:
66% of toxic posts were categorised as Politics.
34% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: refugees, migrant, لاجئ, migrante, asylanten, immigrati, deutschland, vluchtelingen, fachkräfte, réfugiés.
Key Findings: Several posts specifically mention Germany, with criticisms directed at its government's policies regarding refugees, particularly from Ukraine. There's a recurring theme of associating immigrants with crime, such as "knife-wielding," "rapists," and "drug dealers." Some posts express frustration with the perceived burden on social systems and taxpayer money, with phrases like "taxpayer money of citizens" and "social parasites." Political figures and parties are frequently mentioned, including Angela Merkel and the Green party in Germany, often depicted negatively in relation to immigration policies. There's also a sense of division within countries, with one post highlighting perceived differences and animosity between people from eastern and western Germany, suggesting that people from the east feel their status is lower than that of migrants. The language used is often aggressive and derogatory, employing slurs and insults directed at both immigrants and those perceived as supporting them. Some posts also touch upon the idea of national identity and the perceived threat to it from immigration, with calls for deportation and stricter border controls.
Selected News items
• Las redadas 'trumpistas' en Lavapiés que indignan al barrio: "A su padre lo pararon por ser negro"
• La inmigración y los urgentes interrogantes de la integración | El Correo
• Feijóo viaja a Melilla para presentar su alternativa migratoria a Sánchez y Vox
Baseline Antisemitism
Volume: 1.9M of 9.6M posts (20.2%) in this channel were identified as toxic.
Platforms: Most data was observed on X (77%), and YouTube (14%).
Languages: Posts are mostly written in English (58%).
Toxicity Analysis: The average toxicity score was 0.30. Posts with very high toxicity (≥0.8) totaled 119k, and are most prominent on Gab (5%).
Toxicity Categories:
83% of toxic posts were categorised as Racism.
37% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: jews, اليهود, jewish, israel, judíos, juifs, يهودي, israël, żydów, juden
Key Findings: The provided social media posts contain a significant amount of hateful and derogatory language directed towards Jewish people. Several posts explicitly use slurs and express a desire for violence or eradication. For instance, one post states, "JUDÍA DE MIERDA!" and another, "Filthy Jewish bitch." There are also references to historical events, such as the Holocaust, with some posts invoking Nazi rhetoric and imagery. Some posts specifically target individuals or groups, mentioning names like "Lopez," "Manzo," "Netanyahu," and "Tucker Carlson," often in a derogatory context. The posts also touch upon geopolitical issues, with frequent mentions of "Israel" and "Palestine," often framed within a narrative of conflict and blame. The language used is consistently aggressive and dehumanizing, employing a wide range of offensive terms and stereotypes. Some posts also express a desire for a "White America" and advocate for the removal of various minority groups, including Jewish people. The overall sentiment across these posts is overwhelmingly negative and hostile towards Jewish individuals and the Jewish faith.
Selected News items
• Antifašistička liga obilježila Kristalnu noć i upozorila na porast mržnje u društvu - Novi list
• Antifašisti u Zagrebu obilježili Kristalnu noć, čuvala ih policija
• Analyser: Brandväggen mot judehat på väg att rämna
• "Gut erhaltener Judenstern": Deutsches Auktionshaus wollte Holocaust-Zeugnisse versteigern
Baseline Roma
Volume: 252.2k of 1.5M posts (16.6%) in this channel were identified as toxic.
Platforms: Most data was observed on X (52%), and YouTube (37%).
Languages: Posts are mostly written in Russian (31%), English (18%), Portuguese (12%), and Spanish (11%).
Toxicity Analysis: The average toxicity score was 0.20. Posts with very high toxicity (≥0.8) totaled 15k, and are most prominent on Gab (6%).
Toxicity Categories:
86% of toxic posts were categorised as Racism.
14% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: gitanos, gitano, gypsies, gypsy, zigeuner, roms, moros, cygan, roma, jews
Key Findings: The provided social media posts reveal a prevalent and often aggressive discourse targeting specific ethnic and social groups. Several posts contain explicit slurs and derogatory language directed at individuals based on their perceived ethnicity or origin, such as "negro de mierda," "gitano," "sudaca," "amarilla," "rumano," and "bougnoule." These terms are frequently used in conjunction with accusations of criminality, disease, and general societal disruption. For instance, one post from Paris describes the city as being overrun by "sauvages, de negres, de bougnoules, de Roms," associating them with death, disease, and violence. Another post from Valencia details a complaint filed against an individual for using insults like "Negro de mierda! Vete al Bioparc!" and "Gitano!" The discussion also touches upon the controversial use of terms like "Zigeuner" (gypsy), with some arguing it's not inherently racist while others condemn its use due to historical associations with persecution. There are also instances where these groups are invoked in fictional contexts, such as comparing characters in a video game to "roaming thieving gypsies." The posts collectively illustrate a pattern of dehumanising language and the association of negative stereotypes with specific communities.
Selected News items
• Šutarjev zakon: pomisleki iz vrst pravosodja, romske skupnosti in nevladnikov
• Kolumna Marka Crnkoviča: Bežimo, tecimo, Cigani gredo!
• Az egri Momentum szerint cigányozott egy jóízűt a DK helyi jelöltje
• Антисемитизмът е все още жив. Толкова жив, колкото не искаме да признаем
• Europarlamentarul Diana Iovanovici-Șoșoacă solicită desființarea institutului Elie Wiesel
The distribution of toxicity levels across baselines shows marked differences in how hostility manifests online.
Sexism and anti-Roma content show the highest shares of neutral posts (62.0% and 63.3%), indicating that a substantial portion of conversations in these channels remains non-hostile.
Anti-Muslim and anti-refugee narratives also show comparatively high neutral proportions (45.7% and 47.0%), though both still register meaningful amounts of low and medium toxicity.
Anti-LGBTQ+ content stands out for having the largest share of low-toxicity posts (44.2%), suggesting frequent but less severe forms of bias or ridicule.
Antisemitic content presents a more even distribution across neutral, low, and medium levels, reflecting a broad spectrum of hostility.
High-toxicity posts remain low across all baselines - ranging from 0.4% to 1% - but are proportionally highest in the antisemitism and anti-Muslim categories, indicating a slightly greater presence of intensely aggressive or harmful expressions in these narratives.
Hate speech by category
The distribution of categories across the six baselines highlights distinct thematic patterns shaping each form of online hate.
Antisemitic content continues to show the strongest alignment with racism (82.8%), followed by substantial overlaps with political (38.1%) and religious framing (39.9%).
Anti-Muslim narratives are even more heavily driven by religion (80.0%), combined with relevant shares of racism (47.9%) and threat-related language (21%), underscoring a blend of ideological and racialised hostility.
Anti-LGBTQ+ content is most clearly characterised by sexism (62.1%) and obscenity (9.2%), supported by notable levels of ridicule (11.2%) and profanity, reflecting a strong gendered and derisive dimension.
Sexism-related discourse itself shows substantial links to sexism (37.4%), ridicule (9.9%), and threatening language (11.3%), suggesting interactions between contempt, gender norms, and verbal aggression.
Anti-refugee content is dominated by political framing (46.5%) and racism (38.6%), with threats (16.3%) also playing a key role, indicating how exclusionary narratives are shaped by debates around migration, identity, and security.
Anti-Roma hate shows a strong racial basis (27.2%) alongside politics (11.1%) and threats (9.0%), although overall levels across categories remain somewhat lower than in the other baselines.