Monthly Report October 2025 (Copy)
Data for October 2025 was based on 10.8M messages in 12 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
Throughout October 2025, overall toxicity levels remained stable, fluctuating only slightly between 0.19 and 0.21. The month began with steady values around 0.20-0.21 during the first week, followed by a mild decline to 0.19 in the middle of the month. This lower range persisted until the final days, when levels returned to 0.20. Overall, the data indicate consistent moderation in toxicity, with no pronounced increases or sharp fluctuations observed across the period.
Baseline channel analysis
Antisemitic content maintained the highest average levels overall, ranging from 0.29 to 0.31, indicating consistent activity with limited variation throughout the month. Anti-LGBTQ+ narratives followed, showing moderate toxicity values, fluctuating between 0.26 and 0.28, with a slight upward trend toward the end of the month. Anti-Muslim content displayed averages between 0.24 and 0.26, showing a slightly narrower and more stable distribution. Sexism-related content remained the lowest across categories, averaging consistently around 0.16-0.17, with minimal deviation over time. Anti-refugee discourse recorded values between 0.20 and 0.23, remaining relatively stable but with small increases around mid- and late October. Anti-Roma content presented slightly lower averages, between 0.16 and 0.19, with gradual increases during the latter half of the month.
Baseline Sexism
Volume: 3.6M of 38.2M posts (9.4%) in this channel were identified as hate speech.
Platforms: Most data was observed on X (51%), and YouTube (34%).
Languages: Posts are mostly written in English (51%).
Toxicity Analysis: The average toxicity score was 0.16. Posts with very high toxicity (≥0.8) totaled 154k, and are most prominent on Chirpwire (12%).
Toxicity Categories:
49% of toxic posts were categorized as Sexism.
34% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: women, woman, الله, stupid, girl, femmes, femme, charlie, quran, الصهاينة
Key Findings: The provided social media posts reveal a strong undercurrent of anger and frustration, often expressed through highly aggressive and offensive language. Several posts target specific individuals or groups, using derogatory terms and accusations. For instance, one post in German uses extremely vulgar language directed at men, including threats of violence. Another post in Spanish contains a string of insults aimed at a political figure, referencing "niños en maralago" and employing deeply offensive epithets. Arabic posts also feature intense vitriol, with one referencing "Daeash" and "Nasrallah," and another discussing the "Arab Spring" and criticizing "Saudis." German posts continue to express strong negative sentiments, with one calling someone an "Arschloch" and another criticizing "Nazis." The posts also touch upon themes of political figures and their alleged wrongdoings, with mentions of "pederasta," "terrorista," and "genocida." Some posts express a desire for societal change or critique current political situations. The language used is consistently harsh, employing insults, threats, and dehumanizing terms across multiple languages, indicating a volatile and confrontational online environment.
Selected News items
• Stupro di gruppo:giallo su audio vittima, difesa verso revisione - Notizie - Ansa.it
• El fallo en el cribado del cáncer de mama en Andalucía obliga a avisar a 2.000 mujeres
• Prozess in Österreich: Jugendliche missbrauchen Lehrerin - Haftstrafen in Wien
• Stupro di gruppo, confermata condanna per uno degli imputati - Notizie - Ansa.it
Baseline Anti-LGBTQ+
Volume: 785.7k of 5.2M posts (15.2%) in this channel were identified as toxic.
Platforms: Most data was observed on X (71%), and YouTube (16%).
Languages: Posts are mostly written in English (52%), and Spanish (11%).
Toxicity Analysis: The average toxicity score was 0.27. Posts with very high toxicity (≥0.8) totaled 30k, 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, lesbica, lesbiana, lesben, lesbian, charlie, trans
Key Findings: The provided social media posts contain a significant amount of hateful and derogatory language directed towards LGBTQ+ individuals, particularly transgender people. The posts frequently use slurs and associate LGBTQ+ identities with criminality, immorality, and mental illness. Several posts specifically target individuals with derogatory terms like "travesti" (transvestite), "homosexual," and "gay," often linking these identities to accusations of being terrorists, drug traffickers, corrupt, or pedophiles. There are also instances of threats of violence, such as calls for castration or imprisonment. The language used is consistently aggressive and dehumanizing, employing a wide range of offensive terms across multiple languages, including Spanish, German, French, Russian, and Arabic. Some posts express a desire for the eradication or elimination of LGBTQ+ individuals, framing them as a threat to society or humanity. The content suggests a strong undercurrent of animosity and prejudice, with individuals being attacked based on their perceived sexual orientation or gender identity.
Selected News items
• "Don´t Say Gay" - Wie ein Gesetz Queers in Florida bedroht
• Schwul, liberal, gegen Migration: Rob Jetten könnte nächster Premier der Niederlande werden
• Noorse kerk biedt excuses aan voor decennialange discriminatie van LGBTQ+ gemeenschap
Baseline Anti-Muslim
Volume: 1.6M of 9.3M posts (17.2%) 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 (47%), Arabic (17%), and French (11%).
Toxicity Analysis: The average toxicity score was 0.25. Posts with very high toxicity (≥0.8) totaled 123k, and are most prominent on Chirpwire (11%).
Toxicity Categories:
80% of toxic posts were categorized 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 provided social media posts contain a significant amount of anti-Muslim rhetoric, often expressed in highly offensive and derogatory language. Several posts target "moros" (Moors), a term frequently used in Spain to refer to Muslims, associating them with criminality, illegality, and negative societal impacts. For instance, one post states, "Spain is full of shit people, thieves, criminals, Moors and North Africans and scum living on aid useless parasites who for a few coins give their vote and elect the trash of the moment... Spain needs to purge so much Muslim shit..." Another post directly calls for "mass deportations" of "Muslims and blacks" who are described as "parasites" and "useless people." Some posts also link these sentiments to political ideologies, with terms like "fachas" (fascists) appearing in conjunction with anti-Muslim statements. The posts originate from various European countries, with a notable concentration from Spain, and are written in multiple languages including Spanish, Arabic, German, and Dutch.
Selected News items
• União Europeia renova sanções a terroristas islâmicos até outubro de 2026
• El Estado Islámico publica fotografías de cristianos "masacres" en el Congo
• Hamas, abandonată și de țările arabe: "Nu are niciun rol" în viitorul Gazei - HotNews.ro
• Egiptul spune că Hamasul nu are "niciun rol" de jucat în viitorul Gazei
Baseline Anti-Refugee/Migrants
Volume: 335k of 2.9M posts (11.5%) in this channel were identified as toxic.
Platforms: Most data was observed on X (74%), and YouTube (15%).
Languages: Posts are mostly written in English (34%), Spanish (16%), and German (13%).
Toxicity Analysis: The average toxicity score was 0.22. Posts with very high toxicity (≥0.8) totaled 13k, and are most prominent on Gab (1%).
Toxicity Categories:
67% of toxic posts were categorized as Politics.
34% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: migrant, refugees, لاجئ, migrante, asylanten, immigrati, deutschland, vluchtelingen, israel, fachkräfte
Key Findings: The social media posts reveal a strong undercurrent of anti-immigrant and anti-refugee sentiment across various European countries. Several posts express frustration with perceived economic burdens, with one user in Germany stating, "Raus mit dem Pack und 15% mehr Rente. Die sollen nicht flüchten sondern kämpfen in Ihrem eigenem Land aber Deutschland in Ruhe lassen." This sentiment is echoed by a Spanish user who laments that "Dinero hay.. pero solamente para chiringuitos, putas, farlopa, sindicalistas, furcias mediáticas, funcionarios inútiles, mantener a inmigrantes ilegales..." The issue of crime is frequently raised, with recurring accusations of migrants being "Mörder & Vergewaltiger" (murderers and rapists), as seen in a German post: "Bis heute heißen wir nur asoziale Analphabeten und Pädophile, künftige Vergewaltiger und Mörder willkommen!" A Dutch user in Sweden states, "Het zijn dobbernegers!! Het zijn geen vluchtelingen. Het is crimineel uitschot!! Vuile jihadisten, rovers, moordenaars en verkrachters zijn het." Some posts also target specific nationalities or groups, such as a German user complaining about "ein paar Rumänen" and "Kopftücher und Kanacken" in their neighborhood. There's also a sentiment that native populations are being disadvantaged, with a Slovenian user stating, "Ilegalni migranti povečujejo kriminal in socialne stroške." The posts also reflect a distrust of political establishments, with accusations of "Linksgrünen FASCHOs" and "Altparteien" pampering migrants while neglecting their own citizens. For instance, a German post criticizes politicians for allegedly "pampert Migranten und Ukrainer, ihr deckt Mörder und Vergewaltiger." The language used is often aggressive and dehumanizing, employing terms like "Gesindel," "Pack," "Abschaum," and "Invasoren."
Selected News items
• Detenido un inmigrante ilegal que había logrado plaza de Policía en Chicago
• Kommentar: "Geht nach Hause zu euren Familien" - Trump verhöhnt die Ukraine erneut
• 62% dintre germani vor ca refugiații ucraineni să fie trimiși înapoi în țara lor
Baseline Antisemitism
Volume: 1.7M of 8.2M posts (20.3%) 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 (57%).
Toxicity Analysis: The average toxicity score was 0.30. Posts with very high toxicity (≥0.8) totaled 105k, and are most prominent on Gab (5%).
Toxicity Categories:
83% of toxic posts were categorized 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 social media posts reveal a disturbing trend of hateful rhetoric and conspiracy theories. Several posts directly target Jewish people with derogatory language and accusations, often linking them to negative stereotypes such as greed and manipulation. For instance, one post claims "Jewish shekels changed your nose..." and another refers to "filthy sub human Jew lover." There's a recurring theme of accusing Jewish people of controlling global systems, with phrases like "ZOG" (Zionist Occupied Government) and references to "Jewish masters." Some posts also express a desire for violence, with explicit calls to "kill all Jews" and comparisons to Hitler and the Holocaust. The posts also show a tendency to conflate Jewish people with the actions of the Israeli government, with statements like "Israel is today Nazi Germany, and the Palestinian is what the Jew was." Additionally, there are instances of Holocaust denial, with one post stating "The holocaust is fake and ghey. No where did 6 million Jews die." The language used is consistently aggressive and dehumanizing, employing slurs and violent imagery. Some posts also attempt to justify these sentiments by referencing religious texts or historical events, often in a distorted manner.
Selected News items
• Nemecký kancelár Merz varoval pred novou vlnou antisemitizmu
• Antisemitismus: Wir Juden sind nirgendwo sicher
• Paris : un tag antisémite découvert devant l'Hypercacher de la porte de Vincennes
• Innen- und Rechtsausschuss: Antisemitischer Aushang - Ermittlungen kurz vor Abschluss
• Ubojica Židova u Manchesteru se usred napada zakleo se na vjernost ISIL-u
Baseline Roma
Volume: 214.7k of 1.2M posts (17.9%) in this channel were identified as toxic.
Platforms: Most data was observed on X (55%), and YouTube (33%).
Languages: Posts are mostly written in Russian (26%), English (19%), Portuguese (13%), and Spanish (12%).
Toxicity Analysis: The average toxicity score was 0.22. Posts with very high toxicity (≥0.8) totaled 13k, and are most prominent on Gab (6%).
Toxicity Categories:
87% of toxic posts were categorized as Racism.
13% of those toxic posts contained violent rhetoric.
Commonly used words in those toxic posts included: gitanos, gitano, gypsies, zigeuner, gypsy, roms, moros, cygan, jews, roma
Key Findings: The social media posts reveal a deeply concerning trend of anti-Roma sentiment expressed through hateful and derogatory language. Several posts explicitly use slurs and stereotypes, associating Roma people with criminality, dishonesty, and undesirable social behaviors. For instance, one post states, "Ich rate mal ins Blaue....: Entweder das "Volk", dass schon immer für Müll und Dreck, sowie Diebstahl, Scams usw bekannt ist...Zigeuner!" (I'll guess blindly... Either the "people" who have always been known for trash and dirt, as well as theft, scams, etc... Gypsies!). Another post from Portugal states, "Os ciganos são uns cobardes porcos,só são valentes em matilha 10 contra 1 Esta merda de raça,só vai aprender quando lhes fizermos o mesmo" (Gypsies are dirty cowards, they are only brave in a pack 10 against 1 This shit of a race will only learn when we do the same to them). There are also calls for expulsion and violence, such as "Zigeuner raus! Dieses verkommene Gesindel gezielt hereingeholt durch antideutschen Abschaum." (Gypsies out! This degenerate rabble deliberately brought in by anti-German scum.) and "Gypsies, these scum should be sent to the gas chamber." The posts also show a pattern of generalizing negative behaviors to the entire Roma population, ignoring individual differences and perpetuating harmful stereotypes. This rhetoric contributes to a hostile environment and normalizes prejudice against Roma communities.
Selected News items
• La Fundación Secretariado Gitano estrena sede en el centro de Pontevedra
• rizospastis.gr - Οι (Π)Αλαντίν και το τζίνι του φασισμού
• (KOMENTAR) Jezik skrajnežev postaja del besednjaka uradne politike
• Kocsis Máté felfedte, kivel alkothatta meg "Zsolti bácsi" karakterét Juhász Péter
Sexism
Low toxicity: 20.1%, indicating visible but relatively contained expressions of everyday hostility or bias.
Medium toxicity: 17.8%, showing a moderate presence of antagonistic or dismissive language.
Anti-Roma
Low toxicity: 17.7%, reflecting mild yet consistent stereotyping and prejudice.
Medium toxicity: 20.1%, indicating a recurring use of moderately demeaning or racialised language.
High toxicity: 0.6%, showing a small share of explicitly hostile or derogatory content.
Anti-Muslim
Low toxicity: 29.2%, suggesting moderately frequent expressions of bias or hostility.
Medium toxicity: 25.5%, marking a smaller but notable share of explicit or targeted hate.
High toxicity: 1.1%, representing the highest share of threatening posts, together with the antisemitism baseline.
Anti-LGBTQ+
Low toxicity: 44.3%, the largest proportion across baselines, reflecting frequent but contained derogatory language or bias.
Medium toxicity: 24.6%, pointing to sustained levels of moderate hostility or ridicule.
High toxicity: 0.5%, indicating rare instances of extreme aggression or dehumanising rhetoric.
Antisemitism
Low toxicity: 34.3%, suggesting a consistent presence of prejudiced discourse framed in milder terms.
Medium toxicity: 33.1%, highlighting a significant share of moderately antagonistic or conspiratorial language.
High toxicity: 1.1%, reflecting limited but intense expressions of overt hate.
Anti-Refugee
Low toxicity: 31.4%, denoting a substantial share of content with subtle prejudice or exclusionary framing.
Medium toxicity: 21.0%, showing a stable level of moderate hostility or xenophobic tone.
Hate speech by category
This data highlights clear thematic distinctions across the six monitored baselines, reflecting how different hate narratives are shaped by intersecting forms of prejudice and discourse.
Antisemitic content shows the strongest association with racism (82.1%) and religion (42.0%), making it one of the most ideologically driven forms of hate speech. It also aligns closely with political discourse (39.5%) and threatening language (21.2%), underlining its persistent link to hostility and extremist framing.
Anti-Muslim narratives similarly combine religious framing (77.7%) and racism (49.5%), reinforcing their dual ideological and racial dimensions. High overlaps with political content (28.6%) and threats (20.9%) further indicate that these discussions are often confrontational and polarising.
Anti-LGBTQ+ hate is primarily structured around sexism (61.5%) and ridicule (11.7%), emphasising its gendered and mocking character. Elevated levels of obscenity (8.8%) and threats (9.5%) reveal a tendency toward both verbal degradation and aggression.
Sexism-related discourse is dominated by sexism itself (35.9%), while intersecting with ridicule (10.0%), political themes (14.8%), and threats (12.1%). This suggests that gendered hostility often appears alongside moralised or ideological arguments, frequently expressed through contempt or verbal abuse.
Anti-refugee content remains highly political (44.8%) and racialised (37.6%), reflecting exclusionary narratives around migration and identity. The relatively strong share of threat-related content (16.3%) highlights the persistent framing of refugees as a social or security threat.
Anti-Roma discourse displays a dominant racial component (29.8%), combined with political framing (11.1%) and a smaller but visible share of threatening (8.2%) and religious (4.8%) elements. This reflects enduring patterns of stigmatisation and marginalisation.