Live Tracker · Updated September 26, 2026 at 15:30 UTC
Documented activity map
29 documented locations in 14 states
Selected, event-specific reports from January 10 through October 8, 2026. Each dot is a reported place, not a confirmed perpetrator or a hate-crime conviction. The map includes documented white nationalist activity and related bias incidents; a bias incident alone does not establish a white nationalist affiliation. The headline feed below includes analysis, follow-ups and stories outside this map.Vandalism 13 · Hate flyers 7 · Threats 3 · Demonstration 2 · Assault / beating 1 · Arson 2 · Violence 1
29 of 29 documented locations shown
Pins mark approximate city centres, not incident addresses. Rocky Hill and Wethersfield are separate town reports from one possibly connected overnight spree. Nearby flyer drops and graffiti reports may be connected, but each town has its own pin. Flyer distribution may be protected speech; a police investigation does not establish a crime. The Jackson arson and Temecula threat cases are related bias incidents; no white nationalist affiliation has been established. Use the linked reports for details.
Sources and event details
| Date | Activity | Location | Report and status |
|---|---|---|---|
| Demonstration | Danbury, CT | White supremacist group demonstrates outside Danbury police headquartersA masked group assembled outside the police station; city officials confirmed and condemned the demonstration. · CT Insider | |
| Hate flyers | Petaluma, CA | Antisemitic pamphlets left at homes in east PetalumaLocal reporters reviewed pro-Nazi pamphlets; police called this a hate-related incident and had not identified a crime. · Petaluma Argus-Courier | |
| Vandalism | San Anselmo, CA | Swastika and antisemitic writing found at Brookside ElementaryPolice opened a hate-crime investigation after the markings were reported Saturday night. · KTVU | |
| Vandalism | Glenside, PA | Swastikas and anti-Black slur spray-painted at Weldon playgroundAbington Township police investigated; residents and staff cleaned the playground. · The Forward | |
| Vandalism | Waxhaw, NC | Swastika painted on school-board candidate's sign in Union CountyDeputies documented the vandalism near Waxhaw and said the investigation was continuing. · WBTV | |
| Hate flyers | Clintonville, Columbus, OH | Ku Klux Klan flyers found near Clintonville schoolsWOSU confirmed flyers near two schools. Similar flyers appeared in Worthington; the distributor has not been identified. · WOSU Public Media | |
| Hate flyers | Edgewood Borough, PA | Racist Klan flyers distributed in Edgewood and nearby streetsPolice investigated flyers claiming a Klan faction. The report does not link their distributors to a separate local homicide. · CBS Pittsburgh | |
| Hate flyers | Worthington, OH | Police confirm Ku Klux Klan flyers left in WorthingtonPolice said the flyers were placed Monday night, including near schools. Similar flyers were found in Clintonville. · WOSU Public Media | |
| Vandalism | Yardley, PA | Swastikas and racial slur carved into Yardley golf coursePolice said the damage occurred overnight on September 25. · FOX 29 | |
| Threats | York, PA | White supremacist flyers threaten Black and Brown York residentsPennsylvania's Human Relations Commission and local police described the threatening flyers. · Pennsylvania Human Relations Commission | |
| Vandalism | Rocky Hill, CT | Racist and antisemitic graffiti found on signs and vehicles in Rocky HillPolice said this may be connected to vandalism in neighboring Wethersfield the same night. · WFSB | |
| Vandalism | Wethersfield, CT | Swastika and antisemitic writing found on vehicle in WethersfieldPolice also documented markings at Mill Woods Park; this may be linked to Rocky Hill's vandalism. · NBC Connecticut | |
| Vandalism | Manchester, NH | Moka Pot cafe windows smashed and white supremacist graffiti paintedPolice later announced an arrest; the September 25 arrest report describes the September 16 event. · New Hampshire Public Radio | |
| Vandalism | Ventura, CA | Swastikas found at Ventura synagogue, church and parkVentura police said three sites were vandalized between September 4 and 6 and are investigating them as connected felony hate crimes; no suspect affiliation is established. · Ventura Police Department | |
| Threats | Temecula, CA | Temecula mosque reports letter threatening mass killingFederal prosecutors charged a man with mailing threatening communications. The charge remains an allegation; no white nationalist organization link is established. · U.S. Department of Justice | |
| Vandalism | Yorktown, NY | Swastika and other hateful markings found on Yorktown bridgeTown parks staff received a report August 24; police documented the markings before town workers removed them. The date of carving is unknown. · Yonkers Times | |
| Hate flyers | Dearborn Heights, MI | Police investigate Klan recruitment flyers in Dearborn HeightsPolice confirmed flyers on several streets and warned against assuming a link to a nearby demonstration without evidence. · CBS Detroit | |
| Hate flyers | Melvindale, MI | Klan recruitment flyers found in Melvindale neighborhoodsThe police chief said residents reported flyers apparently tossed from a vehicle; no person was identified as the distributor. · CBS Detroit | |
| Vandalism | Ormond Beach, FL | Swastikas spray-painted on property signs in Ormond BeachA police incident report cited by the local newspaper says two signs were damaged; the suspect and motive have not been established. · Ormond Beach Observer | |
| Vandalism | Jackson Township, NJ | Racist and antisemitic graffiti found after break-in at Jackson schoolPolice investigated damage at a former school being acquired by a Jewish institution; they identified seven young people on surveillance video. This is separate from vandalism at a local nonprofit. · NBC10 Philadelphia | |
| Hate flyers | Coto de Caza, CA | Sheriff investigates antisemitic flyers near Coto de CazaSwastika-bearing flyers appeared in multiple places over the preceding month; the distribution dates and perpetrators are unknown. · ABC7 Los Angeles | |
| Threats | Fells Point, Baltimore, MD | Police recover threatening Klan-linked note in Fells PointA note referencing a Klan offshoot was found in a weighted plastic bag; Baltimore police collected it as evidence. · The Baltimore Banner | |
| Vandalism | Palo Alto, CA | Police find anti-Black restroom graffiti at seven Palo Alto sitesCity workers reported seven additional locations July 9. Police are investigating; dates of the individual markings and any common perpetrator are unknown. · Palo Alto Police Department | |
| Demonstration | Washington, DC | Patriot Front members march near Union Station and Capitol HillMasked marchers carried group imagery and anti-immigrant messages. D.C. police monitored the demonstration and reported no major incident. · NBC4 Washington | |
| Vandalism | Millburn, NJ | Swastika discovered near Millburn public library entranceThe library reported the symbol to police on June 23; the township says the investigation was ongoing. · Millburn Township | |
| Arson | Chicago, IL | Cross burned in Chicago's Grant ParkChicago police later filed arson and hate-crime charges over the cross burning. The charges are allegations; no group affiliation was established. · CBS Chicago | |
| Assault / beating | Ceres, CA | White supremacist prison gang members beat Black inmate in Ceres jailFour gang members pleaded guilty to assault and gang charges. Reporting says a tablet-queue dispute preceded the attack; a racial motive for this specific beating was not established. · The Modesto Bee | |
| Violence | San Diego, CA | Three killed in attack on Islamic Center of San DiegoThe FBI said the two deceased attackers shared white supremacist views; investigators described the mosque shooting as a hate-crime investigation. · Associated Press | |
| Arson | Jackson, MS | Federal arson indictment follows fire at Beth Israel synagogueFederal prosecutors allege an antisemitic arson. A white nationalist group affiliation has not been established; the charges remain allegations. · U.S. Department of Justice |
A headline is mapped only after a U.S. location, event date and specific reported action can be verified. Multiple reports about one location and event share one pin. These are selected reports, not a nationwide census. Send a correction or missing report.
Latest reported activity
Updated October 14, 2026 at 01:31 UTC. Headlines link to the original reporting and are not rewritten.
Get alerted
New incidents can be pushed to your phone as they are recorded. Install the free
ntfy app, subscribe to this
site’s alert topic, and each run that finds something new sends a notification with the
location and a link to the reporting. Updates also post to
@kemeticMinds on Telegram.
How to read this map
A shaded state with no dot on it means incidents were recorded there but the reporting
did not name a specific town precisely enough to place a point honestly. Dots are only
drawn where a named place resolved to real coordinates — an approximate dot in the middle
of a state would imply a location no source stated.
Methodology: this tracker is built by scripts/hate_incident_tracker.py from public reporting — not a language model. Each entry is an incident a named outlet documented, with a link to that report; nothing here is an allegation this site is making on its own. Categories, locations and group names come from fixed keyword rules over the source’s own text, and coordinates from OpenStreetMap’s geocoder. This tracks incidents, not people: no private individual is named, profiled or catalogued. Counts are of incidents reported in the window shown, not of all incidents that occurred — under-reporting in this subject area is well documented, so treat every number here as a floor.
Language analysis: how the words are being used
Updated October 13, 2026 at 23:37 UTC · 21,631 public texts from news reports, Bluesky, Mastodon, BitChute, 4chan /pol/ and Telegram · every excerpt links to its primary source.
What the numbers say
- On Bluesky, the most frequent kind of language is violence & militancy (4.6 uses per 1,000 words, in 12% of 3,164 texts).
- On Mastodon, the most frequent kind of language is violence & militancy (2.6 uses per 1,000 words, in 9% of 126 texts).
- On BitChute, the most frequent kind of language is dehumanizing language & slurs (2.7 uses per 1,000 words, in 13% of 245 texts).
- On 4chan /pol/, the most frequent kind of language is dehumanizing language & slurs (4.2 uses per 1,000 words, in 10% of 17,419 texts).
- On Telegram, the most frequent kind of language is violence & militancy (0.9 uses per 1,000 words, in 2% of 437 texts).
- Combining replacement, antisemitic, dehumanizing and violent language, 4chan /pol/ scores highest (8.6 per 1,000 words) and Telegram lowest (0.9).
- News reports use these terms mostly to describe events; posts and comments use them in their own voice or in reply to critics. That is why every number below is split by platform.
Where the text comes from
3D language map
Each dot is one text; dots that use similar words sit close together. Colour is the platform, larger dots contain more of the categories below. Drag to rotate, scroll to zoom, hover to read, click a dot to open the source.
How much of each kind of language, by platform
Bar height is uses per 1,000 words. Compare bars of one colour across kinds of language, or one kind of language across colours.
| Kind of language | News reports | Bluesky | Mastodon | BitChute | 4chan /pol/ | Telegram |
|---|---|---|---|---|---|---|
| Replacement & demographic conspiracy Claims that white people are being deliberately replaced or eliminated. (The search phrases themselves, such as “great replacement,” are not counted.) |
0.0 per 1k words
2% of texts |
0.1 per 1k words
0% of texts |
0.2 per 1k words
1% of texts |
0.0 per 1k words
0% of texts |
0.0 per 1k words
0% of texts |
0.0 per 1k words
0% of texts |
| Antisemitic conspiracy Language blaming Jewish people for social or political events, or denying the Holocaust. |
0.1 per 1k words
3% of texts |
0.1 per 1k words
0% of texts |
0.0 per 1k words
0% of texts |
1.5 per 1k words
7% of texts |
1.6 per 1k words
4% of texts |
0.0 per 1k words
0% of texts |
| Dehumanizing language & slurs Words that describe people as less than human, plus racial and ethnic slurs (counted, never quoted here). |
0.0 per 1k words
2% of texts |
0.1 per 1k words
0% of texts |
0.5 per 1k words
2% of texts |
2.7 per 1k words
13% of texts |
4.2 per 1k words
10% of texts |
0.0 per 1k words
0% of texts |
| Violence & militancy Talk of killing, armed struggle or collapse. Counted; quoted only from news reporting. |
1.1 per 1k words
23% of texts |
4.6 per 1k words
12% of texts |
2.6 per 1k words
9% of texts |
2.5 per 1k words
8% of texts |
2.9 per 1k words
6% of texts |
0.9 per 1k words
2% of texts |
| Recruitment & mobilization Language that invites people to join, meet, or take visible action. |
1.3 per 1k words
30% of texts |
0.7 per 1k words
2% of texts |
1.0 per 1k words
3% of texts |
0.1 per 1k words
1% of texts |
0.5 per 1k words
1% of texts |
0.3 per 1k words
1% of texts |
| Grievance & victimhood Framing white people as persecuted, silenced or under attack. |
0.1 per 1k words
4% of texts |
0.0 per 1k words
0% of texts |
0.0 per 1k words
0% of texts |
0.3 per 1k words
1% of texts |
0.1 per 1k words
0% of texts |
0.1 per 1k words
0% of texts |
| Coded symbols & Nazi references Numbers, symbols and phrases that signal white supremacist or neo-Nazi identity. (Search phrases such as “nazi” and “white power” are not counted.) |
1.1 per 1k words
24% of texts |
1.2 per 1k words
3% of texts |
1.5 per 1k words
5% of texts |
2.6 per 1k words
6% of texts |
0.8 per 1k words
2% of texts |
0.0 per 1k words
0% of texts |
| Anti-immigrant & nativist Charged anti-immigrant framing, beyond ordinary policy vocabulary. |
0.1 per 1k words
7% of texts |
0.3 per 1k words
1% of texts |
0.3 per 1k words
1% of texts |
0.3 per 1k words
2% of texts |
0.5 per 1k words
1% of texts |
0.0 per 1k words
0% of texts |
| Anti-LGBTQ & gender panic Language casting LGBTQ people as a threat or as corrupting children. |
0.0 per 1k words
0% of texts |
1.0 per 1k words
3% of texts |
0.0 per 1k words
0% of texts |
0.3 per 1k words
1% of texts |
0.6 per 1k words
2% of texts |
0.0 per 1k words
0% of texts |
Trend over the last 14 days
Each ribbon counts the texts containing that kind of language on each day, across all platforms.
Tone and distinctive words
| Platform | Average tone | Texts with strongly negative tone |
|---|---|---|
| News reports | -0.12 | 53% of 240 |
| Bluesky | -0.02 | 31% of 3,164 |
| Mastodon | -0.20 | 40% of 126 |
| BitChute | -0.10 | 37% of 245 |
| 4chan /pol/ | -0.13 | 36% of 17,419 |
| Telegram | -0.12 | 28% of 437 |
Words that stand out on each platform (compared with the others):
News reportsbbleeadiossoffpretdfaafranciscodanburyrepubliccollinseht
Blueskyconstitution tariffsdestruction americatrumpstein fileswars crashingamerikkka you'reamerica amerikkkadistract trumpsteintrashing constitutiontariffs destructionyou're distract
Mastodonuspolchristofascismcapitalismdigoscdnpoliderpoliticsnazismpartyafd
BitChuteking charlescharles iiitruthslapprophet mohammedreset kingzero targettarget plannet zeroplan killiii net
4chan /pol/kekpres trumpchuganonn*****jeetsliberatedr******n******shill
Telegramrutubejabalrybarsanaenglish versionaden forcesukrainian-russian warday situationtaizlive help
In their own words
Examples chosen by a fixed rule: highest engagement first, at most two per platform per category, one per source page.
Replacement & demographic conspiracy — examples
Antisemitic conspiracy — examples
Dehumanizing language & slurs — examples
Quoted only from news reporting; posts and comments in this category are counted but not reproduced.
Violence & militancy — examples
Quoted only from news reporting; posts and comments in this category are counted but not reproduced.
Recruitment & mobilization — examples
Grievance & victimhood — examples
Coded symbols & Nazi references — examples
Anti-immigrant & nativist — examples
Anti-LGBTQ & gender panic — examples
Reference
Glossary: what the coded terms mean
- Great Replacement: A conspiracy theory that white populations are being deliberately replaced by immigrants and non-white people. It has been cited in several mass-shooting manifestos.
- 14 words: A white supremacist slogan (“We must secure the existence of our people and a future for white children”). ADL lists it among the most common white supremacist slogans.
- 1488: A numeric symbol combining “14” (the 14 words) and “88” (the eighth letter, H, twice: “Heil Hitler”).
- Active Club: Local, gym-style groups that present as fitness or fight clubs; researchers and the SPLC link many to white supremacist networks.
- Accelerationism: The belief that violence should be used to hasten the collapse of society. In extremist spaces it is often paired with talk of a “race war.”
- Remigration: A euphemism for mass removal of non-white or immigrant populations, used to make expulsion sound administrative.
- ZOG: “Zionist Occupied Government”: an antisemitic conspiracy theory that Jewish people secretly control governments.
- Globalist / globohomo: Terms that can be used as coded antisemitism or anti-LGBTQ language, depending on context.
- Dog whistle: A phrase that sounds ordinary to outsiders but carries a specific meaning to an in-group.
Definitions follow the ADL Hate Symbols Database.
How this was measured, and what it cannot tell you
- Sources. Public text only, no login: Google News articles, Bluesky search, Mastodon hashtag timelines, BitChute video search plus each video’s public comment section, 4chan /pol/ threads and replies (public JSON API; poster ids and flags are never read), and Telegram public channel previews (t.me/s/). Telegram covers only channels that have been verified public; researchers do not publish channel lists, so none are assumed.
- Rolling window. 21,631 texts from the last 21 days, refreshed daily. Numbers change as the window moves.
- Counting. A fixed word and phrase list (below) is matched against each text. “Per 1,000 words” divides matches by the platform’s total words; “% of texts” is how many texts had at least one match. Tone uses the VADER sentiment lexicon. Distinctive words use TF-IDF. The 3D language map turns each text into a word-pattern vector and reduces it to three dimensions (SVD): texts that use similar words sit near each other.
- What a match means. That the language appears, not that its author believes it. Critics, journalists and people replying to extremists use the same words. A list cannot read sarcasm, quotation or context.
- Sampling. Search surfaces what platforms return for these queries, not everything posted. Small samples are flagged. Platforms differ in text length and culture, so compare rates cautiously.
- Privacy. No usernames, display names, profile images or account ids are collected or shown. Excerpts are short, slurs are masked with asterisks, and violent or slur-heavy categories are quoted only from news.
- No AI. No language model wrote or scored anything here. The word list and rules are fixed and listed below.
Terms measured (examples):
- Replacement & demographic conspiracy: replac, white extinction, demographic, 14 words, fourteen words, secure the existence of our people, white…
- Antisemitic conspiracy: zog, zionist occupied, jewish, synagogue of satan, goy, holohoax, holocaust, jew world order, globalist…
- Dehumanizing language & slurs: vermin, trann…
- Violence & militancy: race war, boogaloo, accelerat, day of the rope, civil war, shoot, kill, armed, militia…
- Recruitment & mobilization: join, rall, flyer, banner drop, white pride worldwide, proud boys, sign up, get involved…
- Grievance & victimhood: war on whites, white, our people, white guilt, reverse racism…
- Coded symbols & Nazi references: 1488, 14/88, hh, sonnenrad, black sun, wolfsangel, totenkopf, sieg heil, heil hitler…
- Anti-immigrant & nativist: third world, mass migration, go back to, send them back, migrant crime, mass deportation…
- Anti-LGBTQ & gender panic: globohomo, gender ideology, woke agenda…
How this tracker works
- What it tracks: news about reported white nationalist activity in the United States.
- Where headlines come from: Google News searches for “white nationalist” and for named groups (Patriot Front, Proud Boys, neo-Nazi) covering the past three days; stories from outside the U.S. are excluded.
- What it adds and doesn’t: each entry is the outlet’s own headline, linked to the original. Nothing is rewritten or summarised by AI.
- Labels: tags say what kind of source a story comes from (official source, wire service, national, local or international outlet, academic source, or social media post). They are not ratings of accuracy or trust. Opinion pieces are tagged OPINION and kept out of the Latest box.
- Updates: twice a day, at about 8:30 a.m. and 8:30 p.m. Central. The Latest box shows the four newest non-opinion stories, one report per story, with a count of other outlets that covered it.
- Current source mix (last 25 entries): news outlet 18, local outlet 3, national outlet 2, international outlet 1, opinion 1, from 21 different outlets.
- Spot an error? Tell us and we will correct it.
Research behind these features
- Curry, A. L., & Stroud, N. J. (2021). The effects of journalistic transparency on credibility assessments and engagement intentions. Journalism, 22(4). Explaining how and why a story was produced raised readers’ credibility ratings, across political ideologies. This is why each tracker explains its sourcing and filters.
- Otis, A. (2024). The effects of transparency cues on news source credibility online: An investigation of ‘opinion labels’. Journalism. Labelling opinion content on story previews increased the perceived credibility of the source. Opinion pieces here are labelled and kept out of the Latest box.
- Aslett, K., et al. (2022). News credibility labels have limited average effects on news diet quality and fail to reduce misperceptions. Science Advances, 8(18). Trust ratings on sources did not, on average, change what people read or believe. So the tags here describe what kind of source a story comes from, not whether to trust it.
- Wineburg, S., & McGrew, S. (2019). Lateral reading and the nature of expertise. Teachers College Record, 121(11). Professional fact-checkers judge a claim by quickly checking what other sources say. Each latest story links to other coverage of the same headline.
- Pennycook, G., et al. (2021). Shifting attention to accuracy can reduce misinformation online. Nature, 592, 590–595. A brief prompt to think about accuracy improved the quality of news people chose to share. The Latest box ends with that prompt.
