The worst primary in America just ended
Read the original article at Vox →
What the model flagged
Analyzed 2026-08-19 04:01 UTC. Articles are sometimes updated after publication — if a quote below isn't in the current version, the piece has changed since.
Findings may include language the outlet is quoting rather than asserting. We flag manipulation techniques wherever they appear, including inside quotations — so check each quote against the original before drawing a conclusion about the outlet.
Loaded Language 75%
The superlative 'worst' is an emotionally charged judgment designed to trigger a negative reaction rather than convey neutral factual information about the primary — for example: 'The worst primary in America'
“The worst primary in America just ended”
Exaggeration/hyperbole 75%
Declaring this the absolute worst primary in the entire country is an extreme superlative claim that goes beyond what evidence could reasonably support — for example: 'The worst primary in America'
“The worst primary in America just ended”
Name Calling 85%
Both subjects are labeled with derogatory identity-based terms rather than being described by their actions or policy positions — for example: 'Islamophobic Congress member' and 'antisemitic influencer'
“Islamophobic Congress member”
“antisemitic influencer”
Appeal To Fear 70%
The phrase implies a threatening future trajectory for the Republican Party, framing the outcome as an ominous harbinger of things to come — for example: 'a sign of the GOP to come'
“It may be a sign of the GOP to come”
Appeal To Prejudice 90%
The quote promotes hostility toward Muslims as a group, exploiting in-group/out-group identity dynamics — for example: 'We need more Islamophobia, not less'
“We need more Islamophobia, not less”
Scapegoating 90%
The quote collectively assigns the label of 'evil' to all Palestinians as a group, blaming an entire people rather than specific actors — for example: 'Palestinian is a synonym for evil'
“Palestinian is a synonym for evil”
Black-And-White Fallacy 80%
The absolute equation of Palestinian identity with evil leaves no room for nuance or distinction, presenting a binary of good vs. evil — for example: 'Palestinian is a synonym for evil'
“Palestinian is a synonym for evil”
False Equivalence 70%
The phrase 'As on the Democratic side' draws a parallel between Democratic and Republican divides over Gaza, implying moral equivalence between the two parties' internal conflicts — for example: both sides equally divided
“As on the Democratic side, divides over Gaza and October 7 have created a schism inside the GOP”
Transfer 80%
Superimposing dogs on the Gadsden flag borrows the patriotic symbolism of the 'don't tread on me' flag to lend legitimacy to anti-Muslim sentiment — for example: 'images of dogs superimposed on a Gadsden "don't tread on me" flag'.
“images of dogs superimposed on a Gadsden "don't tread on me" flag”
Card Stacking 75%
The sentence lists inflammatory actions from both sides (Bilzerian's pro-Holocaust video, a swastika on Fine's sign, Fine's FBI complaint) without context or verification, juxtaposing unverified or unqualified claims alongside each other to paint a maximally damaging picture of the campaign — for example: 'Bilzerian posting a pro-Holocaust video, a swastika spray-painted on a Fine campaign sign'
“Bilzerian posting a pro-Holocaust video, a swastika spray-painted on a Fine campaign sign, and Fine attempting to get Bilzerian investigated by the FBI”
Conspiracy Appeal 88%
The 'Trojan horse' framing implies a hidden, coordinated infiltration of American government by Muslims — for example: suggesting Muslim officeholders are secretly working to undermine the country from within
“every single Muslim holding public office in America is a Trojan horse”
Repetition 90%
The anti-Muslim sentiment is repeated across multiple politicians' statements and Roy posted on the topic 244 times, constituting deliberate rhetorical repetition to normalize the message — for example: posting on the topic 244 times in three-and-a-half months
“posted on the topic 244 times in the first three-and-a-half months of the year on the topic”
Analyzed automatically with Biasly's fine-tuned model. These are manipulation techniques, not political tilt — and finding one is not a claim that the article is false. The model has known false positives on strong-but-legitimate language, so treat each finding as a prompt to read closely, not a verdict. How articles are chosen →