Fetterman, already at odds with his party, makes surprise video for GOP convention

NPR · left 6 techniques found 24.1% of sentences flagged

Read the original article at NPR →

What the model flagged

Analyzed 2026-09-10 04:00 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.

Flag-Waving 93%

Patriotic, religious, or in-group identity invoked to validate a position.

Glittering Generalities 75%

The phrase 'steel way of life' is a vague, emotionally resonant expression invoking identity and heritage without specifying any concrete policy or action — for example: 'defend the steel way of life'

“fight and defend the steel way of life”

Name Calling 85%

The term 'socialism' is used as a derogatory label to discredit a political position rather than engage with it substantively — for example: 'socialism'

“socialism”

Loaded Language 85%

The phrase 'anti-American way of life' uses emotionally charged language designed to trigger a negative reaction and associate opponents with being un-American — for example: 'anti-American way of life'

“anti-American way of life”

Black-And-White Fallacy 70%

The framing implies one must either reject 'extremes and socialism' or embrace an 'anti-American way of life', presenting a binary with no middle ground — for example: 'reject the extremes and socialism and that anti-American way of life'

“reject the extremes and socialism and that anti-American way of life”

Repetition 80%

The word 'smears' is used here and again in the next sentence, functioning as a repeated loaded label to discredit the reporting — for example: 'smears'

“smears”

Analyzed automatically with Semblen'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 →

Get notified as this improves — new detection models, the browser extension, and outlet scorecards. Occasional email, no spam.