Minnesota GOP divided over Mike Lindell's potential gubernatorial nomination
Read the original article at NPR →
What the model flagged
Analyzed 2026-08-09 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 70%
The phrase 'targeted the state' carries a threatening connotation beyond neutral reporting, implying deliberate hostile action — for example: targeted
“targeted the state for Medicaid cuts”
Repetition 50%
Repeats key words or phrases to hammer ideas into the audience and make them seem more credible through familiarity — for example: "party" ×3 and "samsundar" ×3.
“"party" ×3”
“"samsundar" ×3”
Conspiracy Appeal 82%
Hidden plots, secret coordination, or shadowy actors invoked to explain away evidence.
Transfer 75%
Qualls borrows authority and legitimacy from his military service as an artillery officer to validate his political candidacy — for example: when I was on active duty as an artillery officer
“When I was on active duty as an artillery officer, when the conventional forces couldn't get it done, we brought in the special forces”
Glittering Generalities 70%
The call for 'different leadership' is vague and emotionally positive without specifying any concrete policy or substance — for example: different leadership
“we need different leadership in a state today”
Name Calling 80%
Klobuchar uses the derogatory label 'rubber stamp' to discredit a potential opponent rather than engage with their positions — for example: rubber stamp of Donald Trump
“We do not need a governor who is a rubber stamp of Donald Trump”
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 →