2026 midterms · 12 days to go

What are the political ads where you live really saying?

6,331 Senate and governor ads, every one read for rhetorical techniques, with who paid for it and the claims checked.

Maine784 ads Iowa415 Alaska372 Michigan332 Ohio252 Texas244 Governor races36 states

This weekMichigan and Kansas Senate debates, every answer analyzed.

Know what you’re reading.

Reading something right now? Paste an article for the TL;DR, then see how it’s written, who gets to speak and whether the claims hold up. Videos and posts work too.

Check an article →

Or add Semblen to Chrome: it checks the page you're on, in a side panel.

New: government statements, and the money behind the 2026 ads. The White House, both parties' leaders in Congress and federal departments, analyzed nightly — and every Senate race now shows who is paying for the ads. See what's new →
Why this exists

Balance stopped being a rule in 1987. Now fewer companies own the news.

On August 4, 1987, the FCC voted 4–0 to abolish the Fairness Doctrine — the rule that since 1949 had required radio and television broadcasters to cover controversial issues and to air opposing views. Congress passed a bill to keep it; President Reagan vetoed it. Nothing replaced it, and cable news and the internet were never covered at all.

72%

1976

of Americans trusted the mass media to report the news fully, accurately and fairly.

28%

2025

say the same today — the lowest figure Gallup has recorded in fifty years of asking.

~40%

of local newspapers gone since 2005

Close to 3,500 papers, and 212 counties now have no local news source at all (Medill, 2025).

This year, ownership has been concentrating fast:

Aug 2026

No national cap on TV stations

On August 6, the FCC repealed the rule that kept one owner's stations from reaching more than 39% of TV households; deals are now judged case by case. In March it had waived the cap to approve Nexstar's $6.2 billion purchase of Tegna: 265 stations in 44 states.

Oct 2026

CBS News and CNN, one owner

On October 6, Paramount Skydance completed its $111 billion purchase of Warner Bros. Discovery.

→

So a second read is yours

News is a business that sells attention, and fewer companies own more of it. Semblen gives you a second read on any article: the TL;DR, how it’s written, who gets to speak, whether the claims hold up and what’s missing — and leaves the conclusion to you. Check an article →

Each of these is one cause among several — the 24-hour cable cycle and the collapse of advertising-funded local news did their part too. None of it is a verdict on any company or deal; it is why a second read is worth having.

What's new

Government statements, campaign money, and claim checks

★

Government, analyzed nightly

What the White House, both parties' leaders in Congress and federal departments said — releases, statements, and the leaders' floor speeches from the Congressional Record — run through the same analysis every night. Every item is listed, flagged or not, and both parties sit side by side under one rule. See what they said →

$

2026 ads — and the money behind them

The Facebook, Instagram and YouTube ads in this year's Senate races, analyzed line by line, with FEC filings for every candidate: outside spending for and against them, linked to the spender's own ads; money bundled through PACs, with every PAC that forwarded $100,000 or more called out, whatever its cause; and each super PAC's largest donors. Browse by race or look up your ZIP →

✓

Claim checks with cited sources

Semblen pulls out the checkable claims an article makes, and up to 5 it reports others making, and looks for primary sources, fact-checkers and wire reporting. Each verdict links to the passage it rests on — and when the evidence is thin it says “No authoritative evidence found” rather than guess. Loaded language inside a quotation is marked as the speaker's, not counted against the outlet.

How it works

How Semblen works

Semblen reads an article alongside you: the TL;DR, the rhetorical techniques it uses — loaded language, name calling, false equivalence, scapegoating and more — who it quotes, and whether its own factual claims hold up against cited sources. Pasted text gets the same, and videos get a TL;DR as well.

1

Article, text, or video

Paste a URL, drop raw text, or paste a video link — YouTube, TikTok, Instagram, Facebook, or X. For articles we follow pagination automatically and analyze up to 500 paragraphs. For videos we analyze the transcript in 1-minute segments.

2

A trained model, then Claude

A model we trained on hand-reviewed news text reads each paragraph and finds the words that use a rhetorical technique. Claude then names the technique, explains in a sentence how it works on a reader, and says whether the words are the outlet’s own or someone it quotes. Separately, it writes the TL;DR and checks the article’s factual claims against cited sources.

3

Read it with the notes

The TL;DR comes first, then a card for each question a careful reader asks: how it’s written, who gets to speak, whether the claims hold up, what’s missing and how other outlets covered it. In the article itself, the words that use a technique are underlined in amber with a note on why; language inside a quotation is marked as the speaker’s and kept out of the count.

Historical context

See the history this article leaves out

When an article has a real backstory — a geopolitical conflict, a financial crisis, a decades-long policy fight, a public-health debate, a civil-rights movement — Semblen’s “What’s missing?” card surfaces the historical background most coverage skips: what this article leaves out, measured against what it addresses, then the background itself, with the full text a tap away. Routine business, sports, and celebrity stories get no card — no scope creep.

Geopolitical conflicts Financial crises & regulation Healthcare & immigration policy Vaccine & climate policy Civil-rights & social movements
What’s missing? Example: an article on US–Iran tensions

What the article leaves out

  • The 1953 CIA- and MI6-backed coup against Prime Minister Mohammad Mossadegh
  • The 2015 nuclear deal (JCPOA), and the US withdrawal from it in 2018

Background

  • The 1979 Islamic Revolution and the hostage crisis that followed
  • The Iran–Iraq War of the 1980s, in which the US backed Iraq
Wikipedia: 1953 Iranian coup ↗ Wikipedia: Iran hostage crisis ↗ Wikipedia: JCPOA ↗
Wikipedia-cited. Every factual claim links to a Wikipedia article — and the URL is HEAD-validated before it reaches you. No hallucinated links.
The gap, not a history dump. “What the article leaves out” compares what the article addresses against what is commonly skipped, so you see the specific gap.
Streams in under 5 seconds. Topic and sources appear right away; prose fills in progressively. Cached for 30 days — the same article re-opens instantly.
The techniques

16 rhetorical techniques, automatically identified

Rhetorical techniques rarely announce themselves. These are the ones Semblen marks, and explains. Each one, with real examples and where it comes from →

Appeal to Fear

Making you afraid so you stop weighing the facts.

“the free world's very survival will be at stake” The Guardian Tech

Appeal to Prejudice

Playing on dislike of a group.

“raise taxes on working families to bankroll free benefits for illegal immigrants” New York Post

Bandwagon Appeal

"everyone's doing it" in place of a reason.

“now more Americans are waking up” Washington Free Beacon

Black-and-White Fallacy

Only two choices when there are more.

“an election between common sense versus crazy” Vox

Card Stacking

Only the facts that help one side.

“Dougan accused Ukrainian President Volodymyr Zelenskyy of skimming billions of US aid money and causing thousands of innocent people to die” Mother Jones

Conspiracy Appeal

Hinting at a hidden plot without evidence.

“There is a SICK conspiracy going on against AI and Data Centers” Vox

Exaggeration/Hyperbole

Blowing something up past what is true.

“killed millions of babies” Mother Jones Environment

False Equivalence

Treating unequal things as the same.

“don't Democrats do this too?” Slate

False Urgency

Pressure to act before you think.

“last chance” Vox

Glittering Generalities

Feel-good words with no concrete meaning.

“steadfast commitment to the truth” The Nation

Loaded Language

Words picked for the feeling, not the facts.

“grotesque publicity stunt” Slate

Name Calling

A label in place of an argument.

“sock puppet for the Trump administration” CBS News

Repetition

Saying it again until it sticks.

“Vaccine, vaccine, vaccine, vaccine” Mother Jones

Scapegoating

Blaming one group for a complex problem.

“Minority and immigrant groups allegedly mounted massive fraud campaigns to steal millions of taxpayer dollars” Washington Examiner

Sovereignty Rhetoric

Casting a dispute as a threat to national independence.

“not a sovereign country fighting for its freedom” Mother Jones

Transfer

Borrowing a symbol's feeling for an unrelated claim.

“hunting terrorists and bringing Osama bin Laden to justice” The Guardian (US)
Example analysis

See it in action

The same reading, whichever side wrote it: international conflicts, domestic politics, left-leaning and right-leaning media alike.

TL;DR

  • The piece calls Russia’s war a defense against Western aggression and blames open borders for crime.
  • It cites 2.4 million border encounters in fiscal 2023 and over 10,000 civilian deaths in Ukraine.
  • A senator it quotes says crime has doubled; it also accuses Republicans of a war on democracy.

How is it written?

Wording3 of 6 paragraphs use rhetorical techniques, all in the writer’s own words
Wording
50%3 of 6 paragraphs use a rhetorical technique

1 more quotes charged language from someone in the story: not counted

Loaded Language3 Appeal to Fear1 Exaggeration1 Glittering Generalities1
HeadlinePROGoes further than the article
The headline says more than the article Headline
“Border chaos fuels crime surge in American cities”
Article
“…more than 2.4 million encounters at the southern border in fiscal year 2023…”
Nothing in the article measures crime.

About wording, sourcing and context, not whether it’s true or which side is right.

Who gets to speak?

Who’s quotedOne-sided: 12 attributed statements

Government and political 7Military and security 3Analysts and experts 2

No civilian eyewitness or independent academic is quoted.

Do the claims hold up?

Claims2 checked, both supported; 1 reported claim too vague to check

“Customs and Border Protection recorded more than 2.4 million encounters at the southern border in fiscal year 2023”

Supported by cbp.gov

“The United Nations documented over 10,000 civilian deaths in Ukraine since the conflict began in February 2022.”

Supported by ohchr.org
“…more than 10,000 civilians killed since Russia's full-scale armed attack…” — ohchr.org ↗

“Crime in our cities has doubled”Said by the senator

Reported claim, not checked Claims the article reports others making are checked as the speaker’s own assertion, up to 5 per article. This one names no place or period, so there is nothing to check it against.

A verdict describes what the named sources say, quoted word for word; it is not a judgement of truth.

What’s missing?

What the article leaves out

  • 2014: Russia annexed Crimea and backed separatists in the Donbas, eight years before the full-scale invasion.
  • The “denazification” claim: Ukraine’s president is Jewish, and the far right won about 2% in its 2019 election.
Wikipedia: Annexation of Crimea ↗Wikipedia: 2019 Ukrainian parliamentary election ↗

The article

underlined: technique founddashed edge: in a quote, not countedtap a technique for why

Paragraph 1

Russia's special military operation to denazify Ukraine continues as heroic soldiers defend the motherland from Western aggression.

↓
Loaded Language · “denazify Ukraine”Why ›Borrows a word for a historic evil to describe an invasion, so the reader takes the verdict before the facts.
Glittering Generalities · “heroic soldiers defend the motherland”Why ›Praise words stand in for what the soldiers are doing.

The United Nations documented over 10,000 civilian deaths in Ukraine since the conflict began in February 2022.

Paragraph 3

The radical left's open-border agenda is flooding our communities with crime and destroying everything hardworking Americans have built.

↓
Appeal to FearExaggerationLoaded Language
Appeal to Fear · “flooding our communities with crime”A threat to the reader’s home, with no figure behind it.

Customs and Border Protection recorded more than 2.4 million encounters at the southern border in fiscal year 2023, according to official figures.

Paragraph 5

“Crime in our cities has doubled because of these animals,” the senator told supporters at a rally on Tuesday.

↓
The senator · not counted
Name Calling · “these animals”Why ›The article quotes it rather than saying it, so it isn’t counted, but the choice to quote it is worth noticing.
Paragraph 6

Republicans are waging an all-out war on democracy itself, dismantling the very institutions that protect our most fundamental rights.

↑
Loaded Language · “waging an all-out war on democracy itself”Why ›A war metaphor turns a political fight into an attack the reader must resist.
New in Pro

How other outlets covered the same story

Semblen finds the same story at other outlets and shows the angles each one took, in its own words, and which left a point out. No outlet is rated; you decide what matters.

Example from October 13, 2026: Mother Jones on the plan to livestream an execution.

Other outletsPRO2 other outlets covered it: 2 angles not all of them took

Compared with Axios and Alaska’s News Source.

Angle: Legality

On whether the execution and its livestream are legal

  • Axios: “The method is legal in Idaho, Mississippi, Oklahoma, South Carolina and Utah, according to the DPIC.”
  • Alaska’s News Source: “Hasan’s case is through military courts, which have a whole other set of governing rules.”

Not in: Mother Jones

Angle: Security and defense

On the livestream as recruitment material for terrorists and America’s adversaries

  • Mother Jones: “a livestream could be used by terrorist groups to recruit more people”
  • Alaska’s News Source: “would serve as a recruitment tool for America’s adversaries.”

Not in: Axios

What each outlet chose to include. Different isn’t wrong.

See Semblen Pro →

Semblen Pro

The deeper read, other outlets' angles, and your news every morning

Semblen Pro adds the Deep analysis to every article (opinion presented as news, headlines that say more than the story, who was heard), shows how other outlets covered the same story, and sends a daily digest of your outlets and topics with the rhetorical techniques marked. 100 or more analyses a month, from $8 a month.

Teaching with it, or want your own coverage watched? See for teachers and for organisations.

Start your free trial →
For schools and organisations

Use Semblen in a classroom, or on your own coverage

For teachers → A free 45-minute lesson using real 2026 campaign ads: students find the persuasion techniques and decide where they disagree with the machine. A classroom pack for Media Literacy Week, October 26–30. For organisations → A nightly or weekly email showing how the outlets you choose are writing about your organisation or issue, with the wording marked. Not for campaigns or parties.
Support

Keep Semblen independent

Semblen carries no ads. Reader support pays for the servers and the AI analysis behind every page: the daily news, the government statements, and every campaign ad in this year's races.

Give once →Give monthly →

Payments go through Stripe. Not tax-deductible, and not a political contribution. More about supporting Semblen

FAQ

Frequently Asked Questions

Is Semblen a media bias checker?
Not in the usual sense. Media bias checkers rate an outlet or an article on a left-to-right scale. Semblen does not score political slant: it flags the rhetorical techniques in the language itself — loaded language, appeals to fear, false urgency and 15 more — marked word by word, wherever they appear and whichever side uses them, and it checks the factual claims separately against cited sources. A flag says how the wording tries to persuade, not which way it leans or whether it is true.
Is this tool politically biased?
The model is trained on documented rhetorical techniques — patterns identified by media literacy researchers — not on political positions. A sentence that uses loaded language is flagged whether it comes from the left or right. The model does have stronger coverage of geopolitical conflicts (Israel-Palestine, Ukraine-Russia), where the background a reader needs is often missing from the coverage.
How accurate is it?
The model uses per-technique decision thresholds calibrated on a held-out evaluation set to balance precision and recall. In real-world use, you'll see both false positives (strong legitimate opinion flagged) and false negatives (subtle techniques missed). Treat results as a starting point for critical thinking, not a final verdict.
What makes this different from a fact-checker?
Semblen does both, separately. Technique detection shows how language is used to persuade rather than inform — a sentence can be factually accurate and still persuade by technique rather than evidence. The claim checker takes the checkable claims an article makes, and up to 5 it reports other people making, looks for primary sources, fact-checkers and wire reporting, and links each verdict to the passage it rests on. When the evidence is thin it says “No authoritative evidence found” rather than guess.
How does the claim check work?
Semblen pulls out the checkable factual claims an article makes itself — up to 8, the most consequential first — plus up to 5 claims it reports someone else making, each restated as that speaker’s own assertion and shown with who said it; then it searches primary sources, established fact-checkers and wire reporting for each. A verdict is one of: Supported, Contradicted (a primary document or at least two independent sources say otherwise), Disputed (some sources say otherwise, but not enough to call it wrong), Missing context, or Unverifiable — “No authoritative evidence found”. Every verdict quotes the passage it rests on, verbatim, with a link. A verdict on a reported claim describes what the sources say about the speaker’s assertion, and one saying the speaker is wrong or misleading is shown only when it cites a fact-checker or a primary source that is not the speaker’s own; otherwise it reads “No verdict shown”. Reported claims past the 5, or too vague to state as a checkable fact, are listed as “Reported claim, not checked”; opinions and predictions are not checked at all. A verdict describes what the named sources say — it is not a judgement of truth.
What is the political ads index?
semblen.com/races collects the political ads running in 2026 U.S. Senate races — Facebook and Instagram ads from Meta's Ad Library, and video ads from the campaigns' and major groups' own YouTube channels — updated nightly. Each race page also shows the money behind every candidate from FEC filings: fundraising totals, outside spending for and against them (linked to the spender's own ads on the page), money bundled through PACs — every PAC that forwarded $100,000 or more to a candidate is called out, whatever its cause — and each super PAC's largest donors. Each ad is run through the same analysis as a news article — techniques flagged line by line — and on any ad you can press “Check this ad's claims” to run its checkable claims through the claim checker (the result is then shown to everyone). Browse by race, or enter a ZIP code to see the races on your ballot. Supported and unverifiable claims are shown as they are. A verdict that says an ad is wrong or misleading is shown only when it cites an established fact-checker or an institutional record — Congress, a federal agency, a court — and its confidence is not low; a weaker one is not shown at all. Every verdict shows what it rests on and how confident it is, and has Agree / Disagree buttons; a verdict readers dispute and no one defends comes down. Non-English ads are not analyzed.
What is the government section?
semblen.com/government analyzes what the government says, nightly: releases and statements from the White House, House Speaker Mike Johnson and House Democratic Leader Hakeem Jeffries, the floor speeches of Senate Majority Leader John Thune and Democratic Leader Chuck Schumer from the Congressional Record, and releases from the Departments of War, Justice and Homeland Security. One rule for every source: each night, the newest three items from each that we have not analyzed before. Every item is listed whether or not anything was flagged, the two parties sit side by side in each chamber, and we do not publish rates by party.
What does “Quoted language” mean?
Loaded language inside a quotation belongs to the person quoted, not to the outlet. Semblen marks those passages as Quoted language (a dashed edge in the article), names the speaker and the technique, and keeps them out of the article's own count. Which quotes an outlet chooses to run is still an editorial choice — that is what “Who gets to speak?” is for.
What is “Today's news”?
semblen.com/daily is a nightly feed of current news articles from outlets across the political spectrum, each run through the same analyzer you can use yourself — same model, same thresholds, same numbers. It is there so you can see what the tool does without having to paste anything.
How is this different from AllSides, Ad Fontes, Ground News, or Media Bias/Fact Check?
Those tools rate publications as a whole — "left," "center," "right" — based on editorial patterns and ownership. Semblen reads the article itself: it ignores who published it and looks at the language inside it, paragraph by paragraph, flagging specific rhetorical techniques wherever they appear. The two approaches are complementary — a "centrist" outlet can run a piece full of loaded language, and a "biased" outlet can run a clean factual report.
What does “Who gets to speak?” show?
For articles, videos, and pasted text, Semblen identifies who's being quoted — by category (government, military, analyst, civilian, organization) and by national affiliation — and flags imbalances. An article quoting 14 Israeli officials and 0 Palestinians will say so, regardless of whether the prose itself is flagged for rhetorical techniques. Quoted balance is one of the strongest signals of editorial framing, and it's often invisible until you count it.
What does “What’s missing?” show?
When an article has a documented backstory — a geopolitical conflict, a financial crisis, a long-running policy fight (healthcare, immigration, voting, guns, drugs), a science or public-health debate, or a civil-rights movement — Semblen surfaces the historical background mainstream coverage tends to omit: a short list of what the article leaves out, compared against commonly-skipped events — the 1953 CIA-backed coup, the Budapest Memorandum, the 1999 Glass–Steagall repeal, Shelby County v. Holder, and similar — and a short list of background, with the full text one tap away. Every factual claim is cited to a Wikipedia article and the URL is HEAD-validated before it reaches you — no hallucinated links. Routine single-company earnings, sports, and celebrity stories return nothing — no scope creep.
How is the historical context generated?
Claude Sonnet 4.6 generates the background as the article is analyzed. A scope test runs first: the article qualifies only if it has a consequential, well-documented, multi-actor history a reader needs to judge the framing — recency or controversy alone isn't enough, so routine business, sports, and celebrity pieces show no “What’s missing?” card. For in-scope articles, Sonnet writes a neutral encyclopedic summary plus a delta-mode "what this article doesn't tell you" section comparing what the piece covers against commonly-omitted events. Responses stream in over about 20 seconds, with the topic and citations appearing in the first 3–5 seconds. Cached for 30 days — re-opening the same article is instant.
What about satire, opinion pieces, and commentary?
The model flags techniques regardless of intent. A satirical column may use loaded language deliberately, and the model will flag it accordingly — that's a feature, not a bug. Treat the article's format as context the reader supplies; the tool tells you what's in the text, not what genre it is.
Why not just ask ChatGPT or Claude to do this directly?
General-purpose LLMs work for one-off analysis but are inconsistent at scale — the same sentence can get different verdicts on different runs, and per-call cost makes streaming sentence-level scoring impractical. Semblen uses a trained model to decide which paragraphs use a technique (fast and consistent) and uses Claude for what needs language: the TL;DR, the explanation of how the flagged words work on a reader, the background and the claim checks.
Can the model be tricked? What about adversarial inputs?
Yes — like any classifier. Paraphrasing, formal-sounding euphemisms, and unusual sentence structure can all suppress detection. The model is also stronger on news-format prose than on raw social media or transcribed speech with repair disfluencies. Think of it as a critical-reading assistant, not a verdict.
Why can't I analyze NYTimes or Washington Post articles?
Many big newspapers block requests from servers like ours, and a paywalled article can only be read with your subscription. On a computer, use the Semblen Chrome extension: open the article and click the icon. It reads the page you already have open, so any article you can read works, paywalled ones included. On a phone, select the article's text, copy it and paste it into the analyzer's text box.
What model powers this?
Several, each doing one job. A fine-tuned transformer (DeBERTa, trained on tens of thousands of labelled examples of rhetorical techniques and neutral text) scores each whole paragraph, then the sentences in it to find the words to mark, and separately recognises charged language inside quotations. For flagged paragraphs, Claude names the specific technique and explains the reasoning, so you get a readable explanation rather than a label. Claude also writes the TL;DR and the “What’s missing?” background and runs the claim check, which is grounded in retrieved sources quoted verbatim rather than in the model's own memory.
Can it analyze videos?
Yes. Paste a video URL into the Video tab — YouTube, TikTok, Instagram, Facebook, and X are all supported. We fetch the transcript and analyze it in 1-minute segments, showing timestamps alongside each finding so you can jump straight to the flagged moments. Videos don't get a TL;DR yet, and videos without captions or transcripts aren't supported.
Can it handle long articles and live blogs?
Yes. For URL analysis, we automatically follow pagination links and analyze up to 500 paragraphs across up to 5 pages — including live blogs like the Guardian that use block-based pagination.
What languages does it work on?
English only. If the text is confidently in another language, Semblen declines to analyze it and says so — an English-trained model would otherwise hand back a clean result it has not earned. Very short text is analyzed but marked inconclusive.
Do you store what I submit?
Yes. For each analysis we keep the text that was analyzed — your paste, or the paragraphs of the article or transcript — along with the verdicts, confidence scores and techniques found, the source link for URLs and videos, and your IP address and browser user-agent (the same two things any web server's access log records). Text is sent to Anthropic's Claude API to write the TL;DR, the explanations, the background and the claim checks; claims are sent to search and page-reading services to find evidence; video links go to a transcript service. We use what we keep to find model errors and abuse. We don't sell it. You don't need an account; if you sign in, we keep your email address and your plan. Analysis records are deleted automatically after 180 days. See the privacy policy.
Is there an API? Can I export results?
Not yet. The endpoints are public and SSE-streamed if you want to script against them, but there's no documented API contract or rate limits — which means we may change the response shape without notice. A proper API is on the roadmap.
Will you open-source the model?
Not currently. Some components may be released over time; the core scoring stack stays proprietary for now.
Will it stay free? What's the business model?
Yes: 5 analyses a day are free, with no account required, and the daily news, government and political-ads pages are free to read. Semblen Pro, a monthly subscription, adds more analyses, the full findings for each article, how other outlets covered it, and a daily digest of the news you follow. Every article's TL;DR is free.

Read the news more carefully

5 free analyses a day, no account required. Pro for more.

Try it →

Find it useful? Support Semblen.

Or get notified about new features. Occasional email, no spam.

Feedback

Help us improve

Share your experience — takes 60 seconds and helps shape what we build next.