BiasChecker.ai

Our Methodology

Every BiasChecker.ai analysis is an AI-assisted close reading of an article, run under a fixed set of instructions we have refined over thousands of analyses. The instructions are the same whichever AI model you choose — model choice changes the judge, never the rules.

This page explains how an analysis runs, the principles every analysis follows, and what each of the eleven analysis types looks for.

How every analysis runs

  1. Content capture. The extension extracts the article's main content from the page and checks the extraction quality before submitting — page furniture like menus, ads and cookie banners is stripped out.
  2. AI analysis. The article and the analysis instructions go to the AI model you selected, which reads it closely and reports what it finds.
  3. Evidence verification. Each finding must cite verbatim quotes. We locate every quote in the original article; quotes we find are marked verified, quotes the model paraphrased are shown but not counted as verified.
  4. Scoring — by our code, not the model. Severity ratings and overall levels are computed from what was found, using fixed classifications of each bias category and technique.
  5. Re-checks are instant. If an article has already been analysed with the same model, you get that result back straight away rather than waiting for a fresh run — and subscribers are never charged for it.

Principles every analysis follows

  • No model-emitted scores. AI models are not calibrated judges of their own findings — the same article can get different numbers from different models. All scoring is done by our code from the findings themselves, so the same evidence always produces the same rating.
  • Verbatim, verifiable evidence. Findings must quote the article exactly, and we check every quote against the source text. You can always see — and judge — the evidence yourself.
  • Decisive about the text, careful about people. Analyses state plainly what the writing does, but stay conditional about unverified claims and never attack the character or motives of a named person. Findings describe the article, never the journalist.
  • Honest about knowledge limits. Analyses that touch real-world facts are instructed to trust the article over potentially outdated training data for recent events, and never to "correct" the article from stale knowledge.

The eleven analyses

Each analysis type gives the model a different expert role and a different question to answer about the same article.

Intent Analysis

Identifies what the text says versus what it steers readers to believe or do. Intent is treated as a property of the writing — the direction the text pushes — never as a claim about the author's private motives. The result names the text's primary purpose and direction, and gives you a quick verification question you can check for yourself in seconds.

Persuasion

Looks for recognised persuasion techniques — emotional appeals, framing devices, logical fallacies — and explains how each one works on the reader, quoting the exact passage where it appears. How serious each technique is comes from our fixed catalogue of techniques, not from the model's mood on the day.

Bias Analysis

Describes how the text leans, with reference to specific words and structural choices, across our catalogue of bias categories — framing, omission, loaded language, source imbalance, political slant and more. Each finding cites the passages that drive it, and the overall bias level is computed by our code from the categories found.

Historical Parallels

A historian's lens: finds past events and patterns that genuinely parallel the situation described — analogies, not a timeline of the topic itself — and explains how each parallel resembles the present case and how it played out, ranked by relevance.

Legal Risk

Assesses whether actions or proposals described in the text raise questions under applicable law, working out the relevant jurisdiction from the text itself. It stays strictly conditional about people and unverified facts: it flags what would need checking, it does not pronounce anyone guilty.

Moral Lens

An applied-ethics read of both the writing and the conduct it reports: how the text treats its subjects and its readers, and how the actions described measure against widely shared moral principles — keeping the two judgements clearly separate.

Scientific Assessment

Checks the scientific rigor of the text: whether claims are supported by the evidence offered, how methodology and statistics are used, what is established versus genuinely contested, and where the text overclaims beyond what its own sources can carry.

Critique

The "does it hold up?" lens. Instead of cataloguing rhetorical tricks, it evaluates the substance of the approach, argument or plan: are the goals right, is the approach sound, is it feasible, and what is being missed — engaging the strongest version of the argument before critiquing it.

Omissions

A dedicated "what's missing" audit: what a well-informed reader would expect to see but the article leaves out, why each gap changes the impression the article leaves, and what kind of source would fill it. Only consequential, established context counts — and a thorough, balanced article comes back with an empty list.

Background

A neutral briefing for someone landing mid-story: what this topic is, how it developed, and where it stands now. Pure orientation — factual, concise, and without taking sides — rather than a critique of the article.

Rhetorical Roast

A warm, playful media critic that turns genuine weaknesses in the writing and reasoning into lighthearted rhetorical questions. The flaws must be real — the comedy is in the angle, and a well-written piece gets nothing to roast.

A living method

The analysis instructions are continuously refined — clearer guardrails, sharper checks. The principles above are the stable part: they survive every revision, and they are what we ask you to hold us to.