See Through Media

Augmented Media Literacy

How to Use

A reader's guide to installing our browser extension and deconstructing the news with objective AI analysis.

Part I

Installing the Chrome Extension

Augmented Media Literacy works directly inside your web browser as you read the news. Installing our verified extension takes less than a minute through the official Chrome Web Store.

1

Open the Chrome Web Store

Launch Google Chrome and navigate to the Chrome Web Store.

2

Search for the Extension

Use the search bar at the top left of the store to type in "Augmented Media Literacy".

3

Add to Chrome

Select our extension from the list and click the blue "Add to Chrome" button located at the top right of the page.

4

Confirm and Pin

When the permissions pop-up appears, click "Add extension". For the best experience, click the puzzle piece icon in your Chrome toolbar and pin Augmented Media Literacy for easy access.

Part II

Launching the Analysis Engine

Once installed, Augmented Media Literacy acts as an unobtrusive analytical lens. It automatically scans web pages in the background to detect genuine news reporting and stands ready to assist whenever you want a deeper look.

Is this news?

Our extension features intelligent page detection. When you visit a genuine article, our floating badge appears in the bottom-right corner. If you navigate to a shopping site, search engine, or home page, the engine automatically pauses to keep your browsing clean and conserve processing power.

On-Demand Analysis

Click the floating badge on any article to open your side analysis panel. Simply click "Execute Analysis", and our computational linguistics engine will immediately deconstruct the page — analyzing tone, separating viewpoints, and comparing headlines in seconds.

Part III

The Analytical Toolkit

We believe an AI should act as an objective X-ray, not a referee. Our tools never tell you what to think; instead, they transparently highlight how a story is constructed so you can form your own unbiased conclusions.

Diagnostic 1

Headline Comparison & Clickbait Taxonomy

Publishers often use different headlines for different audiences — sometimes using a calm title on their home page while testing sensational, click-driven headlines on social media or search engines. We compare three distinct headline tiers simultaneously:

  • On-Page Headline: The main title visible when you read the article.
  • SEO Search Title: The headline optimized to attract clicks on search engines.
  • Social Share Title: The headline embedded in metadata for sharing on social platforms.

The 7-Category Engagement Taxonomy

When evaluating headlines, our engine checks against seven standard psychological engagement hooks. Inside the extension, you can hover over any of these categories for interactive definitions and examples:

1. ListiclePromises an easy, numbered, digestible read (e.g., "7 Superfoods to Boost Your Memory").
2. CliffhangerAbruptly cuts off information to force a click (e.g., "What Happened Next Stunned Everyone").
3. Provokes EmotionsUses exaggerated, highly charged language to bypass logic and trigger outrage or shock.
4. Curiosity GapWithholds critical context, telling you that something happened without explaining what or why.
5. Challenge to EgoAppeals to pride or identity by daring the reader (e.g., "Only 1% of People Can Solve This").
6. Defies ConventionDirect contradictions of established consensus or common sense to generate intrigue.
7. Induces FearPreys on personal anxieties, health concerns, or safety by framing everyday items as looming threats.
Diagnostic 2

Emotional Density & Tone

Objective reporting relies on verifiable facts rather than dramatic adjectives or subjective framing. This tool evaluates the distribution of emotive language across an article with complete transparency.

Dual-Metric Word Counts

We calculate emotional density using two distinct measurements: the article as a whole vs. the reporter's own narrative voice. To ensure complete clarity, we display exact fractions (e.g., "52 out of 1,146 words"), allowing you to see exactly how much emotion comes from external quotes versus authorial commentary.

Observational Tone

Rather than labeling an article as "good" or "bad," our engine provides a strictly observational summary of where subjective words appear (e.g., highlighting when dramatic terms like "combative" or "fiery" are used to frame an event).

"Emotion does not make reporting inherently invalid, but measuring its density is a powerful indicator of how subjective or narrative-driven an article is."

Diagnostic 3

Source Attribution & Viewpoint Diversity

A well-balanced article incorporates multiple perspectives. However, simply quoting ten different people does not mean an article is balanced if all ten people agree. Our engine separates people from ideologies.

Entities Quoted

An exhaustive extraction of every human speaker, official organization, or document cited in the text.

Distinct Viewpoints

We group quoted sources by the actual arguments or ideological stances they represent, showing true perspective diversity.

Share of Voice

A visual donut chart and data table displaying the exact percentage of text volume dedicated to representing each viewpoint.

Perspective Diversity Diagnostic: Articles are automatically evaluated and rated as presenting either Multiple Distinct Perspectives, a Primary Perspective with Minor Stances, or a Single Dominant Perspective.
Diagnostic 4

Sentence Consensus

Coming soon

What is Augmented Media Literacy Consensus Analysis?

When a major news event breaks, our editors track how it is reported across the wider media landscape. If this article were part of an actively tracked event, this tab would reveal verbatim overlaps (Lexical Consensus) and ideological framing differences (Semantic Consensus) from thousands of other publishers, helping you isolate universal facts from publisher-specific spin.

Our Editorial Criteria

Like editors of a major publication, we focus our computational resources on the news that is significantly impacting the world each day. Our team selects 1 to 5 major events per day for deep consensus analysis based on strict criteria:

  • Volume & Velocity: The topic must generate more than 500 distinct articles across the media landscape within a 48 to 72-hour window.
  • Substance: We filter out brief updates and wire summaries, focusing our analysis on articles containing more than 300 words.
  • Complex Framing: We prioritize events that trigger multiple points of view, diverse opinions, and debated narratives.

Lexical Consensus: Identifies when independent reports use the exact same verbatim words, which often indicates reliance on a single shared press release or wire service.

Semantic Consensus: Identifies when multiple outlets confirm the same underlying factual meaning, even if described using completely different words.

Part IV

Interactive Reading Features

We designed our interface to respect your vertical screen space and seamlessly integrate into your normal reading routine.

In-Page Highlighting

When analysis is complete, our engine gently highlights potential emotional words directly within the article text using soft purple underlining. Hover your mouse over any highlighted word on the page to see an instant tooltip explaining its linguistic framing and identifying who spoke it.

Collapsible Accordions

To keep long articles from overwhelming your screen, detailed breakdowns — such as individual word cards, full source tables, and headline tier comparisons — start neatly tucked inside collapsible drawers. Click any section header to expand or close the details as you explore.

Educational Tooltips

Anytime you encounter a technical media literacy term inside the extension — such as "Curiosity Gap" or "Challenge to Ego" — the text is underlined with dotted formatting. Hovering over the term opens a floating educational guide with definitions and real-world examples.

Ready to Go?