ORTHOS: News Manipulation Detector

ORTHOS is a news analysis system that uses text embeddings to detect emotional framing and linguistic manipulation in news headlines.

Based on the TRIBE V2 research, ORTHOS maps headlines against neural coefficients to estimate the likely impact of language on the brain—distinguishing between Subcortical Arousal (emotional triggers) and Cortical Load (analytical processing).

---

How it Works

The engine analyzes headlines using a linguistic mapping system derived from fMRI research.

• S-Arousal (Subcortical): Identifies language patterns associated with emotional spikes and threat detection.
• C-Load (Cortical): Measures the density of factual and logical markers that require analytical processing.
• Subject Variability: Simulates how different psychological archetypes (Analytical, Emotional, and Tribal) might respond to the same text.
• Position Weighting: Prioritizes the beginning of the headline, where linguistic "hooks" are most effective.
• Context Gap Detection: Flags headlines where high-arousal adjectives are used without supporting factual context.
• Performance: Fast local inference (~500ms-1s) using all-mpnet-base-v2.
• Privacy: 100% local execution.

---

Supported Outlets

ORTHOS is optimized for and officially supports real-time extraction and analysis across the following major news platforms:

• AP News
• BBC
• Bloomberg
• CNBC
• CNN
• Fox News
• NBC News
• Newsweek
• NPR (manual page refresh required)
• Politico
• Reuters
• The Guardian
• The Hill
• The New York Times
• The Wall Street Journal
• The Washington Post
• Time (manual page refresh required)

---

Installation

1. The Backend

1. Clone the repository:

   git clone https://github.com/ItsNotGold/orthos-neural-detector.git
   cd orthos-neural-detector

2. Install dependencies:

   python -m pip install -r requirements.txt

3. Start the server:

   python -m orthos_backend.server

2. The Chrome Extension

1. Open Chrome and go to chrome://extensions/.
2. Enable Developer mode.
3. Click Load unpacked and select the orthos_extension folder.
4. Pin the extension to the chrome Toolbar
5. Open the extension popup and ensure the port matches your server (usually 8000).

---

Classification Levels

• 🔴 High Arousal Risk: Headlines using markers designed to trigger immediate emotional responses.
• 🟠 Sensationalist: Moderate emotional framing used to amplify interest.
• 🟢 Objective / Neutral: Factual reporting that supports rational evaluation.

---

Support the Knight First Amendment Institute

Every day, we scroll through a digital landscape engineered to bypass our logic and trigger our emotions. While ORTHOS helps visualize and detect these manipulation patterns, lasting change requires legal and systemic action.

To support this mission, I have set up a GoFundMe campaign: Supporting the Knight First Amendment Institute.

• 100% of the proceeds go directly to the Knight First Amendment Institute at Columbia University to support their legal fight for free speech, public transparency, and digital privacy.
• ORTHOS is completely open-source and free; I do not receive any personal funds from this campaign.

---

Status & Credits

• Foundation: Built on the TRIBE V2 research architecture and calibrated against narrative stimuli from the ds003020 fMRI dataset.
• Project Status: ORTHOS is an active passion project. Contributions are not accepted at this time.
• License: CC BY-NC 4.0

Disclaimer: ORTHOS is an experimental tool for media literacy. Scores are predicted linguistic impacts based on research patterns and should be used as one of many signals for evaluating content. It does not measure actual brain activity.