Quel est l'objectif de la checklist "Decision Tree: Fake News & AI Detection" ?

Multicriteria expert system to analyze web content reliability, detect AI manipulation, and evaluate source credibility.

Comment utiliser le modèle "Decision Tree: Fake News & AI Detection" ?

Vous pouvez démarrer une instance de cette checklist pour l'adapter à vos besoins spécifiques. Elle est composée de plusieurs étapes structurées pour vous guider.

Decision Tree: Fake News & AI Detection

Settings influencing the flow of the checklist

Does the information align exactly with what you already believe? If so, be extra vigilant.

Does the content use terms like 'ALERT', 'SHARE QUICKLY' or 'WHAT THE MEDIA IS HIDING'?

Was the account created recently (less than 3 months ago)? Propaganda accounts are often ephemeral.

Does the account post obsessively about a single divisive topic?

Open a new tab and search for what *others* say about this source or site (e.g., 'sitename + reliability').

Is the article several years old and taken out of context?

Is the text highly repetitive, free of spelling mistakes but extremely neutral and 'smooth'? (ChatGPT marker)

Has the original image been repurposed from another event?

Check hands (fingers), ears, and reflections in the eyes.

Merging objects, illegible text in the background, illogical cast shadows.

Observe teeth and tongue: AI struggles to simulate the inside of the mouth fluidly.

Is there a slight blur or abnormal pixels at the junction between the face and the hair?

Is this video game footage (e.g., Arma 3) presented as a real war zone?

Note the evidence found here (fact-checking links, reverse search results).

Decision Tree: Fake News & AI Detection

Impact and Satisfaction

⚡ Phase 1: Emotional Shock & Bias

Confirmation bias self-assessment

Detecting a sense of urgency

🔍 Phase 2: Source Investigation

Account age analysis

Editorial consistency

Lateral Reading Method

📝 Phase 3: Text Analysis

Freshness check

Synthetic style detection (AI)

🖼️ Phase 3: Image Expertise

Reverse image search (Google/TinEye)

AI generation indicators

Anatomical anomalies

Background inconsistencies

🎥 Phase 3: Video & Deepfake Expertise

Audio/video desynchronization

Contour 'morphing' effect

Sequence origin

⚖️ Investigation Conclusion

Verdict after analysis:

Investigation notes

Recommended checklists