P
✍️RedactionAdvancedAll AIs

Real-Time Fact-Checking During Events

Quickly analyze and verify claims made during debates, conferences, and live events.

Paste in your AI

Paste this prompt in ChatGPT, Claude or Gemini and customize the variables in brackets.

Tu es un journaliste fact-checker expérimenté. Ta mission est d'analyser rapidement des affirmations formulées lors d'un événement en direct et d'évaluer leur véracité.

Type d'événement : [Débat politique / Conférence de presse / Table ronde / Webinaire / Discours officiel / Autre]
Thématique principale : [Sujet de l'événement]
Date de l'événement : [Date]
Intervenant(s) : [Noms et fonctions des personnes citées]

Affirmation(s) à vérifier :
[Coller ici la ou les affirmations exactes à vérifier, en les numérotant]

Pour chaque affirmation, produis une analyse structurée :

  1. Affirmation originale : Reprends la citation exacte.

  2. Verdict : Attribue un niveau de véracité parmi :

    • VRAI : L'affirmation est factuelle et vérifiable
    • PLUTÔT VRAI : Globalement exact avec des nuances
    • TROMPEUR : Contient des éléments vrais mais présentés de façon biaisée
    • PLUTÔT FAUX : Les éléments principaux sont inexacts
    • FAUX : L'affirmation contredit les faits établis
    • INVÉRIFIABLE : Impossible à confirmer ou infirmer avec les données disponibles

  3. Contexte factuel : Présente les données, chiffres et faits vérifiés qui permettent d'évaluer l'affirmation. Cite les sources.

  4. Nuances importantes : Signale les éléments de contexte que l'intervenant a omis ou qui modifient l'interprétation.

  5. Sources de référence : Liste les sources fiables consultées (organismes officiels, études, bases de données).

Règles à respecter :

  • Reste strictement factuel et non partisan
  • Distingue les faits des opinions
  • Signale explicitement quand tu n'as pas assez d'informations
  • Privilégie les sources primaires (INSEE, Eurostat, OMS, etc.)
  • Date tes données pour éviter les anachronismes

Personalize this prompt with Léa

Léa rewrites this prompt for your job and your exact goal — 3 quick questions.

Why this prompt works

<p>This prompt transforms AI into a structured fact-checking assistant. The six-level veracity scale and the requirement for primary sources ensure rigorous and nuanced analysis, essential for journalism and information monitoring.</p>

Use Cases

Fact-check a live political debateVerify figures from a press conferenceAnalyze claims from an industry webinar

Expected Output

Structured veracity analysis with verdicts, context, and sources

Improve this prompt

Run this prompt through the Optimizer to strengthen its context, constraints and expected format.

Improve this prompt with the Optimizer

Comments

  • LéaAI

    Pour un usage en direct, ajoutez une contrainte de délai (ex : « analyse en moins de 30 secondes ») et demandez à l’IA d’indiquer le temps de recherche estimé pour chaque source, afin de prioriser les vérifications les plus fiables et rapides.

Glossary Terms

ContextEmbeddingFew-Shot PromptingFine-TuningGPTGuardrailsInferenceLLM

📬 Get new prompts every week

Join our newsletter and never miss a prompt.

Go further

Improve this prompt with the Optimizer

Analyze the prompt, strengthen the context and get a more precise version.

Prompt Engineering

Cross-model principles for writing more reliable prompts.

Similar Prompts

✍️RedactionIntermediateAll AIs

GitHub Copilot Prompt for Translating a Text

GitHub Copilot, the AI development assistant created by GitHub and OpenAI, is not limited to code generation. Integrated directly into your editor (VS Code, JetBrains, Neovim), it can also serve as a powerful translation tool for your multilingual projects. Whether you need to translate localization files (i18n), code comments, technical documentation, or user interfaces, Copilot understands your project context and produces translations consistent with your existing terminology. Its main advantage lies in its ability to maintain source code formatting intact — HTML tags, interpolation variables, JSON keys — while translating only the textual content. This editor-integrated approach avoids back-and-forth with external tools and ensures your translations respect your codebase's technical conventions. Here is an optimized prompt to fully leverage GitHub Copilot in your translation tasks.

0417
✍️RedactionIntermediateAll AIs

Perplexity Prompt for Translating a Text

Perplexity stands out from traditional translators due to its ability to understand the overall context of a text before translating it. Unlike Google Translate or DeepL, which work sentence by sentence, Perplexity analyzes the intent, register, and domain of the content to produce a coherent and natural translation. The tool excels particularly for technical, marketing, or literary texts where tone and cultural nuances matter as much as lexical accuracy. By crafting a structured prompt, you can guide Perplexity to adapt the language level, respect sector-specific terminology, and preserve the original author's style. This prompt-based approach also allows you to request justifications for translation choices, making it a linguistic learning tool as well as a simple translator. Whether you are translating a business contract, a blog article, or personal correspondence, a well-constructed prompt turns Perplexity into a contextual translator capable of rivaling a human professional.

0477
✍️RedactionAdvancedAll AIs

Write Professional Blog Posts and White Papers

Create engaging blog posts and professional white papers that strengthen your industry expertise.

01299
✍️RedactionAdvancedAll AIs

Climax Scene Brainstorm

Planning or revising your story's climax

0467