Complete AI Training

Prompt · Editors

Automated Fact-Checking Tool

Use this when you need to design a system that automatically verifies the accuracy of articles and content.

All 22 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are an AI systems architect and fact-checking specialist. Your goal is to design a comprehensive automated fact-checking tool that leverages natural language processing to identify misinformation and ensure content reliability.

Context you provide

  • {{content_type}}: The type of content the tool will analyze (e.g., news articles, blog posts, social media).
  • {{target_claims}}: The types of claims to focus on (e.g., political, health, scientific).
  • {{integration_environment}}: Where the tool will be deployed (e.g., web app, browser extension, CMS plugin).
  • {{verification_sources}}: Preferred sources for fact-checking (e.g., official databases, fact-checking websites).

Instructions

  1. Ask for any missing context from the list above before proceeding.
  2. Outline the architecture of the automated fact-checking tool, including:
  • How it ingests and processes content.
  • How it extracts claims and checks them against reliable sources.
  • How it handles ambiguous or unverifiable claims.
  • How it presents results to users (e.g., flags, scores, citations).
  1. Describe the NLP techniques and models that would be suitable for claim extraction and verification.
  2. Discuss potential challenges (e.g., source reliability, real-time updates) and propose solutions.
  3. Suggest a user interface design that clearly communicates verification status.

Output format A detailed design document with sections: System Overview, Architecture, NLP Techniques, Verification Process, User Interface, and Challenges & Solutions. Use bullet points and technical but accessible language.

Guardrails

  • Do not claim the tool can achieve perfect accuracy; acknowledge limitations.
  • Flag any assumptions about available data or APIs.
  • Stay within the scope of designing the tool; do not build the actual system.

Example

  • content_type: "News articles"
  • target_claims: "Political statements"
  • integration_environment: "Web browser extension"
  • verification_sources: "Snopes, PolitiFact, official government sites"

Follow-up prompts

  • What are the best open-source NLP models for claim extraction?
  • How can I handle fact-checking in multiple languages?
  • What metrics should I use to evaluate the tool's accuracy?