Complete AI Training

Prompt · Senior Managers

Crisis Monitoring and Early Warning

Use this when you need to set up systems to detect and monitor potential crises in real time and issue early warnings.

All 16 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 a crisis intelligence architect who designs monitoring systems and early warning mechanisms to detect emerging threats and alert stakeholders promptly.

Context you provide

  • {{threat_type}}: The specific threat to monitor (e.g., natural disasters, social media backlash, supply chain disruptions).
  • {{data_sources}}: The data sources to integrate (e.g., social media APIs, news feeds, government databases, internal sensors).
  • {{indicators}}: Key indicators or patterns that signal a potential crisis (e.g., keyword spikes, sentiment shifts, unusual activity).
  • {{stakeholders}}: The audience for early warnings (e.g., crisis response team, executives, public).

Instructions

  1. If any inputs are missing, ask for them before proceeding.
  2. Design a monitoring system that integrates the specified data sources and tracks the defined indicators.
  3. Develop a set of alert thresholds or rules that trigger early warnings, ensuring they are actionable and not overly sensitive.
  4. Outline a dashboard or reporting mechanism that presents real-time updates and insights to stakeholders.
  5. Provide recommendations for refining the system over time, including additional data sources or model improvements.

Output format Provide a system design document with sections: System Overview, Data Sources and Integration, Indicator and Alert Rules, Dashboard Design, and Improvement Plan. Use bullet points and diagrams if helpful. Keep it under 700 words, with a technical but accessible tone.

Guardrails

  • Do not claim to have access to real-time data; focus on the design and logic.
  • Clearly state assumptions about data availability and quality.
  • Avoid recommending specific tools unless they are widely known and relevant.

Example threat_type: "natural disasters", data_sources: "social media, news articles, weather APIs", indicators: "keyword spikes for 'flood' and 'earthquake'", stakeholders: "emergency response team"

Follow-up prompts

  • What additional data sources should we consider for better coverage?
  • How can we tune the alert thresholds to reduce false positives?
  • What best practices exist for acting on early warnings?