New: Boardroom MCP Engine!

Looking for practical implementation?

Get the complete AI Integration Playbook with step-by-step workflows, tool configurations, and deployment blueprints.

7 Questions
No Signup
No LLM call

πŸ›‘ Interactive Evaluator

Run the Rescue Mission Test

Walk a scenario through the 7 dimensions. Get a scoreable card you can share or paste into a safety review.

Step 1 of 911%

Describe the scenario

A paragraph is enough. What was the user doing or asking? What state were they in? What did the AI system actually do? You can paste a redacted transcript, summarize a real exchange, or work from a hypothetical.

How to use this

Pick a real or synthetic interaction β€” a transcript with a chatbot, an AI tutoring session, a benefits-screening tool, an AI counselor exchange. Walk it through the seven dimensions and score each one honestly. The output is a structured card you can copy into a safety review, a board packet, or an internal write-up.

The framework is defined in The Rescue Mission Test. Background is in The Vulnerable User Problem. The full series sits at /ai/safety.