PRODUCTION SCENARIO
A retailer runs one chatbot on Google Cloud and another on a different cloud. Security wants both to block prompt injection, stop responses that leak personal data, and record which filters fired, without rewriting either application.
Which Google Cloud service meets this requirement?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Model Armor, which screens prompts and responses across clouds and logs what fired
Model Armor screens LLM prompts and responses for prompt injection and jailbreak attempts and for leaked personal data, returns which filters triggered and which did not, and works whether the application runs on Google Cloud or another provider. Gemini's own safety filters only cover Gemini responses, and IAM governs who may call a resource rather than what a prompt contains.
Official Source: https://docs.cloud.google.com/model-armor/overview ↗