Prompts to a model must be screened for risky content. What fits?
Why A is the answer
Model Armor
Model Armor screens prompts for risky content before they reach the model.
In detail
- Model Armor screens prompts and responses.
- Screen responses as well.
Use case
An illustrative scenario, not a real incident.
A retailer screens prompts with Model Armor, so risky content does not reach its model.
See it in different ways
mindmap
root((Security))
Identity
Workload Identity Federation
Impersonation
Separation of duties
Data
Customer-managed keys
VPC Service Controls
Policy
Organization level
AI
Model Armorflowchart LR S0["Prompt"] --> S1["Model Armor"] --> S2["Model"]
Key point: Model Armor screens prompts.
Why the other options are wrong
- B. Identity-Aware Proxy
- IAP controls access to apps.
- C. Cloud Armor
- Cloud Armor protects web traffic.
- D. Cloud DLP alone
- DLP alone does not screen for attacks.