FeatureOct 9, 2026

Model Armor includes enhanced prompt injection and jailbreak protection for Workspace data.

Model Armor includes enhanced prompt injection and jailbreak protection for Workspace data. This capability improves detection accuracy and minimizes false positives when screening Workspace content, such as emails, documents, and files.

This capability is available on multi-region endpoints in the US (us) and EU (eu) for templates and floor setting configured with filter version v3 or later, including the Stable and Latest aliases.

Read the original announcement

All Model Armor updates →

Template-specific exclusion rules are available in Preview.
This feature lets you configure dictionary (word and phrase lists) and regular expression rules to mitigate false-positive detections for prompt injection and jailbreak detection and responsible AI filters.
Model ArmorPreviewSep 28, 2026
You can use the following filters in Melbourne (australia-southeast2) with data residency enforcement enabled:
You can use the following filters in Melbourne ( australia-southeast2 ) with data residency enforcement enabled:

Model Armor includes enhanced prompt injection and jailbreak protection for Workspace data.

More from Model Armor

Model Armor includes enhanced prompt injection and jailbreak protection for Workspace data.

More from Model Armor