Policy: GCP > Model Armor > Floor Setting > Settings > Responsible AI > Hate Speech

Minimum confidence level at which the live floor's Hate Speech RAI filter must be enforced.

When set to any value other than None, the Settings control alarms unless the live filterConfig.raiSettings.raiFilters[] contains an entry for HATE_SPEECH whose confidenceLevel is the configured value or stricter. The Set Floor Setting action inserts or raises that entry on Enforce: Enabled per .... None means the Hate Speech filter is not required.

Targets

This policy targets the following resource types:

GCP > Model Armor > Floor Setting

Primary Policy

This policy is used with the following primary policy:

GCP > Model Armor > Floor Setting > Settings > Responsible AI

Controls

Setting this policy configures this control:

GCP > Model Armor > Floor Setting > Settings

Policy Specification

Schema Type	`string`
Default	`None`
Valid Values [YAML]	`None` `Low and above` `Medium and above` `High`
Examples [YAML]	Medium and above

In Your Workspace

Policy Settings by Type report

Developers

Category URI

tmod:@turbot/turbot#/control/categories/security

Policy Type URI

tmod:@turbot/gcp-modelarmor#/policy/types/floorSettingSettingsResponsibleAiHateSpeech

GraphQL

query policyType(id: "tmod:@turbot/gcp-modelarmor#/policy/types/floorSettingSettingsResponsibleAiHateSpeech") { … }

query policySettings(filter: "policyTypeId:'tmod:@turbot/gcp-modelarmor#/policy/types/floorSettingSettingsResponsibleAiHateSpeech'") { … }

query policyValues(filter: "policyTypeId:'tmod:@turbot/gcp-modelarmor#/policy/types/floorSettingSettingsResponsibleAiHateSpeech'") { … }

CLI

Get Policy Type

turbot graphql policy-type --id "tmod:@turbot/gcp-modelarmor#/policy/types/floorSettingSettingsResponsibleAiHateSpeech"

Get Policy Settings

turbot graphql policy-settings --filter "policyTypeId:tmod:@turbot/gcp-modelarmor#/policy/types/floorSettingSettingsResponsibleAiHateSpeech"