‎Types of Guardrails | Sprinklr Help Center

Support

Sprinklr

Enterprise Offerings
Sprinklr Digital Twin
Sprinklr Service
Overview
Inbound Voice
Outbound Voice
Social Customer Service
Live Chat Support
Communities
Conversational AI Platform
Conversational IVR
Knowledge Base Software
Unified Agent Desktop
Agent Assist
Omnichannel Routing
AI-Powered Quality Management
Supervisor Console
Conversational Analytics
Workforce Management
Reporting and Analytics
Omnichannel Surveys
Guided Workflows
Service Command Center
Sprinklr Social
Overview
Social Publishing and Engagement
Employee Advocacy
Social Marketing for Distributed Teams
Works Best With
Social Listening
Social Advertising
Conversational Commerce
Social Customer Service
Sprinklr Insights
Overview
Social Listening
Competitive Insights and Benchmarking
Product Insights
Location Insights
Visual Insights
Media Monitoring and Analytics
Sprinklr Surveys
Sprinklr Marketing
Overview
Content Marketing and Campaign Planning
Social Advertising
Marketing Analytics
Ad Comment Moderation
Platform
Overview
Sprinklr AI
Sprinklr AI+
AI Studio
Security
Integrations
Sandbox
Display
Presentations
Active Data Retention
API
CPG
Financial Services
Government
Retail
Technology
Telecommunications
Travel & Hospitality
Marketing Teams
Overview
Crisis Management
Content Lifecycle Management
Social Media Management
Compliant Digital Selling
Franchise Digital Management
Executive Reputation Management
Industry Insights and Competitive Intelligence
Creative Insights and Production Optimization
Customer Service Teams
Overview
Omnichannel Contact Center
Customer Self-Service
Digital Customer Service
Outbound Campaign Management
Workforce Engagement Management
Contact Center Intelligence
Social Customer Service
Conversational Commerce
Proactive Digital Engagement
Unified-CXM
Customer Stories
Sprinklr Champions
Sprinklr Community
Our Story
Leadership
Newsroom
Partners
Careers
Investor Relations
Security & Data Privacy
Sustainability
Sprinklr Social ROI Estimator
AI Maturity Assessment
CX-WISE
Analyst Reports
Sprinklr Demo Center
eBooks & Reports
Events & Webinars
Blog
Unified-CXM Guide
Forbes Influential CMO 2024
Our Services
Training
For Agencies
Help Center
Release Notes
Contact Us

Pricing

Home
Platform
AI+ Studio
Guardrails
Types of Guardrails

Types of Guardrails

Articles

(5)

Prompt Guard Guardrail

The Prompt Guard Guardrail acts as a protective layer that evaluates user inputs before they reach the model. Its purpose is to detect unsafe or malicious behavior and prevent the system from being manipulated.Prompt Guard classifies each input into categories such as:Benign – Safe inputs that can b

Sprinklr Service

New Article

20.10

Knowledge Base Article

•

Types of Guardrails

Harmful Content Guardrail

The Harmful Content Guardrail helps prevent the system from generating or engaging with unsafe, abusive, or violent material. This configuration is designed to maintain a safe environment for users and ensure compliance with organizational and regulatory standards.When you enable this guardrail, the

Sprinklr AI+

Sprinklr Service

New Article

20.10

Knowledge Base Article

•

Types of Guardrails

Word Filters Guardrail

The Word Filters Guardrail allows you to block specific words or phrases from appearing in user interactions. This configuration is useful for restricting profanity, sensitive terms, or any language that violates organizational policies.When a filtered word is detected, the system can:Block interact

Sprinklr AI+

Sprinklr Service

New Article

20.10

Knowledge Base Article

•

Types of Guardrails

Denied Topics Guardrail

The Denied Topics Guardrail allows you to restrict the use of specific topics within the system. When you configure this guardrail, the platform actively monitors user interactions for any reference to the denied topics and blocks them from being processed.You can define a list of topics that are no

Sprinklr AI+

Sprinklr Service

New Article

20.10

Knowledge Base Article

•

Types of Guardrails

Protected Material Guardrail

The Protected Material Detection guardrail identifies content in model outputs that may contain copyrighted, trademarked, or otherwise legally protected material. This guardrail helps ensure that generated content complies with intellectual property and legal requirements.When protected material is

Sprinklr Service

New Article

20.10

Knowledge Base Article

•

Types of Guardrails

Core Components

Governance

Sandbox

Integrations

AI+ Studio

Introduction to AI+ Studio

AI+ Studio - Provider and Models Settings

Add Support for OpenAI

Add support for Azure OpenAI

Add Support for Amazon Bedrock

Add Support for Google Vertex

Add Support for BYOM

Add Support for Anthropic

Add Support for xAI

Use-Case Deployments

Supported List of Use-Cases

Manage Copilots

Introduction to Security and Compliance in AI+ Studio

PII Masking

Guardrails

Prompt Engineering Best Practices

Sprinklr Accessibility and Compatibility Features

Sprinklr AI Features

How can we help you?

Products

Sprinklr Service
Sprinklr Insights
Sprinklr Marketing
Sprinklr Social
Unified CXM Platform

Solutions

Industries

CPG
Financial Services
Government
Retail
Technology
Telecommunication
Travel & Hospitality

Teams

Marketing
Customer Service

Company

Our Story
Newsroom
Partners
Careers
Investor Relations
Security & Privacy
Customer Stories

Support

Developer Portal
Integrations
Help Center
Contact Us

Resources

Analyst Reports
Sprinklr Demo Center
Podcasts
Blog
Guides
eBooks
Events & Webinars
Masterclass
Services
Training

Social Media Topics

Social Listening Guide
Social Media Monitoring
Social Media Management
Social Media Glossary

Customer Service Topics

CCaaS Guide
Customer Service
Contact Center AI (CCAI)
Customer Service Glossary

Sprinklr Website Home