Ejento AI
GuidesQuickstartRecipesREST APIsRelease NotesFAQs
GuidesQuickstartRecipesREST APIsRelease NotesFAQs
Ejento AI
  1. Ejento AI Shield
  • Getting Started
  • Basic Operations
    • Features
      • Teams → Projects → Assistants Hierarchy
    • Guides
      • Login/Signup
  • Assistants
    • Features
      • Introduction to Assistants
      • Assistant Access Control
      • Caching Responses for Assistants
      • Assistant Evaluation
      • Evaluation Metrics
      • URL-based Chat Thread Creation and Prepopulation
      • Reasoning Modes
    • Guides
      • Add Assistant
      • Evaluate Assistant
      • Edit Assistant
      • Embed Assistant
      • Delete Assistant
      • Add Favourite Assistants
      • View Assistant Id
      • View Dataset Id
  • Workflows
    • Features
      • Overview and Types of Supervisors
    • Guides
      • Add Workflow
      • Workflow Chat
  • Corpus
    • Features
      • Introduction
      • Corpus Permissions
      • PII Redaction
    • Guides
      • Assistant Corpus Setup
      • Assistant Corpus Settings
      • Corpus Access Control
      • Corpus Connections
      • ETag Setup for Corpus Incremental Refresh
      • View Corpus Id
      • View Document Id
      • Tagging
        • Corpus tagging
        • Document tagging
  • Tools
    • Features
      • Introduction
    • Guides
      • Tools Overview
      • Create External Tool
      • Connect Tool to Assistant
  • Analytics
    • Guides
      • Analytics
  • User Settings
    • Features
      • Ejento AI User Access Levels
    • Guides
      • Assistant Edit Access
      • Add new user
      • Add User in a Team
      • Remove User from a Team
      • View my Access level in a Team
      • View my User Id
  • Chatlogs
    • Guides
      • Chatlogs
      • View Chatlog & Chat thread Id
  • API Keys
    • Guides
      • API Keys
  • Integrations
    • Email Indexing
    • Microsoft Teams
    • Sharepoint Indexing
    • MS Teams Integration Setup
  • Teams
    • Features
      • Introduction
    • Guides
      • Add a Team
      • Edit a Team
      • Delete a Team
      • View Team Id
  • Projects
    • Features
      • Introduction
    • Guides
      • Add a Project
      • Edit a Project
      • Delete a Project
      • View Project Id
  • Ejento AI Shield
    • Introduction
  1. Ejento AI Shield

Introduction

Customizable Guardrails for Enterprise-Grade AI Safety

Deploy AI with Confidence. Govern AI with Guardrails.#

Ejento AI Shield is an enterprise-grade safety layer designed to ensure AI-generated content remains compliant, secure, and aligned with your organizational policies.
Built on NVIDIA's NeMo Guardrails framework and integrated with Azure AI, it provides intelligent, configurable guardrails that adapt to your risk posture, helping you scale LLM adoption responsibly across your organization.
Disclaimer: Ejento AI Shield is available as an add-on feature and requires a separate subscription.

Build Safe, Compliant AI Workflows#

AI has the potential to drive massive efficiencies but without oversight, it can introduce risk:
Hallucinated facts
Data exposure
Sophisticated prompt manipulation
Lack of ethical moderation
Ejento AI Shield equips you with layered, intelligent guardrails that keep your AI systems operating within safe and defined boundaries.

What Ejento AI Shield Does#

Multi-Layer Prompt Protection#

Detects and blocks advanced jailbreak attempts
Prevents ASCII smuggling
Protects against system prompt leakage, LLM evasion techniques, and prompt injection attacks

Intelligent PII Redaction#

Uses Azure AI Language services to detect and redact personally identifiable information (PII)
Offers granular control over:
Entity type to be excluded from the redaction
Specific terms/values to be excluded from the redaction

Factual Grounding#

Performs real-time hallucination detection
Validates AI output against your data sources to maintain factual accuracy

Content Safety Moderation#

Identifies and prevents ethically harmful, offensive, or inappropriate content across a range of categories

Topic Compliance#

Enforces strict topical boundaries
Ensures alignment with your business policies and use cases

Built for the Enterprise#

Cloud-Native Azure Integration#

Runs entirely within your Azure environment
No data leaves your cloud infrastructure—ensuring complete privacy and compliance

Universal LLM Compatibility#

Seamless integration with any LLM-powered application or platform via flexible API architecture

Governance-Ready#

Centralized policy enforcement
Detailed audit logs for enterprise compliance and regulatory needs

Advanced Observability#

Natively supports integrations with observability tools like Langfuse and Sentry

Advanced Admin Panel#

A powerful, web-based interface for real-time configuration management, enabling you to:
Modify guardrail prompts
Adjust model parameters
Configure guardrail's threshold
Manage API credentials (LLM, Azure AI service)
View detailed logs of all requests and responses
All without code deployment
Changes made via the admin panel take immediate effect across your deployment.

Dynamic Prompt Engineering#

Customize prompts for:
Content safety
Jailbreak detection
Topic control
Hallucination prevention
Tune guardrail behavior to match your risk posture and operational requirements—all from the admin interface

Production-Scale Performance#

Built on FastAPI
Supports async processing
Includes database persistence
Robust error handling ensures enterprise-level reliability

The Guardrail Infrastructure for Responsible AI#

Ejento AI Shield empowers organizations in regulated industries, handling sensitive customer data, or upholding strict brand standards, to:
Mitigate AI risks
Adapt guardrails in real time
Ensure AI consistently operates within defined boundaries
All through an intuitive, enterprise-ready admin interface.
Previous
View Project Id