O

Platform Overview

Understanding Outharm's content moderation platform

Outharm is a content moderation platform designed for modern applications that need to maintain safe, compliant user-generated content. We provide both AI-powered automated detection and human expert review capabilities through a simple REST API.

🎯What We Do

Content moderation is critical for any platform that accepts user-generated content. Whether you're building a social media app, marketplace, forum, or any application where users can post text, images, or other media, you need reliable ways to detect and handle harmful content.

Outharm handles this complexity for you by providing enterprise-grade moderation tools that can detect various types of harmful content including adult material, violence, hate speech, harassment, drug-related content, and self-harm promotion across multiple content categories.

⚙️How It Works

🔧

1. Configure

Define your content structure using Schemas that describe your data fields. Choose which categories of harmful content to detect and set sensitivity thresholds.

🤖

2. Moderate

Submit content via our API for instant AI analysis or queue it for human review. Every submission is tracked and can be referenced later.

📊

3. Monitor

Review moderation results, manage human review queues, and track usage analytics through our web-based Console dashboard.

🧠Core Concepts

Organize your applications into Projects, each with its own API tokens, schemas, and settings. This allows you to manage multiple applications or environments separately.

📋

Schemas & Components

Schemas define the structure of content you want to moderate. Each schema contains Components that represent individual fields like "title", "description", or "images".

🏷️

Categories

Categories represent types of harmful content we can detect, such as Adult Content, Violence, or Hate Speech. You can enable/disable categories and set confidence thresholds.

📄

Submissions

Every piece of content you send for moderation creates a Submission record that tracks the content, results, timestamps, and can be referenced later for appeals or auditing.

Moderation Types

AI-powered instant analysis that provides immediate moderation decisions. Perfect for high-volume applications that need real-time content filtering.

Results in milliseconds
Cost-effective for high volume
Configurable confidence thresholds
Detailed category scoring

Human expert review for content that requires nuanced understanding, cultural context, or complex policy decisions that AI might miss.

Context-aware human decisions
Cultural and linguistic nuance
Complex policy enforcement
Quality assurance and appeals

🔄Typical Integration Workflow

1

Create a Project

Set up your project in the Console and generate API tokens for authentication

2

Define Schemas

Create schemas that match your application's content structure and data model

3

Configure Categories

Choose which content categories to detect and set appropriate sensitivity thresholds

4

Integrate the API

Add our REST API endpoints to your application's content submission workflow

5

Monitor and Optimize

Track performance and adjust thresholds based on your content patterns and requirements

Ready to Get Started?

Ready to add content moderation to your application? Start with our Quick Start guide to get up and running in minutes, or explore the Console to understand the platform better.

Related Documentation