👋Step-by-step instructions

Unitary's content classification API is designed to classify harmful content accurately.

Integrating Unitary's API will help your organization effectively moderate user-generated content and ads, reduce user reports, protect brand reputation, and maintain strong partnerships to ensure the continued success of your platform.

Unitary's AI will allow your human teams to focus on higher-value initiatives.

You'll be able to stop worrying about users and advertisers losing trust in your platform.

If you need help or have any feedback, please get in touch with us at [e-mail].

Steps to integrate with Unitary's API

1. Request an API Key from Unitary ([e-mail]).

2. Decide which API endpoint(s) are best for your use case:

   • The Items and Characteristics endpoints provide granular signals for identifying harmful or sensitive elements within content, such as 'smoking_and_tobacco' or 'knife'.

   • The Detoxify endpoints should be used for identifying harmful or offensive language in user-generated text content, such as captions or comments.

   • The Standard & Premium: Policy Classification endpoints provide a more advanced understanding of content that goes beyond simply detecting potentially harmful signals - by taking context and intent into account.

   • The Virtual Moderator endpoint provides an easy way to integrate with our Virtual Trust & Safety Agents which leverage a blended solution of AI and humans for our most advanced understanding of content, giving you human-level accuracy quickly and reliably.

3. Integrate the relevant API endpoint(s) - it's recommend you start by reading the Features and requirements and API Authentication.

4. (Optional) Integrating Webhooks will allow you to build a more robust integration that enables you to act on the API classifications as soon as they have been processed.

Your integration experience should be as seamless and straightforward as possible. If you encounter any issues or have feedback to share, please reach out to [e-mail].