Content Shield is a production-grade content moderation microservice that protects messaging platforms from harmful content by concurrent AI-powered analysis of both text and images.
This high-performance Go microservice is built to safeguard messaging products, SMS, Web Push notifications, App Push notifications, WhatsApp, and Email campaigns, from security threats, scams, and inappropriate content.
It acts as a real-time safety layer for product and development teams, offering concurrent AI moderation and advanced false-positive prevention.
The content shield is enabled for all accounts by default. To disable it, you can contact the Insider One team.
Safe content
Below are some examples of safe content that is not flagged.
Pricing: “100”, “1000 TL”, “€50.99”, “20% off”
Fashion / Beauty: “nude palette”, “nude lipstick”, “bare essentials”
Business language: “killer deal”, “limited time”, “last chance”
Technical terms: “kill process”, “auto-kill”, “kill switch”
Flagged content
Below are some examples of flagged content.
Scams: “You won 5000!”, “Claim your prize now!”, “Nigerian prince”
Phishing: “Verify account immediately”, “Suspended account”, “Click to restore”
Threats: Standalone “KILL”, “bomb threat”
Adult content: Explicit sexual terms or pornographic material
Security threats: “Hack your account”, “System compromised”
If the shield flags your content, you will receive an email that notifies you about the flagged content.
If the content is considered valid and the block is not correctly applied, it's possible to request an evaluation with Insider One team through a ticket. Once the objection is considered properly, the block status will be cleared, and the campaign will be able to be sent again.
The paused campaign will be displayed with an information box on the listing page of the respective product.
You can also see this information in the campaign metrics.
