Alibaba Cloud Content Moderation is a pioneer in the field of content moderation. It originates from Alibaba Cloud's many years of security technology accumulation and relies on the content governance experience, AI algorithms, and cloud computing infrastructure in multiple fields such as e-commerce, community, and online payment to provide enterprises with stable, out-of-the-box, and mature Content Moderation solutions. Content Moderation helps enterprises and developers quickly detect various risks in text, images, audios, and videos in complex and ever-changing Internet environments. This can help you improve quality of platform content and user experience.
Feature description
The algorithm models for content moderation has been trained on a large number of samples and evaluated in a variety of ways. Routine iterations and upgrades are also performed continuously. The Content Moderation service is available out-of-the-box, enabling enterprises and developers to quickly roll out automatic content moderation. In addition, systematic methods and processes are provided for you to customize algorithm models based on business requirements, modify moderation thresholds, and perform other optimizations such as optimizations for specific moderation objects.
Currently, the Content Moderation service provides you with the content moderation APIs and the console to conduct tasks related to content moderation.
Content moderation APIs
You can use the content moderation APIs to call the API operations provided by Alibaba Cloud to moderate text, images, videos, and audios for pornography, politically sensitive content, violence, horror content, ad violations, spam content, and undesirable content. The Content Moderation service is applicable to online services whose content is accessible to the public such as video websites, live streaming platforms, social platforms, media platforms, vertical communities, forums, e-commerce websites, storage platforms, content delivery network (CDN) platforms, and AI-generated content (AIGC) products.
Console
You can use the console to perform analysis and management such as viewing statistics, analyzing moderation results, configuring custom moderation, and managing risk libraries. The console is suitable for you if you want to perform analysis, have different moderation requirements in various scenarios, or require custom management.
Relationship between Content Moderation 2.0 and Content Moderation 1.0
Content Moderation 2.0 is a new content moderation service. Compared with Content Moderation 1.0, it offers preset common moderation services, more labels, and more flexible configurations, and is more cost-effective.
Capability comparison between Content Moderation 2.0 and Content Moderation 1.0
Item | Content Moderation 2.0 | Content Moderation 1.0 |
Billing method and pricing |
Note For more information about the pricing of Content Moderation 2.0, see Pricing of Content Moderation. |
Note For more information about pricing of Content Moderation 1.0, see Pricing of Content Moderation. |
Moderation configurations | Moderation scope (over 10 categories) Custom libraries | Moderation scope (over 5 categories) Custom libraries |
Default capacity |
|
|
Content to be moderated and moderation services | Content to be moderated: images, text, voices, and videos Preset moderation services: common baseline moderation, social entertainment and live streaming moderation, and voice moderation | Content to be moderated: images, text, and videos Preset moderation services: common moderation service |
Moderation results | Over 50 interpretable labels (Multiple labels that indicate violations can be returned at a time.) Scores of confidence levels | Over 40 interpretable labels (Only one label that indicates a violation can be returned at a time.) Handling suggestions |
Activation and billing
For more information about activation and billing of Content Moderation 2.0, see Activation and billing.
For more information about activation and billing of Content Moderation 1.0, see Activation and billing.
Item | Content Moderation 2.0 | Content Moderation 1.0 |
Pay-as-you-go | After you activate the pay-as-you-go billing method for Content Moderation 2.0, you are charged based on the types (images, text, and voices) and quantity of content that is moderated. In scenarios where multiple risks are detected for the same content, the moderation cost decreases by 50% to 70% compared with that in Content Moderation 1.0. | After you activate the pay-as-you-go billing method for Content Moderation 1.0, you are charged based on the types of moderated content, moderation scenarios, number of requests per day, moderation results, and regions where content is moderated, such as the China (Shanghai) region. The types of moderated content include images, text, and videos. The moderation scenarios include but not limited to pornography detection, QR code detection, and text anti-spam. The moderation results include review, block, and pass. |