WangYiYiDun

Overview

Netease Yidun is a digital content risk control brand under NetEase Zhihui (Netease Enterprise). Leveraging over 20 years of NetEase security expertise, it provides full-chain risk control services covering content security, business security, and application security. As a leader in content risk control in the LLM era, Yidun uses cutting-edge AI technology to help enterprises build intelligent risk control systems, reduce compliance costs, and ensure healthy business growth.

On the technical front, Yidun has developed its own deep learning engine that supports multimodal content moderation for images, text, audio, and video, with over 99% recognition accuracy. It also offers business security products such as behavioral CAPTCHA, marketing anti-fraud, and application hardening, effectively defending against fake registrations, click fraud, and malicious attacks. Its AIGC content risk control capabilities perform compliance checks on content generated by large models, helping enterprises get algorithm registration approved.

Yidun has served more than 400,000 enterprise customers across industries including social media, e-commerce, gaming, finance, and education, processing over 10 billion reviews daily. Whether for startups or large corporations, flexible API integration and visual policy configuration enable rapid deployment of risk control capabilities, ensuring all-round protection from content publishing to user behavior.

Core capabilities

Image moderation

Precisely identifies pornographic, violent, politically sensitive images, with custom blacklist/whitelist support.

Text moderation

Filters political, abusive, prohibited, advertising and other harmful content, supports multiple languages.

Audio moderation

Real-time detection of violations in live streams/voice, with emotion recognition support.

Video moderation

Frame-by-frame analysis of video and audio, covering long and short video scenarios.

Behavior captcha

Minimalist slide/click verification, effectively distinguishes humans and bots, with 99.9% interception rate.

Marketing anti-fraud

Identifies fake registrations, coupon abuse, fake orders and other black-market behaviors, reducing operational losses.

App hardening

Protects against reverse engineering, tampering, and debugging, securing mobile applications.

AIGC risk control

Detects compliance of large model-generated content, assisting algorithm filing and content ecosystem governance.

Use cases

Real-time review of social platform posts, filtering political and pornographic content.

Detection of violating product images on e-commerce platforms to prevent banned items from being listed.

Abuse and porn detection in game chat systems to maintain a healthy community.

Audio content monitoring on live streaming platforms to promptly identify violating speech.

Document security scanning in enterprise office systems to prevent confidential leaks.

Compliance detection for AIGC-generated content to ensure output meets regulations.

Behavior verification during financial app registration to block bot registrations.

Anti-fake-order for marketing campaigns, ensuring fairness and ROI.