Douyin's 2024 Crackdown on Online Malicious Activities: AI-Powered Protection for User Safety

Douyin's 2024 Crackdown on Online Malicious Activities: AI-Powered Protection for User SafetyIn 2024, Douyin significantly intensified its efforts to combat online malicious activities. By upgrading its AI large language model (LLM) technology, iterating risk control strategies, and optimizing its reporting mechanisms, the platform effectively enhanced its ability to identify and manage these activities

Douyin's 2024 Crackdown on Online Malicious Activities: AI-Powered Protection for User Safety

In 2024, Douyin significantly intensified its efforts to combat online malicious activities. By upgrading its AI large language model (LLM) technology, iterating risk control strategies, and optimizing its reporting mechanisms, the platform effectively enhanced its ability to identify and manage these activities. This "2024 Crackdown on Online Malicious Activities" report details the platform's specific achievements in combating online fraud, online water armies (bots), and illegal outbound links, demonstrating Douyin's commitment and actions in protecting user safety and fostering a healthy online environment.

I. Precisely Targeting Online Fraud to Protect User Assets

In 2024, fraudulent activities impersonating leaders, relatives, and official customer service representatives became a key focus. Malicious actors used various methods, such as phone calls, text messages, and instant messaging applications, to broadly contact potential victims. They typically induced victims to share their screens or download third-party applications to obtain bank account information, guiding users step-by-step to transfer funds and perpetrate fraud.

Brush-order scams and shopping scams were also key targets. Brush-order scam groups often used job offers, part-time work, or get-rich-quick schemes as bait, or disguised themselves as singles seeking companionship, to induce users to download third-party software for fraudulent activities. Another type of group used gimmicks like insider concert tickets, free My Little Pony cards, or low-priced game skin transfers to lure target users into transactions on third-party platforms, ultimately committing fraud.

To effectively counter these fraudulent activities, the Douyin platform implemented several measures:

• Multi-channel Anti-fraud Reminders: The platform intercepted malicious actors from reaching victims in multiple scenarios, including comment sections, group chats, and private messages, and issued anti-fraud reminders through various on-site and off-site channels, protecting an average of 2 million potential victims daily.

• Launch of Anti-fraud Tools: Launched the anti-fraud customer service "Verification Assistant" to help users identify and prevent fraud risks.

• Continuous Iteration of Anti-fraud Strategies: Based on the characteristics of malicious activities, the platform iterated over 200 anti-fraud strategies per month, continuously improving its anti-fraud capabilities.

These efforts yielded significant results. In 2024, based on clues provided by Douyin, the police cracked down on 69 fraud gangs, arresting 451 suspects, effectively deterring online fraud.

II. Precisely Identifying Online Water Armies to Maintain a Healthy Platform Ecosystem

Selling water army services, malicious comment control, fake traffic generation, and malicious reporting were the four main types of water army activities Douyin targeted in 2024. Malicious actors sold illegal services such as boosting follower counts and likes, or systematically and purposefully posted comments in large quantities to mislead users; they also used cheating methods to fabricate fake traffic, hype up popularity, and even exploited the platform's reporting rules for profit, maliciously reporting others resulting in account suspensions.

To effectively combat water army activities, Douyin platform implemented the following measures:

• Building User Models: Douyin built user models for bot operators and buyers to accurately identify risky accounts and intercept abnormal operations, blocking water army's fraudulent pathways.

• Enhanced Identification Capabilities: Utilizing large model technology to enhance the identification of water army accounts, with differentiated handling of bot accounts and deceived users.

Through these measures, the Douyin platform achieved significant results in 2024: intercepting an average of 62 million illegal pieces of information daily, banning 1.1 million accounts, and reporting suspected illegal activities to relevant authorities. Additionally, based on clues provided by Douyin, the police arrested 90 suspects involved in water army activities.

III. Upgrading Technological Capabilities to Block Illegal Outbound Link Risks

In addition to fraud and water armies, illegal outbound links are another major area where malicious activities are concentrated. Malicious organizations used cheating methods to circumvent the platform's risk control strategies, inserting QR codes, contact information, images, and URLs in various channels, guiding users to click through to third-party illegal websites, which typically offer illegal services such as pornography and gambling.

To counter the new methods of malicious activities, the Douyin risk control team continuously improved technological capabilities:

• Automated Black Sample Library: Added an automated black sample library, utilizing AI technology to automatically identify and classify illegal behavior and take action.

• Upgraded Video Keyframe Model: Upgraded the video keyframe extraction model, quickly locating illegal pages embedded in videos.

• Added Live Comment Scene Model: Added a live comment scene model, effectively implementing illegal interception strategies in this scenario.

These technological upgrades effectively improved the platforms ability to identify and intercept illegal outbound links. In 2024, Douyin banned over 1.2 million related accounts, and the police, based on clues provided by Douyin, cracked down on 8 illegal traffic diversion gangs, arresting 40 suspects.

IV. Summary

In 2024, the Douyin platform achieved significant success in combating online malicious activities. By continuously upgrading its technological capabilities, iterating its risk control strategies, optimizing its reporting mechanisms, and actively cooperating with public security organs in cracking down on these activities, Douyin effectively maintained platform ecosystem safety and protected users' legitimate rights and interests. In the future, Douyin will continue to increase investment and continuously improve its anti-malicious activity capabilities to create a safer, healthier, and more positive online environment for users. The platform will continue to monitor new trends in online malicious activities, actively responding to new challenges, and contributing to the construction of a clear and healthy online space. This is not only the platform's responsibility but also a shared responsibility of society as a whole. Only with universal participation and concerted efforts can we effectively combat online malicious activities and maintain online security.

Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])