Improved AI Governance: TikTok Bans 2.6 Million Accounts in Q1
The social media giant has increased its security efforts, banning illegal accounts.
Owners of TikTok, Douyin recently released its Q1 2025 Report on Combating Illegal Activities, showcasing significant progress in tackling illicit operations. The report reveals that Douyin banned 2.6 million accounts involved in activities such as online water army operations, fraud, and illegal traffic redirection in the first quarter, and reported suspected criminal activities to relevant authorities.
Improved Security
By building an AI-powered governance system, Douyin significantly improved efficiency in patrol, early warning, and analysis of illegal activities. In combating online water armies, Douyin leveraged AI to build intelligent robotic tools, automating risk detection, early warning, patrol, analysis, and post-operation review. These actions have significantly improved the platform's efficiency in identifying and handling "volume boosting," achieving second-level processing time for individual cases, an overall accuracy rate exceeding 85%, a 10.25-fold increase in patrol efficiency for online water army service violations, daily interception of 60 million illegal requests, and the banning of over 200,000 water army accounts.
Douyin is increasing its fraud prevention moves by continuously strengthening anti-fraud alerts, upgrading its "Verification Assistant" tool to help users verify if information originates from official Douyin channels and to view their contact history with official customer service. The platform also improved security models across various scenarios, focusing on addressing fraud in niche chat environments and increasing user alerts for high-risk scenarios such as "pig butchering" scams and illicit adult content redirection.
Safety Enhanced by AI Technology
From the beginning of 2025, Douyin has banned 1.4 million fraud-related accounts, sent over 800,000 anti-fraud SMS alerts daily, and made nearly 170,000 anti-fraud warning calls.
One of the ways Douyin addresses illegal traffic redirection is by establishing a multi-account link identification system to sever connections between illegal accounts across preemptive, in-process, and post-event stages, and uniformly processes violating accounts. Through the extensive application of AI technology, the platform saw a 73.3% decrease in related violation reports in the first quarter, banned nearly 1 million accounts involved in illegal traffic redirection, processed 7.45 million pieces of illegal video content, and reported leads to relevant authorities to assist in cracking down on illegal traffic redirection groups.
The company further stated that it will continue to leverage AI technology to enhance its governance capabilities, protect user safety, and maintain a healthy platform ecosystem.
