MTikGuard System: A Transformer-Based Multimodal System for Child-Safe Content Moderation on TikTok

arXiv — cs.CLTuesday, November 25, 2025 at 5:00:00 AM
  • The MTikGuard system has been introduced as a transformer-based multimodal content moderation tool for TikTok, aimed at detecting harmful content in real-time. This system leverages an expanded TikHarm dataset and a multimodal classification framework that integrates visual, audio, and textual features, achieving high accuracy in content moderation.
  • This development is significant for TikTok as it addresses the growing concerns over harmful content affecting children and teenagers on the platform. By implementing MTikGuard, TikTok aims to enhance user safety and improve the overall experience for its younger audience.
  • The introduction of MTikGuard coincides with TikTok's broader strategy to empower users with more control over their content consumption, including features that allow users to manage the amount of AI-generated content in their feeds. This reflects a growing trend in social media platforms to prioritize user agency and safety amid rising scrutiny over content moderation practices.
— via World Pulse Now AI Editorial System

Was this article worth reading? Share it

Recommended apps based on your readingExplore all apps
Continue Readings
Tech Left Behind in S&P 500’s Latest Rebound | Bloomberg Tech 12/3/2025
NeutralArtificial Intelligence
The latest analysis from Bloomberg highlights a significant lag in the performance of major tech stocks, particularly the so-called Magnificent 7, amidst a broader rebound in the S&P 500. This trend raises concerns about the sustainability of the tech sector's growth, especially as companies like Okta report mixed earnings and adjust profit forecasts. Additionally, TikTok has announced plans to invest over $37 billion in a new data center in Brazil, marking its first major project in Latin America.
Brazil Emerges as AI Hot Spot With TikTok’s Data Center Push
PositiveArtificial Intelligence
ByteDance's TikTok has announced a significant investment of 200 billion reais (approximately $38 billion) to establish a large data center complex in Brazil, reinforcing the country's status as a key player in AI investment in the region.
TikTok plans to invest $37.7B+ to build a data center in Brazil's port of Pecém, partnering with Omnia and Casa dos Ventos, its first Latin American project (Bloomberg)
PositiveArtificial Intelligence
TikTok, the popular video-sharing platform owned by ByteDance Ltd., has announced plans to invest over $37.7 billion to construct a data center in Brazil's port of Pecém. This initiative marks TikTok's first major project in Latin America, in collaboration with local partners Omnia and Casa dos Ventos.
TikTok to Invest More Than $37 Billion in Brazil Data Center
PositiveArtificial Intelligence
TikTok, owned by ByteDance Ltd., has announced a significant investment of over 200 billion reais (approximately $37.7 billion) to construct a data center in Brazil, marking its inaugural project in Latin America. This facility will be located in the port of Pecém and is expected to enhance TikTok's operational capabilities in the region.
AI Forensics: 354 AI-focused TikTok accounts pushed 43K posts made with GenAI tools that hit 4.5B views, including posts with anti-immigrant and sexual material (Dan Milmo/The Guardian)
NegativeArtificial Intelligence
Researchers identified 354 TikTok accounts focused on artificial intelligence that collectively generated 43,000 posts using generative AI tools, amassing 4.5 billion views in a month. These posts included content deemed harmful, such as anti-immigrant and sexual material, raising significant concerns about the platform's content moderation practices.
Irish media commission to probe LinkedIn and TikTok over potential DSA breach
NeutralArtificial Intelligence
The Irish media commission has initiated investigations into LinkedIn and TikTok to determine if they have violated the Digital Services Act (DSA) by failing to facilitate the reporting of illegal content effectively. This inquiry is part of a broader effort to ensure that social media platforms adhere to regulations designed to create safer online environments.
Anti-immigrant material among AI-generated content getting billions of views on TikTok
NegativeArtificial Intelligence
Researchers have identified 354 TikTok accounts that generated 43,000 posts using AI tools, collectively amassing 4.5 billion views in just one month. This content includes harmful material, such as anti-immigrant and sexualized themes, raising significant concerns about the impact of AI-generated content on social media platforms.
Ireland is investigating TikTok and LinkedIn for possible DSA violations
NeutralArtificial Intelligence
Ireland is currently investigating TikTok and LinkedIn for potential violations of the Digital Services Act (DSA), which aims to regulate online platforms and ensure user safety. This inquiry reflects ongoing scrutiny of social media companies regarding their compliance with European regulations.