{
  "id": 17776689,
  "title": "Chamet deploys multimodal AI moderation for live video streams",
  "url": "https://urgent.news/2026/11/02/chamet-deploys-multimodal-ai-moderation-for-live-video-streams",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-02T08:44:24.000Z",
  "source": {
    "name": "Vietnam Investment Review",
    "slug": "vietnam-investment-review",
    "url": "https://vir.com.vn/chamet-deploys-multimodal-ai-moderation-for-live-video-streams-164151.html"
  },
  "original_language": "en",
  "account": "Singapore-based Chamet, a live streaming and video chat platform, is enhancing its safety measures through multimodal AI technology. Operating in Southeast Asia, a key market for the platform, Chamet has implemented a sophisticated moderation system designed to protect users during live video interactions. With over 50 million downloads on Google Play, Chamet faces the challenge of ensuring a safe environment while fostering connections among its adult users (18 and over).\n\nChamet's new AI system goes beyond traditional moderation by integrating multiple data points and signals. It analyzes profile photos, live-stream frames, video, voice, text, self-reported age, and account behavior. Each case is assigned a risk level based on model confidence, and the platform has the capability to restrict content, limit interactions, or suspend accounts as needed. Human reviewers perform a final check on flagged cases to ensure accuracy. In instances where users falsify their age, accounts are permanently suspended.\n\nThe platform employs a multi-model fusion approach, combining outputs from various moderation models and third-party services. These models generate labels and confidence scores for different content types and placements. Chamet maps these outputs onto a unified risk scale, considering factors such as content type, location, and the model's historical performance. This comprehensive evaluation ensures that the most critical cases receive immediate attention and consistent action.\n\nIn addition to AI-powered moderation, Chamet incorporates real-time AI translation in more than 50 languages, providing live subtitles during video calls. This feature facilitates communication across language barriers while maintaining a safe environment. The platform also employs safeguards against unauthorized screen recording, in-app reporting, blocking features, and an appeals process, working in tandem with automated systems to ensure prompt and effective safety measures. David Qi, Chamet's Chief Marketing Officer, emphasizes the importance of swift action in addressing potential harm during live video streams, noting that the combination of AI and human review enables the platform to identify risks more quickly and make consistent decisions.",
  "summary": "Social platform Chamet integrated multimodal artificial intelligence risk scoring systems to scan live video content continuously enforcing age restrictions as Singapore enacts its Online Safety Code",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}