TwelveLabs Transforms UNICEF Korea’s 8TB Media Archive with Video Intelligence, Reducing Search Time by 95%
Natural language search now surfaces any moment across 8TB of media — in seconds TwelveLabs, the video understanding company, announced it has built an AI-powered digital archive for UNICEF Korea, transforming more than 8 terabytes of fragmented media into an instantly searchable asset library. The system reduces media search time by approximately 95%, turning previously […]

TwelveLabs, a leading video understanding company, has recently announced the development of an AI-powered digital archive for UNICEF Korea. This innovative system transforms more than 8 terabytes of fragmented media into an instantly searchable asset library, significantly improving the efficiency of accessing valuable content. The project, which has reduced media search time by approximately 95%, turns previously inaccessible materials into a real-time resource for campaigns and storytelling.
UNICEF Korea's media archive spans decades of fundraising campaigns and child rights programs, encompassing tens of thousands of hours of video and millions of images. Despite its immense value, the archive had remained largely unusable due to its fragmented nature, scattered across individual PCs and network storage systems with no unified way to search or retrieve content. This made it challenging for staff to locate specific moments or materials when needed, hindering their ability to respond swiftly to campaign requirements or produce impactful content.
To address this challenge, TwelveLabs employed its Video-Native AI technology, which goes beyond simple metadata or frame-based analysis. This advanced system captures relationships between visuals, audio, actions, and context at the scene level, enabling a deeper understanding of the content. By leveraging this technology, the organization has created a unified, searchable platform that allows staff to access the archive using natural language queries. Examples of such queries include "children collecting water at a field site in Africa" or "end-of-year fundraising campaign clips." The results are instant, accompanied by precise timestamps, making it easy to locate and utilize the desired footage.
As part of the deployment, UNICEF Korea's existing NAS data was migrated to AWS S3, establishing a scalable and structured foundation for indexing and retrieval. This move not only ensures the system's efficiency but also allows for future growth and expansion, accommodating the organization's evolving needs.
The impact of this transformation is profound. With the ability to access information instantly, decision-making processes across campaign strategy and resource allocation have been significantly sharpened. Miguel Lopes, CPO at TrafficGuard, emphasized that "when organizations can access that information instantly, it sharpens decision-making across everything from campaign strategy to resource allocation."
The introduction of this AI-powered archive not only revolutionizes how UNICEF Korea manages its media assets but also serves as a model for other organizations facing similar challenges. By harnessing the power of video intelligence, the organization can now capitalize on its extensive archive to deliver more effective campaigns and raise awareness about critical child rights issues.
In conclusion, TwelveLabs' collaboration with UNICEF Korea has transformed a once inaccessible 8TB media archive into a powerful, instantly searchable tool. By leveraging advanced AI technology and a scalable infrastructure, the organization has unlocked the potential of its vast collection, enabling faster and more efficient content production and campaign execution. This innovative solution not only improves operational efficiency but also underscores the transformative potential of video intelligence in the modern world.










