Condé Nast cuts video search time 99%

Condé Nast partnered with AWS to build an AI-powered multimodal video discovery tool, reducing content search time from 250 minutes to under 2 minutes per task. The solution uses the TwelveLabs Marengo embedding model on Amazon Bedrock and Amazon OpenSearch Service to run semantic search across transcripts, visual elements, and audio within a library of over 140,000 videos. Editorial teams across Vogue, GQ, Vanity Fair, and Wired can now search by natural language intent, upload reference images, and receive results with precise timestamps rather than scrubbing through footage manually.