How Condé Nast built multimodal video discovery with Amazon Bedrock

Condé Nast partnered with the AWS Generative AI Innovation Center (GenAIIC) to build an AI-powered multimodal video discovery solution for its editorial teams, AWS reported. The solution runs intent-based semantic search across video transcripts, visual elements, and audio, and reduced discovery time from 250 minutes to under 2 minutes per task, according to the company.
Condé Nast had been spending an average of 250 minutes per content discovery task, manually scrubbing through a library of more than 140,000 videos and relying on titles and descriptions. AWS said the core problem was structural: existing search tools cannot look inside video content, so teams depended on institutional knowledge to locate assets, creating single points of failure when specific individuals were unavailable.
The solution is built on Amazon Bedrock and Amazon OpenSearch Service. The team selected the TwelveLabs Marengo embedding model, which AWS said natively encodes visual, audio, and transcript signals jointly. AWS said Marengo is accessed through Amazon Bedrock, while Amazon OpenSearch Service provides managed k-nearest neighbor search with multi-AZ replication and metadata filtering. The architecture separates an asynchronous ingestion and indexing plane from a synchronous query and serving plane. The solution has run in production for six months, AWS said.
Condé Nast conducted a benchmarking workshop in May 2026 to quantify impact. In addition to the 99.2 percent reduction in discovery time, the company measured over 90 percent reduction in manual video review effort and approximately $800,000 in estimated annual operational savings.
Billy Keenly, Global Senior Director, Creative Optimization at Condé Nast, said: "The AWS team, along with the GenAI and TwelveLabs attendees, helped us explain the business case clearly and concisely. I am optimistic about notching more progress towards our workflow goals in the near future."
Based on reporting from the original publisher. Visit the source for full context and later updates.
Publisher excerpt
Condé Nast's editorial teams spent an average of 250 minutes per task searching a library of more than 140,000 videos using only titles and descriptions. Working with the AWS Generative AI Innovation Center, they built a multimodal video discovery solution on Amazon Bedrock and Amazon OpenSearch Service that cut discovery time to under 2 minutes.