AWS Announces Agentic Conversational Video Intelligence

AWS has announced a new agentic conversational video intelligence solution built on Amazon Bedrock, Amazon Rekognition, and Amazon Transcribe. This solution enables users to ask natural language questions about uploaded videos and receive answers within seconds. The technology is designed to address the challenge of reviewing large volumes of video content manually, which is common in industries such as media, security, insurance, and professional services.

The solution uses an agentic architecture that dynamically decides which AWS services to invoke based on the user's query. For previously analyzed content, responses are returned in under a second. Initial analysis of new videos takes 5–10 minutes depending on the length and services required.

A major media and entertainment company has already adopted this approach during an AWS Professional Services engagement. With this solution, their consultants can query recorded discovery session content, extracting design decisions, action items, and stakeholder positions. This has resulted in a reduction in manual review time of approximately 80 percent across a backlog of more than 200 multi-hour recordings.

Source: aws-ml