AWS Launches Elemental Inference AI Service for Automatic Vertical Video and Live Clip Creation

Most live broadcasts are produced in a 16:9 landscape format, while TikTok, Instagram Reels and YouTube Shorts favor 9:16 vertical short-form video. The process previously required editors to crop footage segment by segment, track subjects and select highlights manually, often missing the window for real-time distribution. Amazon Web Services (AWS) has now integrated AI into the encoding workflow, eliminating the need for media companies to build their own models or establish separate post-production teams.

AWS formally launched AWS Elemental Inference on February 24, 2026, as a fully managed, usage-based service, without disclosing fixed pricing. It can intelligently crop 16:9 video into 9:16 format while encoding live and on-demand content, as well as extract highlights from sports including soccer and basketball, with processing latency of about 6–10 seconds. The initial rollout covers four regions: US East (N. Virginia), US West (Oregon), Asia Pacific (Mumbai) and Europe (Ireland).