AWS for M&E Blog
Personalize live viewing, monetize every impression, and reach every device with AWS Elemental at IBC 2026
Live video audiences now expect content-aware streams, per-viewer personalization, and monetization opportunities that respond to what’s happening on screen. But because each has historically required its own vendor, its own pipeline, and its own operational overhead, competing priorities force teams to choose which to pursue first rather than delivering all three at once.
Today, we’re announcing four new capabilities across AWS Elemental that bring AI, personalization, and intelligent monetization to the live video workflow, launching in time for IBC 2026 in Amsterdam.
Here’s what’s new:
- AWS Elemental MediaPackage Dynamic Multiview – Viewers choose their own multi-camera layout from a single set of encoded sources, delivered to any device over standard HTTP Live Streaming (HLS) or Dynamic Adaptive Streaming over HTTP (DASH), without per-layout encoding costs
- AWS Elemental Inference Contextual Metadata – AI-generated metadata extracted from live video in real time, powering personalized video experiences and contextual ad decisioning
- AWS Elemental MediaTailor Yield Optimization – Automated backfill of unfilled ad inventory using Amazon Publisher Services (APS) demand, giving streaming TV publishers incremental revenue with zero setup cost
- AWS Elemental MediaTailor Low-Latency Ad Insertion – Extends personalized ad insertion of MediaTailor to low-latency streams using Apple’s HLS interstitials, so publishers no longer have to choose between fast delivery and ad revenue
Server-side multiview, stitched on demand with MediaPackage Dynamic Multiview
Multi-feed viewing has become a necessity across live content: multiple camera angles for a single sporting event, simultaneous games on one screen, or curated linear bundles combining different content types like news, sports, or entertainment into a single view. But existing approaches force trade-offs. Tiled encoding limits reach to High Efficiency Video Coding (HEVC)-capable devices. Client-side compositing demands custom SDKs per device. Server-side compositing requires pre-encoding every layout combination, each costing as much as a new channel. Large operators using server-side compositing vendors spend $2M–$12M per year on infrastructure alone, and most streaming platforms can’t justify the investment at any price.
MediaPackage Dynamic Multiview stitches multi-camera and multi-feed layouts on demand from encoded sources, delivered as a single adaptive bitrate stream through HLS or DASH. Sources encode one time through AWS Elemental MediaLive. AWS Elemental MediaPackage assembles layouts in the compressed domain with no re-encoding. The output reaches more than 95% of consumer devices with a single digital rights management (DRM) key and requires no SDK, per-device builds, or new vendors.
“AWS Elemental MediaPackage Multiview lets us deliver multiple synchronized camera angles within a single stream, putting our subscribers in control of how they experience live sports. The server-side composition approach scales this premium experience to over 100 million viewers without multiplying origin complexity or CDN costs, giving us a true differentiator in India’s competitive OTT landscape.” — Akash Saxena, CTO (JioHotstar)
Figure 1: Multi-feed live experiences with MediaPackage Dynamic Multiview
By compositing in the compressed domain rather than re-encoding per layout, MediaPackage Dynamic Multiview provides up to 97% cost savings compared to solutions that require a separate video stream for every viewing combination.
Understanding content in real time with Elemental Inference Contextual Metadata
Personalizing video experiences requires understanding what’s happening on screen. Whether that means surfacing the right highlight clip, recommending related content, or matching an ad creative to the moment, it starts with metadata. Until now, generating that understanding at broadcast scale required either manual tagging or bespoke machine learning (ML) infrastructure.
Elemental Inference Contextual Metadata uses AI to generate rich, scene-level metadata extracted from live video in real time. Computer vision models run in parallel with encoding to identify objects, actions, brands, on-screen text, and sentiment from real video content. The metadata is emitted through a standard API, aligned to the Interactive Advertising Bureau (IAB) content taxonomy and the Global Alliance for Responsible Media (GARM) brand suitability categories. As displayed in the following flow diagram, this powers downstream workflows including automated highlight clipping, content discovery, contextual ad decisioning through Elemental MediaTailor, and content moderation, all without custom ML infrastructure.
Figure 2: Contextual metadata flow from encoding to ad decisioning
Maximizing revenue with Elemental MediaTailor Yield Optimization
AWS Elemental MediaTailor Yield Optimization fills available ad slots from Amazon Ads inventory during ad insertion, turning dead air and slate video into monetized impressions. It delivers incremental revenue with no cost to enable and no additional infrastructure. Because the ads are inserted server side (stitched directly into the stream before it reaches the viewer), they play seamlessly on every device, can’t be blocked, and transition without buffering. At launch, MediaTailor Yield Optimization is available to APS streaming TV publishers.
The capability is built for large-scale live events like professional league championships, with low latency and fast response times. It remains brand safe by filtering demand by category to avoid competitive conflicts within the same break. For example, if a sports drink brand is running a direct-sold spot, MediaTailor Yield Optimization won’t fill an adjacent slot with a competing beverage advertiser. Publishers set their own price floors and category rules throughout.
Monetizing low-latency streams with MediaTailor Low-Latency Ad Insertion
AWS Elemental MediaTailor already delivers personalized, server-side ad insertion at scale. MediaTailor Low-Latency Ad Insertion now extends that capability to low-latency streams, using HLS interstitials, a server-guided ad insertion (SGAI) approach. As audiences increasingly expect streams to keep pace with social (a goal scored, a breaking headline, a viral moment), publishers can no longer afford latency that lets the timeline spoil the moment before their stream delivers it.
“Our team set out to monetize low-latency streaming without giving up the latency itself, which had not been done on our platform before. Through DSPORTS’s coverage of the FIFA World Cup 2026, we held six seconds glass-to-glass with ads running across every match, built on AWS Elemental MediaTailor and Amazon CloudFront.” — Claudio Rodrigues da Silva, Head of IT (Streamco)
How these capabilities connect across workflows
These four capabilities share a design philosophy: intelligence layered inside the video pipeline, not bolted on after. They all run within the same AWS Elemental services that power millions of live streams globally, with the same 99.9% service level agreements (SLAs), the same operational model, and no new vendors to onboard.
The value compounds across the stack. AWS Elemental Inference generates metadata that enriches yield decisions, MediaPackage Dynamic Multiview delivers personalized viewing and ad experiences at scale, and MediaTailor low-latency insertion help to avoid impressions being left behind regardless of how fast the stream moves.
Visit us at IBC 2026 in Amsterdam (September 12–15) to see live demonstrations. Get in touch with an AWS for Media & Entertainment expert to learn more.
Get started: MediaPackage Dynamic Multiview and Elemental Inference Contextual Metadata are available today on the AWS Elemental MediaPackage console and AWS Elemental MediaLive console. MediaTailor Yield Optimization and Low-Latency Ad Insertion are available on the AWS Elemental MediaTailor console.
Have questions or building something with these capabilities? Let us know in the comments.