Amazon Web Services Inc. is bringing artificial intelligence into live video streams within the AWS Elemental infrastructure service to help broadcasters meet viewers’ expectations of content-aware streaming and hyper-personalization.

Announced ahead of this week’s IBC 2026 conference in Amsterdam, an event that showcases the latest advances in media, broadcasting and entertainment technology, the updates span server-side dynamic multiview, real-time scene analysis, low-latency ad insertion and automated ad yield optimization.

Grand View Research says the live streaming industry has grown to a more than $100 billion global market, but most streaming platforms still struggle to meet viewer’s expectations. These days, people expect a lot more than just a simple stream – they want customizable camera angles for sports events, real-time personalization and other features. They even want to see more relevant ads, because if they’re going to be forced to watch them, most would prefer that it’s something they’re more interested in.

However, the complexities involved means that few broadcasters have been able to deliver all of this. “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,” said Kali Ridley, senior technical industry product marketing manager for media and entertainment at AWS.

This is why AWS is trying to solve all of the challenges in one fell swoop, embedding machine learning and dynamic stream compositing natively into AWS Elemental’s core video pipeline. It eliminates the need for multiple complex third-party integrations, so broadcasters can deliver personalized, interactive broadcasts without mushrooming overheads.

With AWS Elemental MediaPackage Dynamic Multiview, broadcasters will be able to let their viewers select their own camera angle at all times from multiple feeds of the same event. It works by stitching multiple feed into a single compressed domain from a single set of encoded sources generated by AWS Elemental MediaLive, so there’s no need to re-encode each one. According to Ridley, this approach reduces layout infrastructure costs by up to 97%, streaming to almost any consumer device over standard HTTP Live Streaming or Dynamic Adaptive Streaming over HTTP.

One of the first broadcasters to take advantage of this is the Indian livestreaming giant JioHotstar, which is similar to Netflix but also offers live cricket, football, tennis and other sports.

“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,” said JioHotstar Chief Technology Officer Akash Saxena. “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.”

For editing and ad workflows, AWS is introducing AWS Elemental Inference Contextual Metadata, where computer vision models run in parallel with video encoding to extract real-time, scene-level metadata from live video feeds. This data can then be routed into downstream workloads, Ridley said, to enable instant highlight clipping, contextual ad placement and content moderation.

Advertisers should also be eager to see AWS Elemental MediaTailor Yield Optimization, which is a new capability that helps to solve content monetization problems. It works by automatically filling vacant ad slots on demand, using insights from Amazon Publisher Services.

By stitching these ads into the content before it’s delivered, it ensures that campaigns play smoothly without buffering, while bypassing ad blockers. Broadcasters can use the service to set custom price floors and brand safety filters, to prevent competitors from showing their ads in adjacent slots.

Finally, Ridley introduced AWS Elemental MediaTailor Low-Latency Ad Insertion, which enables targeted ad delivery in low-latency livestreams. That ensures broadcasters can monetize high-stakes live sports and breaking news with dynamic ad revenue.

The combination of AWS Elemental MediaTailor and Amazon CloudFront enabled the South American livestreaming platform DSPORTS to monetize its recent coverage of the FIFA World Cup. “Our team set out to monetize low-latency streaming without giving up the latency itself, which had not been done on our platform before,” said Claudio Rodrigues da Silva, head of IT at StreamCo. “Through DSPORTS’s coverage of the FIFA World Cup, we held six seconds glass-to-glass with ads running across every match.”

Ridley said the new capabilities all share the same key innovation, with intelligence being layered inside existing video pipelines rather than bolted on afterwards. Because they all run within the same AWS Elemental service, they provide compounding value for broadcasters – AWS Elemental Inference generates the metadata needed to choose the right ads, MediaPackage Dynamic Multiview enables the personalized viewing and ad experiences, and then MediaTailor is what stitches them together inside the low-latency streams.

Image: AWS