Where AI Succeeds and Fails in Live Streaming Monitoring Today
The benefits of leveraging AI in live stream monitoring seem self-evident, especially at scale, given the sheer amount of information required to track all aspects of a live stream from ingest to delivery and the speed at which that data arrives. Human eyes on glass can only absorb so much. But where is AI working and not working in 2026 when it comes to live stream observability and the monitoring tasks streaming teams are currently off-loading to it? AWS’s Steph Lone, TAG’s Michael Demb, and Streaming Media’s Nadine Krefetz discuss in this clip from Streaming Media Connect 2026.
AI, Predictive ABR, Anomaly Detection, and Live Monitoring
Krefetz kicks off the discussion by giving examples of some of the prioritizing that goes into traditional eyes-on-glass live stream monitoring, with only so many eyes and so much glass to go around. "When you guys are talking about the different aspects of what you're monitoring, what needs to be delivered, what needs to be watched versus what not to watch, will AI take some of the burden off of existing systems? Where does AI work, and where it does not work?"
Lone offers an analogy from an adjacent area of live streaming tech and workflows that can benefit directly from AI-based monitoring. "Think about AI like content-aware encoding for live, both from a budgetary standpoint as well as just being able to save costs on that bandwidth reduction. AI might adjust that bitrate per scene on the encoding side in real time. When you have more predictive ABR," she explains, AI can help you " see low-bandwidth users and auto-degrade the bandwidth for those users. Or real-time, quality control anomaly detection. AI is used already in a lot of different places to detect things like artifacts or freezing events, lip sync errors, and allowing those alarms to be surfaced. I do believe that it has a place and is being used quite a bit here from an automation standpoint."
She goes on to share "a couple other thoughts on where AI is also being helpful, especially around things like mean time to resolution. We have a use case with F1 where they've built a root cause analysis assistant, and that allows them, very quickly in real time with their F1 premium TV product, to figure out what the problem is and reduce the time to fix considerably using AI."
Simplifying Complexity
TAG's Demb chimes in on the question of where AI excels, and starts with the overall benefit of bringing AI into the monitoring mix: "AI is good at simplifying things for humans. It can take a complex problem and compress it down for us to easily understand," he says.
"One of the key things that I always talk to my customers about in monitoring is not everything that you see is what's actually happening." For example, he says, "You might see some microblocks. You might not even need a tool to detect microblocks because there's always an underlying problem that is causing this. So you need to start thinking backwards and understand, 'Why do I have microblocks? Maybe I have a packet loss on the incoming feed to my encoder.' It might be the encoder itself, or the buffers. There are tons of different reasons."
The difficulty of making a definitive root cause analysis, he goes on to explain, is one reason AI can be so helpful. "Because there are tons of different reasons for every symptom and every problem we see, that's already a complex problem to solve for humans." Tipping his hat to Lone, he says, what AI offers in these scenarios is "reducing time to repair (MTTR). We try to help customers, so we are looking for ways to use AI to simplify this complexity," and make it easier for them to process "all of those different metrics and events and alarms that they're getting in real time. Especially when you're producing a big show and the time is critical, you want to know what's going on right now," as well as gathering information that will "point you towards [solving] the problem in the future."
Looking ahead, he concludes, "Not too far in the future, we're going to be able to use AI to solve those problems for us. We need to start trusting AI a little bit more before we do that. But things like self-healing and solving the problems for humans are definitely not far beyond the horizon for us."
Join us November 9–11, 2026 for more thought leadership, actionable insights, and lively debate at Streaming Media Connect 2026! Registration is open!
Related Articles
In this Q&A, I had the opportunity to discuss the fast-evolving requirements and best practices for live streaming QC at scale with Anupama Anantharaman, VP Product Management at Interra Systems, developers of the ORION-OTT comprehensive monitoring solution for live and OTT. The 2025 Streaming Media All-Star offers insights into how AI is changing live stream monitoring, how performing "deep QC" impacts streaming workflows, the best KPIs for predictive monitoring, and the continuing role of humans in an increasingly AI-supported QC world.
26 Jun 2026
From an observability standpoint and eyes on glass, what are the key metrics to watch to determine if ad insertion is going smoothly during a live stream? Altitude TV's Dave Zur, Sargeway LLC's Sarge Sargent, Qualabs' David Hassoun, and FanServ's C.J. Leonard discuss challenges and best practices for monitoring live streaming ad performance in this clip from Streaming Media Connect 2026.
09 Apr 2026
As with all streaming workflows, AI has steadily crept into the live streaming technology stack. In some cases, the impact is incremental, in others, profound. From production to monetization, here's a quick overview of where AI has become relevant for live event producers and engineers, and some areas where, surprisingly, it hasn't.
26 Mar 2026
Elecard Head of Strategic Partnerships Victoria Tuzova joins Future Frames Podcast Co-Producer Doug Daulton for a candid interview on the show floor at Streaming Media 2025, discussing livestream monitoring for sports and other large-scale events, ways to improve the sports fan experience through ultra-low-latency and personalization, and her work with Women in Streaming Media to heighten women's representation in the industry.
14 Oct 2025