Coming out of NAB, I have a slightly different read than the usual post-show summaries.
Yes, the expected conversations were all there: AI, cloud, software-defined workflows, next-generation fan experiences. But if you actually sat down with the people building and running broadcast infrastructure, a different picture emerged.
This industry is still in the middle of a much more fundamental transition. Broadcast is just now fully moving to IP. That may sound obvious, but it’s important context. While other industries debate AI maturity curves, broadcast engineers are still grappling with the realities of SMPTE ST 2110, multicast at scale, and what happens when a packet is lost in a live production environment.
IN THIS ARTICLE
The conversations themselves were different this year. In multiple conversations, unprompted, we heard variations of the same message: “You guys are doing a great job,” and “The difference is that NETGEAR cares.” That second one came from a large, well-known manufacturer, said in direct comparison to the usual enterprise networking players.
What sits behind this comment is frustration with how this market has been served historically. Broadcast has often been treated as an adjacency to data center networking: important, but not core. And when priorities shift at those vendors, as they inevitably do, so does attention.
We also heard very practical concerns. Supply constraints from some vendors are becoming a real issue. And when your entire revenue depends on infrastructure you can’t get, it stops being an inconvenience and starts being a business problem.
So when we show up with a clear focus on AV and broadcast, a full warehouse, and a firm roadmap, it resonates because we are better aligned with their reality.
One of the sharpest lines of the week came from Helen Matthews at Futuresource, during a Grass Valley session: “The cost of adding things to increase viewer experience isn’t usually the technology, it’s the added headcount.”
That explains a lot of what we are seeing. AI is the shiny object, but operational efficiency is the daily reality: doing more with less. We heard it twenty times at NAB: more feeds, more formats, more personalization, more real-time engagement. The tools are increasingly there, whether in software-defined workflows, remote production, or AI-driven enhancements.
But since people can’t scale the way the workflow does, the burden shifts to the infrastructure. The network, in particular, has to do more than move packets. It has to simplify operations, provide visibility, and reduce the friction of running increasingly complex environments.
In SDI environments, things were point-to-point, single-use, and relatively easy to troubleshoot. In IP, that’s gone. Now you’re dealing with high-bandwidth, uncompressed video streams, tight synchronization requirements, multicast at scale, distributed production models, and hybrid cloud workflows.
That is why topics like observability, system health, and monitoring kept coming up. Very practically: ‘how do I know what’s happening in my network in real time, and how do I fix it NOW when it isn’t working?’ Good to see the progress here with vendors like Providius and NEXOG.
This is also why there’s growing interest in more programmable, software-defined approaches. The network is becoming part of the workflow.
The appetite to deepen partnerships was clear. We had technical, detailed conversations, and in many cases outlined concrete homework on integration, training, and commercial alignment.
Concepts that extend beyond traditional switching are also getting attention. The idea of running third-party applications closer to the network, on-premises, instead of in the cloud, resonated more than expected. There’s genuine interest in experimenting, joining beta programs, and understanding how this could fit into existing environments.
That ties back to the earlier point. If the network is becoming part of the workflow, it also becomes a place for additional functionality to live.
Services matter more than before, too. Support and professional services came up in almost every meeting, as a key part of the value proposition. Customers aren’t just buying hardware. They’re buying confidence that it will work as expected in a high-stakes environment.
And bandwidth is only going one way. We had multiple asks around 400 gig, in the same month where we shipped out the first full 100 gig broadcast switches. The bandwidth conversations happened in the context of planning and roadmap discussions. Between higher resolutions, more feeds, and distributed production, the demand is building quickly.
If you spend time in Silicon Valley or at events like MWC or CES, you would think AI is the only topic. At NAB, it was present, but uneven. There were areas where it was front and center: analytics, personalization, content optimization. Sports, news, and weather report production in particular are moving quickly on this. But in the core infrastructure conversations, much less so.
That reflects where the industry actually is. Before you can fully leverage AI, you need a solid, scalable, and reliable foundation. And for many broadcasters, that foundation is still being built.
Get the plumbing right first.