Does AI Change the Case for AV Standardization?

By Sam Kennedy

For years, we've told enterprise customers to standardize their AV and UC environments, and there are plenty of good reasons why. Standardization simplifies design, deployment, and procurement. It creates a more consistent user experience, and perhaps most importantly, it simplifies operations.

If your organization has hundreds or thousands of rooms, every additional manufacturer introduces another layer of knowledge your team needs to maintain: another management platform, another interface, another API, another set of commands, another troubleshooting methodology, another certification, another set of quirks your technicians need to understand.

So the advice has been pretty straightforward. Reduce the number of variables. Standardize on a smaller number of manufacturers and technologies, and your environment becomes easier and less expensive to operate.

That logic has shaped enterprise AV strategies for years. But AI may be about to challenge part of it, because what happens when the human operator no longer has to personally carry all of that manufacturer-specific knowledge?

Standardization Has Always Had an Operational Argument

There are many reasons to standardize, and I don't think they're going away. Volume purchasing matters. Spares matter. Consistent room experiences matter. Repeatable designs matter. Deployment efficiency matters.

But there's another reason organizations standardize that we don't always talk about explicitly: people can't be experts in everything.

Imagine supporting an environment with equipment from ten different manufacturers. Your team needs to understand how each works. When something breaks, someone needs to know which management platform to open, understand the terminology, know where to look, and understand how that manufacturer's products interact with everything else in the room. And when the problem crosses multiple manufacturers, the troubleshooting gets even harder.

So we reduce complexity by reducing choice. That's rational. But what if AI can absorb some of that complexity?

What If the AI Knows the Manufacturers?

This is where the idea gets interesting. Imagine an AV administrator supporting rooms containing products from multiple manufacturers. Traditionally, that diversity increases the operational burden on the team.

But now put an operational AI layer across that environment. The administrator doesn't necessarily need to remember how one manufacturer exposes a particular setting versus another. They don't necessarily need to know which interface to open first, or memorize every command or device-specific troubleshooting process.

They can start with the problem, like the room isn't showing video, and let the AI help navigate the underlying technologies.

The room may contain a display from one manufacturer, a conferencing system from another, a DSP from another, control from another, and network infrastructure from another. The operator sees a room with a problem. The AI can deal with the complexity underneath it. That's a very different operating model.

We Standardized the Hardware Because We Couldn’t Standardize the Knowledge

That's the thought I keep coming back to. Historically, one of the easiest ways to simplify operations was to simplify the technology stack. If every room uses the same products, technicians can become very good at supporting those products, and the organization's knowledge becomes repeatable.

But AI introduces another possibility. What if the common layer doesn't have to be the manufacturer? What if the common layer becomes the intelligence operating across the manufacturers?

Now, instead of forcing every room into the same technology stack so humans can reasonably support it, the AI layer can normalize some of that complexity for the humans. The devices can be different, the APIs can be different, the management systems can be different, but the administrator's experience can become more consistent. They can talk to Lena.

From Manufacturer-Centric to Outcome-Centric

This could change how organizations think about designing spaces.

Consider two rooms. One is a small huddle space used primarily for quick internal calls. The other is an executive boardroom supporting high-profile hybrid meetings. Should those rooms necessarily use the same technology? Maybe. But the reason should be because the technology is right for both spaces, not simply because that's the standard.

The same question applies to classrooms, training rooms, auditoriums, divisible spaces, and specialty collaboration environments. Different spaces have different requirements, different users have different requirements, and different manufacturers are good at different things.

Historically, organizations have had to balance those needs against operational complexity. The perfect product for a particular application might lose to the standardized product because introducing another manufacturer creates too much support overhead. AI has the potential to change that equation.

Best of Breed Becomes More Practical

This is where organizations may gain something important: choice.

If operational AI can reduce the cost of managing a heterogeneous environment, organizations can become more willing to select the technology that best solves the problem in front of them. Maybe one manufacturer is the best fit for your small rooms, another delivers something particularly valuable in your executive spaces, a third makes sense in classrooms, and a fourth solves a specialized requirement better than anyone else.

Historically, every one of those decisions adds operational complexity. With an AI orchestration layer, some of that complexity can be abstracted away from the people operating the environment. Now the question becomes less which manufacturer we've standardized on, and more what the best technology is for this space and these users. That's a healthier design question.

This Doesn’t Mean Standards Go Away

This is where I want to be very clear. I'm not arguing for an AV free-for-all. AI isn't a reason to install whatever anyone wants wherever they want it.

Organizations still need standards: architectural standards, security standards, network standards, user-experience standards, documentation standards, deployment standards, support standards. They may still have preferred manufacturers, and there will still be tremendous value in repeatable room designs.

But perhaps we need to separate two ideas that have historically been tightly connected: standardizing the experience, and standardizing every piece of technology used to create that experience. Those aren't necessarily the same thing.

Maybe the future is highly standardized at the experience and operational layer while becoming more flexible at the device layer. That's an interesting possibility.

AI Can Become the Translation Layer

Think about what happens when a technician encounters an unfamiliar manufacturer today. There's friction. Where do I manage this? What's this setting called? How does this device report status? What commands are available? What does this error mean? How does this interact with the other technology in the room?

The technician has to translate between manufacturers. AI can potentially become part of that translation layer. The administrator can express intent in a consistent way, like turning on the display, asking why a room isn't showing video, checking whether a room is ready, or asking what changed.

Underneath that request, the actual implementation may be completely different depending on the manufacturer, but that's implementation detail. Increasingly, the human may not need to care. The human defines the desired outcome, and the orchestration layer determines how to interact with the underlying technology. That's one of the reasons I believe AI orchestration will be so important in AV.

Troubleshooting Gets Even More Interesting

This becomes particularly powerful when a problem crosses manufacturer boundaries. A user reports that the display isn't working. The display is from one manufacturer, but the problem isn't actually the display. It's a connected device from another manufacturer.

As we've discussed in our troubleshooting work with Lena, symptoms don't always appear where the root cause exists. In a heterogeneous environment, that becomes even harder for humans. Now the technician needs expertise not only in the first manufacturer, but the second, and the relationship between the two.

AI doesn't have to think in terms of organizational silos or manufacturer boundaries. It can investigate the environment as a system: what's supposed to happen, what's actually happening, and where the behavior diverges from the expected outcome. That is much closer to how we need to operate increasingly complex spaces.

The Economic Argument Changes Too

This ultimately comes back to cost. Organizations don't standardize only because they like consistency. They standardize because complexity is expensive. Complexity requires training and specialized knowledge, makes troubleshooting harder, creates more platforms for administrators to understand, and increases the number of people who need to become experts in different technologies.

If AI can reduce the human cost of that complexity, then the economic argument changes. Not disappears. Changes.

Maybe choosing a second manufacturer used to create too much operational overhead. With AI assisting the operation, perhaps it doesn't. Maybe the incremental support cost becomes small enough that the benefit of choosing a better technology outweighs it. That's where this gets really interesting.

Standardize Where Standardization Creates Value

I've been thinking about this a lot recently, in part because of conversations and ideas being advanced by Byron Tarry at NEXXT around how AI changes the operating model for our industry. It has challenged me to reconsider something we've accepted for a long time: what are we actually trying to accomplish when we standardize?

If the answer is a better user experience, absolutely standardize it. If the answer is better security, standardize it. If the answer is repeatable deployment, standardize it. If the answer is easier procurement, standardization may still be incredibly valuable.

But if the answer is that our people can't possibly know how to operate all these different technologies, AI may give us another way to solve that problem. And that means organizations may be able to standardize where standardization creates value while allowing choice where choice creates value.

The New Standard May Be the Experience

Maybe that's where this is ultimately heading. Instead of saying every room must contain the same manufacturer's technology, we may increasingly say every room must deliver the same expected experience.

Users shouldn't care which API Lena called. They shouldn't care which management platform sits behind the device, or whether the display in this room came from the same manufacturer as the display three floors away. They care that when they walk into the space, it works.

The administrator wants something similar: a consistent way to understand, troubleshoot, and operate the environment regardless of which manufacturer's logo is on the equipment.

If AI can provide that common operational layer, then perhaps the future of standardization isn't less discipline. It's standardizing at a different level. Standardize the outcome. Standardize the experience. Standardize the operational expectations. And then give organizations more freedom to choose the technologies that best deliver those outcomes.

For years, we've reduced complexity by reducing choice. AI may allow us to manage complexity without sacrificing as much choice, and that could fundamentally change how we design the next generation of AV and UC spaces.

To learn more or to schedule a demo:
Visit: www.netspeek.ai
Contact: lena@netspeek.com
Demo: Book a live walkthrough