Integrators and AI

Symbolic AI Image of All-Seeing Eye
(Image credit: Getty Images)

The hype around generative AI is recent, but Pro AV has been applying smart technology and artificial intelligence for a while. What has members of this industry excited is how quickly these systems are evolving, as well as their potential to revolutionize not only the end user experience but how systems are designed and deployed as well.

[Hybrid: The Next Generation]

In the hybrid meeting environment, for example, smart cameras and audio processing technology are leveraging AI to improve conferencing. “If we talk about intelligent edge devices, pretty much there isn’t a single manufacturer that is building any kind of conference room technology where AI isn’t a central part of how the camera is framing the image and composing what we view,” noted Julian Phillips, SVP of global workplace solutions and managing director of the AVI-SPL Experience Technology Group (XTG). "Thanks to AI-driven audio processing, meetings are no longer interrupted by the sound of crinkling papers and tapping pens. With AI, now we have the ability to isolate a lot of those noises and concentrate on the real conversation that’s going on.”

Julian Phillips, AVI-SPL

Julian Phillips (Image credit: AVI-SPL)

Phillips believes that Pro AV is well-positioned to leverage artificial intelligence—more so than others in the tech space. “I think the sooner that we in the AV industry understand the advantages that we have, we will create the next big surge of growth,” he said.

This is because for AI to be trained well, it requires solid data. AV professionals are experts in deploying cameras and microphones—what Phillips argues are among the most powerful sensors out there, capable of capturing, processing, and analyzing rich data in real time. “We know about camera technology. We know about microphones. We also know how to network them,” Phillips said. “It’s a natural advantage we have that puts us in a very powerful position.”

For example, imagine an industrial manufacturer that wishes to streamline its production process. By analyzing video feeds, AI could suggest how to cut costs while at the same time speed up operations. 3D augmented and virtual reality models can assist clients in visualizing how these modifications would look on a practical level.

 [Blueprint for Success: CAIO and Arrivederci]

“And guess what? When that [new] factory is built, just think about all the AV that’s going to be in there—the cameras and microphones watching what’s going on, ingesting information, analyzing it, and spitting it out onto large visual displays or [into] control rooms that are monitoring the way the factory is run,” Phillips illustrated.

The Sales Pitch

Christopher Pennell, Frost & Sullivan

Christopher Pennell (Image credit: Frost & Sullivan)

The hype surrounding AI makes it difficult for customers to determine how this technology can actually help them. Christopher Pennell, industry principal at research firm Frost & Sullivan, noted that the organizations his company works with are trying to understand how AI can solve their business challenges. This means AV integrators should focus on selling solutions rather than specific technology.

“It’s not trying to sell AI per se," Pennel offered. "It’s trying to understand the challenge that the customer is trying to use AI to address."

In some cases, AI may be the best way to resolve these issues—while in others, it may be overkill. AV integrators can play a key role in helping their customers navigate this landscape. “The cost may put a client off doing something, or it might just be that it over-engineers the solution [and] locks down a client to a certain approach that they might not want in two or three years,” Pennel said.

Keith Yandell, AVI Systems

Keith Yandell (Image credit: AVI Systems)

AI also has a role in expediting an AV integrator’s internal functions. Keith Yandell, vice president of innovation at AVI Systems, noted that his organization is exploring how AI can help project teams draft proposals faster. “The scope is an all-important piece of our business: It puts a fence around responsibilities,” Yandell said. “[We’re] leaning on AI—Copilot and other tools—to help us write a scope quickly and work with our customer to understand the job at hand, get it underway, and get it done.”

Managed services is one area where Yandell sees the demand for AI-driven solutions gaining traction, particularly when it comes to monitoring and management. “When a device triggers some kind of alert, how much automation can we have around sending queries to that device to get a better idea of what’s failing so that we can solve the problem sooner?” he asked. “That’s where, on the managed services side, AI is becoming attractive to our customers.”

Keep It Secure

The problem with public, cloud-based AI tools is that they’re public. That means the data these solutions are trained on is available to anyone. If an AV firm is using these systems to generate proposals, it’s potentially sharing its intellectual property with the competition, Yandell noted.

[Cybersecurity and Pro AV]

For this reason, organizations across every industry need to define clear parameters for good security hygiene as it applies to AI. Phillips argued the best way to do this is to bring AI in-house and establish rules for data governance and security without restricting users to the point where they go rogue. “You’ve got to enable people to experiment with this stuff,” he said, “but make people aware of the obligations they have with regards to security.”

Sam Kennedy, Crestron

Sam Kennedy (Image credit: Crestron)

There are also the legalities to consider, noted Sam Kennedy, senior director of product marketing at Crestron. “Certain AI programs that can identify an individual speaker—literally putting a name to a face—are illegal in some U.S. states and global jurisdictions,” he said.

Ethics is another area of concern. Should employers really be using AI to discern what moods meeting participants are in—and then save that information for future reference? And how well can artificial intelligence differentiate between real anger, sarcasm, or simply kidding around?

“This is why an AI program—especially those that are generative or meant to act as a virtual assistant—should default to ‘opt-in,’” Kennedy said. In other words, the system won’t capture this data unless participants allow it to.

David Martin, Nureva

David Martin (Image credit: Nureva)

TJ Adams, Q-SYS

TJ Adams (Image credit: Q-SYS)

The same could be said for conferencing platforms that allow users to create personal voice profiles that are then automatically recognized by the system and stored by the platform developer. “I think the most interesting question is: Are you willing, as a participant in corporate life, to sign up for being a trusted participant?” challenged David Martin, co-founder and CTO at Nureva. “Those are the types of questions that will be facing all of us as we move forward with these types of technologies.”

For TJ Adams, vice president of systems product strategy and development at Q-SYS, the issues related AI security in Pro AV should be an ongoing conversation. “We need to keep talking about this as cultural norms evolve alongside our AI technology adoption journey,” he said. He acknowledged that there may be no easy answers to questions, for example, about privacy and the treatment of user data—at least not yet. “However, if we stay watchful, empathetic, and ethical, we can handle it with integrity together.”

Carolyn Heinze has covered everything from AV/IT and business to cowboys and cowgirls ... and the horses they love. She was the Paris contributing editor for the pan-European site Running in Heels, providing news and views on fashion, culture, and the arts for her column, “France in Your Pants.” She has also contributed critiques of foreign cinema and French politics for the politico-literary site, The New Vulgate.