AI Disruption: Embracing the Change?

Artificial intelligence (AI) can provide operational efficiencies and create new business opportunities. The big consideration for many organizations is whether the potential benefits outweigh the challenges and risks of implementation.

I recently participated in a panel discussion with a pair of executives who have firsthand experience in integrating AI within their companies:

• 

  Jeff Bennett, CEO, Utility Trailer Manufacturing, the oldest privately owned, family-operated trailer manufacturer in the U.S.

• 

  Porter Travis, CFO, The Pacific Companies, a provider of maritime services including logistics, cargo handling, maintenance and repair.

BDO Digital USA, a leading technology advisory firm, co-sponsored the event with BMO. Eric Garrison, Digital Market Leader, BDO Digital, served as the moderator.

Following is a summary of our discussion.

One thing to keep in mind is that AI is a feature, not a product. For Pacific Companies, it meant taking a customer-driven approach to AI.

“As our container terminal operators have moved to automated equipment, it's completely changed the game. We have one customer that generates 100 billion data points a day. There are tens of thousands of potential errors that the equipment can generate, and trying to analyze using the traditional method is not really feasible. AI is a fantastic tool for looking at all that data, parsing what's needed and determining actionable trends that you can correct.”

A lot of organizations struggle with where to start with AI, especially as it seems like new platforms are constantly emerging. Deciding how to approach AI depends highly on a company’s unique operating model. One determining factor could include the technology solutions companies already employ. For Utility Trailer, as a longtime Microsoft customer, using Copilot to analyze data trends was the obvious choice.

“We already have a lot of sensitive data on OneDrive, so asking Copilot to comment on it basically wasn’t that big of a leap,” Bennett said.

Travis noted that Pacific Companies had to consider two key constraints before deciding on the right approach. One constraint was cost.

“When we took our first automated customer, we didn't anticipate having to deal with 100 billion points of data daily,” Travis said. “As a result, we needed to find a way to analyze that volume of data in a cost-effective manner. We started that journey by utilizing open-source code. This led to an iterative process of testing and learning.”

Pacific Companies’ other constraint was customer concerns around data security. “They were highly concerned about the security of the data being in the cloud,” Travis said.

As a company that operates in a highly regulated industry, BMO itself is aware of the data security concerns. That said, AI is not a trend any business can ignore, which is why we’re currently running more than 70 AI-based projects, albeit in a highly controlled and targeted manner.

According to Microsoft’s 2024 Work Trend Index,¹ 75% of global knowledge workers use AI at work. And among those using AI, 78% are using personal AI tools. The lesson here is that if your company doesn’t host an AI platform, you’re exposing yourself to data privacy and cybersecurity risks.

While Utility Trailer uses generative AI tools to analyze data trends, it also uses the technology to update its internal programs written in legacy programming code.

“I'm still amazed at the ability to have AI write programs for us,” Bennett said. “Our IT department is just scratching the surface, but it’s something that’s going to completely change what they do. The leverage you’re going to get from having a personal assistant that can help you do the boring 80% of the work that we all have to do so you can focus more on the 20% that’s the creative side or the strategic side of things is a game-changer.”

The flip side is a lot of that boring stuff used to be tasks that helped junior staff members learn the ropes and move up in the organization. As AI automates those tasks, companies will have to think about finding other ways to train staff.

Travis said this could change the skills needed for specific jobs. “So, for us, all the way from the mechanics on up, we have to look at what skills they currently have. What are they going to need in the future? How do we fill that gap? Do we need a reskilling or upskilling program?”

And for all the data insights that AI can provide, Travis also pointed out that the amount of information AI generates can be overwhelming. That’s going to force companies to innovate and expand their fields of expertise.

"For businesses that want to differentiate themselves over the long term, they're going to need to have AI in their tool belt,” Travis said.

Whether the current benefits of AI—including operational efficiencies and meeting customer demand—outweigh the drawbacks—such as data overload and security concerns—depends on each company’s unique circumstances. But we’re still in the early stages of this revolution, and the steps organizations take now will help them determine how they can move forward as the technology evolves.

¹ _{Microsoft’s 2024 Work Trend Index}