AI Expected to Change the Way Carriers and Their Customers Operate

“Everything is changing, from your tech stack to your processes to the way you interface with customers,” Tinic said. Qualcomm moved its entire tech stack from on-premise to the cloud, which helped with resiliency, scale and innovation, he said. “We moved out of large, monolithic platforms,” but one thing that didn’t change dramatically yet was the “macro workflow,” he added. “The quote-to-cash process was pretty similar to what it was before, but we automated it as much as we possibly could.”

That will change as AI agents take on a larger role, Tinic said. “You have a new digital worker in the environment,” and “the way that worker operates is completely different. ... It’s going to be able to handle more complex solutions” and “massive amounts of data in a way that you and I don’t do every day.”

The “good news” is that companies that have made a lot of progress on their digital transformation are already in “a good position,” Tinic said. “You’re still going to have a fair amount of re-architecture to do to change the way your workflows work.” But companies that fell beyond have “homework to do,” he said. They have “double duty.”

The only thing “big tech companies care about is AI -- they’re frothing at the mouth,” said Kevin Chung, chief strategy officer at Writer, an AI platform. Tech company executives are asking, “What is that next new AI innovation that’s going to come out that we can take advantage of?” he said.

Chung said he also recently met with media and entertainment companies, whose executives are asking whether AI offers anything that could “completely change the way the creative industry is going to work.” They view AI with “skeptical optimism,” he said. “Those two dynamics are completely different, and the vibe is completely different.”

The reality is that, based on a recent survey by the Massachusetts Institute of Technology, large organizations have spent more than $30 billion on AI “and have gotten zero dollars in return on that investment,” Chung said. “That’s rough.”

Vijay Karunamurthy, a founder of Scale AI, said some of the biggest “wins” that telecom providers and other businesses are seeing from AI are in customer service. Measurements of customers' loyalty, churn and ability to find the products they need show that AI is working, he said. “The great thing about voice agents is they can more reliably take in all of these campaigns of information that you want to share and actually integrate them organically.” When consumers call with complaints, the resolution rate using AI agents is probably only 60%-70%, but “we’re still in the early days,” he said.

Ben Kus, chief technology officer for the content management platform Box, said companies are increasingly using AI to handle data that used to be difficult to analyze. Some of the best software developers use AI agents to address 10 different problems at the same time, “almost as if they had 10 different interns.” They're becoming “managers of AI agents” that do the work, he said, a model that's also being used in other parts of businesses.

AI has been “driving value for more than a decade,” but it was just called "data science," said Beena Ammanath, who leads the Global Deloitte AI Institute. AI is “still evolving,” so a year ago everyone was talking about generative AI, while the focus this year is about AI agents, she said. “Next year, it might be about physical AI.” Generative AI made “AI more common, accessible, something anyone can touch and feel,” but for the tech world, the technology has been around for at least 15 years, she said.

AI is “a broad bucket,” Ammanath said, questioning whether 95% of AI projects fail, as MIT has reported. “I certainly have not seen that.” AI has been most successful so far in industries that have a lot of good data available “in a structured way" and staff members who understand the data, she said.

John Watson, group president of Bell Business Markets, said most people don’t know the extent to which Bell Canada has already built “a great tech business.” It's working with Canadian AI company Cohere on ways to make greater use of AI for customers, he said. Bell launched its “AI fabric” in May and secured 500 megawatts of power for its AI centers. AI for business is hard, Watson added. “We don’t make it easy; we make it easier.”

Canada has put more focus on growing “sovereign” capabilities within the nation, but that’s not the same as being separate in everything, said Cohere President Martin Kon. He cited recent comments from Evan Solomon, Canada's minister of AI and digital innovation, who said he wouldn't buy chocolate, regardless of quality, just because it's from Canada. Likewise with technology, Kon said, “you can’t compromise on quality and on technical leadership just because it’s sovereign. ... That’s true for both” Cohere and Bell.

Some companies are taking consumer models for AI or consumer chatbots and “plopping them into enterprise,” Kon said. “That’s not going to work.” If it were easy for businesses to adopt AI models, no enterprise would be able to differentiate itself from competitors, he said.