For the past few years, AI has dominated almost every customer experience conversation. It has been pitched as the answer to rising costs, growing customer expectations and stretched contact center teams. But as customer experience leaders look ahead to 2026, a more pragmatic shift is taking place.
Instead of asking what AI they should buy, enterprises are starting to ask a more uncomfortable question: what problem are they trying to solve?
That reframing marks a turning point in how AI is being adopted across customer experience, and it’s forcing vendors and enterprises alike to rethink their roles, Chris Angus, Vice President, CPaaS and CX Expansion at 8×8, told CX Today in an interview.
“AI has become more practical,” Angus said, noting that the hype and “top down direction we saw the last few years,” are beginning to shift.
The Internal-First AI Strategy Redefining Customer Experience in 2026
Much of the early momentum around AI in CX was driven by executive mandates that pushed teams to “do something with AI” long before they had clarity on where it would add value, Angus said.
“I sat on many CIO panels where the top-down messaging was the driving force, not quite sure how to use or where to implement it, how to put pounds towards it, because it’s hard to give you an ROI without a tried and tested mechanism.”
That was much like the customer service contact center was a decade ago, when it was viewed as a cost center rather than a revenue generator, Angus added.
The result was predictable. AI was often rushed into customer-facing roles before the foundations were in place, leading to disappointing experiences and frustrated customers.
Now, the focus is turning inward. In 2026, Angus expects “a continued shift into the outcomes-based focus for how you use AI.”
“What we’re seeing now is an internal viewpoint first. A lot of focus has been put on how we’re using AI and some other features within the contact center and our communication space to service our internal teams, to streamline internal mechanisms first.”
This internal-first approach reflects a growing recognition that employee experience and customer experience are inseparable.
“We really focused on the employee experience and the customer experience not being decoupled… sharpening the tools we give our staff to use, making their jobs easier, more efficient, more effective, improving how we train them and how we coach them… making them happier in their seat.”
The logic is simple.
“It sounds really simple, and there are no aha moments… guess what? They give a much better customer experience,” Angus noted.
The Unglamorous Reality of AI Data Health
As AI moves from hype to implementation, enterprises are also being forced to confront a less exciting truth: many aren’t ready.
“One of the biggest challenges we face when implementing new policies, procedures or in an AI, is that the data lakes and the data warehouses are not consolidated,” Angus said.
Without centralized, accessible data—whether that’s employee data, contact center transcripts, thought leadership, scripting, knowledge bases or information about customer behavior—even the most advanced AI tools struggle to deliver value, Angus said.
“If you don’t have a centralized data lake… it won’t track all those trends and you can’t use these tools to improve the income.”
There is a growing misconception that AI can fix this problem on its own, Angus said.
“The interpretation of data is where you can use AI, but if you’re running across multiple, disparate systems, you can’t use AI to collect that.”
Enterprise leaders are taking a step back to make sure they are capturing and recording all of their voice calls and bringing together data on customer interactions and purchase history before they put AI systems in place. Skipping that process has resulted in many failed initiatives.
“People don’t think about that till the last minute,” Angus said.
“This kind of step back approach allows us to give recommendations of get your data health together better first, then you can start using predictive insights to enhance employee and customer experience.”
Why Trusted Advisors Matter More Than Ever
As AI matures, so too does the buying process around it. Enterprise buyers tend to be highly knowledgeable about the solutions available on the market, but that can mean they look to buy technology they’ve heard about and feel they need to keep up with the market.
“Typically, by the time they proactively reach out to somebody to purchase, they’re probably 40 percent through their buying cycle already. They’ve made decisions already actively before speaking to you,” Angus said.
That creates a dangerous dynamic: customers arrive with a solution in mind, not a problem statement.
“You don’t know what you don’t know,” Angus cautioned.
“It’s our obligation to slow down that that journey… I would always ask customers to focus on the outcome you’re trying to achieve, and partner with an organization that you can trust to advise you accordingly… If you’re trying to purchase a solution, you don’t need to come ready armed with what you think the solution is… it allows you to be iterative and go on that process together.”
This is where the role of the vendor must change from solution seller to trusted advisor, Angus said.
And that often means saying something uncomfortable.
“I’ve done so many consultations where I say ‘actually, AI is not the solution for you.’”
In some cases, the best answer is not more technology at all.
“We can solve that with some really basic machine learning… You’re just not using tools you have today in the best way possible.”
That honesty is what builds trust and long-term relationships. “You need to have someone you trust to take you through that conversation,” Angus said.
This shift toward outcomes-based thinking is reshaping how AI success is measured.
Rather than deploying AI because competitors are doing it, enterprises are starting to evaluate whether it genuinely improves the customer experience. It’s a slower, more deliberate approach that reflects the reality of constrained budgets and rising expectations.
“Budgets are tight. Every company, every department, has been forced to do more with less.”
That pressure is also influencing channel strategy, investment decisions, and the renewed emphasis on pragmatic innovation rather than flashy transformation.
Incremental Security Improvements Without Disrupting the Customer Experience in 2026
The same maturity is emerging in how enterprises think about security and risk heading into the new year. Rather than bolting on complex tools that disrupt customer journeys, enterprises are looking for incremental improvements to make security gains without customers ever noticing, Angus said.
“You can start small… You can look at things like multi-factor authentication and one-time passwords through… the channels customers use already.”
“You can do it through step changes. You can implement end-to-end encryption for your voice and your digital channels without the customers being impacted at all. It also shows your customer base that you’re investing in their security, not your own return on investment.”
Real-time fraud and anomaly monitoring and detection, such as artificial traffic inflation with CPaaS can help to uncover fraudulent activity. Global organizations also need to consider data residency requirements and how they secure data. “The biggest risk is normally the way you store data, how you’re looking after it,” Angus said.
“These are things you can do behind the scenes to secure your customer interaction with you, without having to increase costs.”
More GDPR-style rulings around AI governance and the impact on customer experience are to be expected in 2026, Angus noted.
“As we process more data through AI, I think we’ll see stricter protocols we have to adhere to on how we process it to stop the data leakages and hallucinations.”
Governments in Europe and elsewhere are likely to put more checks and balances in place to ensure businesses are held to account on how they service customers and manage their data.
What Enterprise AI Maturity Looks Like in 2026
Looking ahead, the enterprises that succeed won’t be those with the most AI, but those with the clearest intent.
“If everything’s a priority, nothing’s a priority,” Angus said.
In that sense, the next phase of AI in customer experience in 2026 isn’t about intelligence at all. It’s about judgment. As AI grows up, leaders are discovering that the smartest move is not always to buy more technology but to ask better questions, earlier, with the right people at the table.
