AI is making it possible for customer experience teams to create more—more insights, more journey maps, more personas, more content, and more recommendations.
But more output does not automatically lead to better customer experiences.
That tension shaped the conversation at AI ❤️ CX: Less Hype. More Signal., held July 15 at The Berkshire Room in Chicago. Hosted by JourneyTrack, Thematic, Alterian, and the Institute for Journey Management, the event brought together CX practitioners for an honest discussion about where AI is delivering value, where it is falling short, and why human judgment is becoming more—not less—important.
The panel featured:
➡️ Stephanie Linville, CCXP, Customer Experience Strategy Lead at Corteva Agriscience
➡️ Dilyana Pavlova, CCXP, Director of Experience Insights at ResMed
➡️ Amy Ravit Korin, SVP of Strategy & Consulting, ConnectedCRM at Publicis Digital Experience
Their experiences varied by industry, organization, and AI maturity, but several powerful themes emerged.
One of AI’s most immediate benefits is its ability to help people get started.
Stephanie described how Corteva uses AI to support CX practitioners across regions, including people who may be new to customer experience or rotating through marketing roles. Instead of asking someone to create a persona, synthesize insights, or begin a journey map from scratch, JourneyTrack's AI can provide a starting point for the team to evaluate and refine.
That distinction matters. The AI-generated output is not the finished work. It reduces the uncertainty and hesitation that often prevent the work from beginning.
This is particularly valuable during organizational change. As Corteva separates its two business units into publicly traded companies, teams must navigate changes to data, systems, processes, and responsibilities. AI is helping new leaders and practitioners get up to speed, bridge gaps between disconnected platforms, and find information faster.
The productivity opportunity is well documented. Research summarized by MIT Sloan found that generative AI tools increased software developer productivity, with particularly significant benefits for less-experienced workers. But the research also reinforces the importance of understanding when and how AI supports the work rather than assuming the benefits will appear automatically.
For CX organizations, this presents an important opportunity: AI can democratize access to CX practices, but only if teams are also taught to question, validate, and improve what it produces.
Perhaps the most consequential theme of the evening was the changing purpose of the CX function.
Dilyana described ResMed’s evolution from a team traditionally measured by the reports, dashboards, research, and other assets it produced into an enablement function that helps product teams make better decisions.
AI accelerates that shift by taking on more of the work involved in creating and retrieving assets. ResMed is using AI to develop research guides, synthesize research, produce reports, improve intake processes, support voice-of-customer closed-loop activities, and make existing knowledge accessible through agents.
But the goal is not simply to automate production. It is to turn the organization’s body of research into a living knowledge layer that product teams can use as they make decisions.
That mirrors a broader change in enterprise AI. McKinsey’s 2025 global research found that organizations are increasingly redesigning workflows to capture value from AI, although the transition from experimentation to scaled business impact remains a work in progress. The organizations seeing the most value are not treating AI as a collection of isolated productivity tools; they are rethinking strategy, operating models, talent, technology, data, and adoption together.
Microsoft’s 2025 Work Trend Index reached a similar conclusion: 82% of leaders said the year represented a pivotal moment to rethink core aspects of strategy and operations, while 81% expected agents to be moderately or extensively integrated into their AI strategies within 12 to 18 months.
For CX teams, the question is no longer simply, “How can AI help us create this deliverable faster?” It is, “How can AI help the organization make a better decision?”
Amy offered a candid example of using an LLM to conduct a competitive website audit. The tool produced an impressive-looking table in seconds, creating the impression that the work was nearly complete. During review, however, a colleague questioned one of its claims. The website did not say what the AI reported it said.
The lesson was simple: the time saved at the beginning of the process must often be reinvested at the end.
Research, information gathering, and synthesis may happen faster, but teams still need to analyze the output, verify the underlying information, add organizational context, and socialize conclusions with the right stakeholders.
This is not merely an early-model problem that organizations can assume will disappear. Stanford’s 2025 AI Index notes that even as AI performance improves rapidly, complex reasoning remains a challenge. Models can perform extremely well on sophisticated benchmarks and still fail to solve logic problems reliably, limiting their use in situations where precision is critical.
For CX teams, an authoritative tone is not the same as an accurate answer. AI outputs should be treated as inputs to professional judgment, not substitutes for it.
One panelist's frustration drew broad agreement: AI in CX is too often reduced to customer-facing chatbots.
Chatbots may be the most visible application, but AI is appearing throughout the CX operating model:
➡️ Synthesizing research and customer feedback
➡️ Creating and refining personas
➡️ Supporting journey mapping and management
➡️ Surfacing insights from organizational knowledge
➡️ Improving research intake and discussion guides
➡️ Enabling sales and service employees
➡️ Connecting fragmented data sources
➡️ Supporting closed-loop voice-of-customer programs
➡️ Helping teams prioritize recommendations and actions
The better question is not, “Where can we put a bot?” It is, “Where can AI meaningfully improve the experience, and where will a human create greater value?”
That choice matters to customers. Qualtrics’ 2025 Consumer Experience Trends found that only 26% of consumers trusted organizations to use AI responsibly. Its research cautioned companies against overusing AI to deflect customers into self-service and emphasized the importance of providing access to a human when something goes wrong.
Similarly, PwC’s 2025 Customer Experience Survey concluded that successful brands use AI intentionally, accelerating service where it adds value while handing interactions to people when empathy and judgment matter.
The best AI experience may not be the one customers notice. It may be the intelligence helping an employee find the right answer, understand the customer’s context, or resolve an issue more effectively.
If AI allows every organization to create hundreds of personas, reports, designs, and recommendations, the competitive advantage cannot be the volume of those assets.
It becomes the ability to determine which ones deserve attention.
Dilyana described discernment and judgment as central to the future value of CX. Researchers and experience professionals understand that evidence cannot always be interpreted through a simple formula. A single customer observation can sometimes reveal a critical risk, while a larger sample can still be insufficient for a particular decision. Context matters.
Stephanie similarly emphasized critical thinking, problem-solving, and the ability to ask better questions. AI may make answers easier to generate, but CX professionals still need to test assumptions, identify what is missing, detect bias, and determine whether an output represents customers accurately.
The World Economic Forum’s Future of Jobs Report 2025 supports this dual need. AI and big data are among the fastest-growing skills, but so are creative thinking, curiosity, resilience, flexibility, leadership, and analytical thinking.
The future CX professional is not choosing between technical fluency and human expertise. They need both.
Most organizations begin AI adoption with experimentation: encouraging employees to explore tools, develop skills, and identify useful applications.
Eventually, experimentation must be supported by structure.
The panelists described organizations at different points in that transition, from early-stage exploration to formal governance teams, approved tools, frameworks, guidelines, and defined guardrails. As Dilyana put it, the process can feel like moving from chaos to “more structured chaos.”
For regulated organizations, the stakes are especially high. Stephanie explained that Corteva’s AI-enabled sales tools must draw from approved proprietary research and keep sensitive information within appropriate boundaries. Regulatory, legal, R&D, and digital teams all need confidence in how the system develops its responses.
The National Institute of Standards and Technology’s 2024 Generative AI Profile recommends managing these risks throughout the AI lifecycle, including defining human oversight responsibilities, evaluating system limitations, documenting risks, and establishing testing and validation practices.
Governance should not exist to prevent adoption. Done well, it gives employees the confidence to use AI more effectively because they understand the boundaries.
One of the most thought-provoking questions concerned early-career professionals.
If AI can create the first draft of a research plan, persona, journey, or synthesis, how do junior practitioners develop the foundational skills required to know whether that draft is any good?
Dilyana described team members trying to learn prompting while also learning how to become strong researchers. Their concern is not only whether AI will change their jobs, but how they will build meaningful careers when parts of the traditional learning process are automated.
This creates a new responsibility for CX leaders. Prompting cannot replace mastery of research, customer understanding, facilitation, design, or analytical thinking. Organizations will need intentional development models that teach employees not only how to work with AI but also how to cultivate the expertise needed to challenge it.
That may mean revisiting career paths, apprenticeship models, review processes, and how senior professionals explain the rationale behind their decisions.
AI is dramatically expanding what CX teams can produce. It can help organizations surface more feedback, generate more insights, identify more opportunities, and propose more actions.
That creates a new problem: noise.
When every team has more recommendations than it can implement, the most valuable capability becomes deciding which customer moments to address, in what order, with what expected outcomes, and how success will be measured.
That is why JourneyTrack has evolved beyond journey mapping and management into decision intelligence for customer journeys. The goal is not to add another stream of AI-generated output. It is to help CX teams connect customer evidence, business priorities, organizational goals, and expected impact so they can make—and defend—the decisions that matter most.
The Chicago panel made one thing unmistakably clear: AI can help CX teams move faster, but speed is not the destination.
Better decisions are.
Subscribe to our blog and stay in the know.