AI adoption is surging faster than CX teams can adapt

AI adoption is moving faster than AI readiness

AI adoption has reached 88% of organizations. Employee readiness is far lower.

McKinsey’s 2025 State of AI report found that 88% of organizations use AI in at least one business function, up from 78% a year earlier. Generative AI adoption reached 79%, compared with 71% in 2024. AI has already moved into mainstream business operations.

The workforce data tells a different story. Gallup reported that 13% of U.S. employees use AI every day at work. Twenty-eight percent use it at least a few times per week. Only one in four employees say their organization has a clear plan for integrating AI into their jobs.

This is the central problem for executives. Technology deployment is running ahead of operating readiness. Companies can buy AI tools quickly. Building the skills, workflows, governance and management practices needed to use them well takes longer.

CX leaders face this gap directly. AI capabilities change quickly, while customer expectations and internal demands are rising at the same time. There is also no stable maturity model that tells every company what successful AI adoption should look like. Each organization must determine where AI creates value within its own customer journeys, data environment and operating model.

The executive response should therefore focus on enablement. AI adoption rates reveal very little about whether a company is improving customer outcomes or employee productivity. Leaders need to examine actual usage, employee competence and the business results produced by AI-assisted workflows.

AI literacy belongs inside this operating model. Employees need to understand how to frame tasks, provide relevant context, assess AI output and recognize when human review is required. Governance must develop alongside these skills. Faster deployment without clear accountability can increase inconsistent outputs, poor decisions and operational risk.

The companies that close this readiness gap can turn widespread AI access into useful organizational capability. For CX executives, that means treating workforce enablement as a core part of AI implementation rather than a training activity added later.

Delaying AI experimentation creates its own risk

Customer service is already well into AI adoption. Salesforce’s State of Service report finds that 85% of service organizations use at least one form of AI. In 2026, 66% are using agentic AI, up from 39% the previous year.

Agentic AI raises the significance of this shift. These systems can pursue defined objectives and carry out multi-step tasks with varying degrees of autonomy. That expands AI’s potential role beyond generating content or summarizing information. It can become part of how service work is executed.

For CX teams, several practical use cases are already clear. AI can synthesize large volumes of customer feedback, analyze behavioral data and identify recurring friction. It can connect customer sentiment with operational and financial information. It can also bring together data from multiple enterprise systems, helping leaders understand what customers experience and how the business is responding.

Customer expectations add urgency. Eighty percent expect highly personalized and anticipatory experiences in real time. Seventy-two percent expect seamless service across digital and physical touchpoints. Sixty percent expect AI-powered experiences that still feel human and consistent with the brand.

Waiting for a settled set of best practices is therefore a weak strategy. AI capabilities, operating practices and customer behavior are evolving together. Organizations that run controlled experiments now can learn which applications improve outcomes in their own environment. Those that wait delay the learning process as well as the deployment.

This does not require uncontrolled adoption. Executives should define narrow use cases, clear owners and measurable outcomes. A CX team might test AI for feedback synthesis, compare its findings with human analysis and measure time saved and insight quality. A service team could introduce an AI-assisted workflow for a limited customer segment and monitor resolution quality, escalation rates and customer satisfaction.

This approach converts experimentation into evidence. Successful use cases can expand. Weak ones can be changed or stopped. Governance can also improve using observed risks instead of assumptions alone.

The strategic risk is therefore broader than choosing the wrong AI product. Companies also face the cost of delayed organizational learning. As AI becomes part of normal service operations, practical experience compounds. CX leaders should start building that experience while keeping scope, accountability and customer outcomes under control.

AI literacy is becoming a core business capability

Seventy-two percent of leaders say AI literacy is important for day-to-day work. Only 35% report having a mature, organization-wide AI literacy program. That 37-point gap exposes a clear operational constraint: organizations are deploying AI faster than they are teaching people to use it effectively.

AI literacy means more than knowing how to access a tool. Employees need to understand how to define a task, provide useful context, evaluate an AI-generated response and improve the result. They also need to recognize errors, uncertainty and situations that require additional review.

These skills matter in CX because the inputs are complex. Customer feedback, behavioral signals, service records and financial data can produce different conclusions depending on how a question is framed and which context is supplied. AI can process this information quickly. Employees still need enough domain knowledge and AI competence to judge whether the output is useful for a business decision.

The absence of a universal AI roadmap makes internal learning more important. AI products and capabilities continue to change, and successful practices will vary by company, function and use case. A CX organization working with sensitive customer conversations has different requirements from a team using AI to summarize survey responses.

Executives should make AI literacy part of operational capability building. Training should connect directly to real workflows. Employees can learn how to use AI for customer-feedback analysis, research, reporting or repetitive administrative work while managers monitor quality and business outcomes.

Governance should develop at the same time. Employees need clear rules for customer data, confidential information, human review and accountability. A mature literacy program gives people the skills to use AI productively within those boundaries.

Regular use is critical. Reading guidance and viewing demonstrations can establish basic knowledge. Practical work develops the ability to ask better questions, supply the right information and identify weak outputs. Organizations that create structured opportunities for this practice can build capability while their AI strategy continues to evolve.

CX leaders should build AI capability through continuous experimentation

Forty-two percent of employees expect AI to significantly change their role within the next year. Seventy-nine percent believe they will need new skills to keep pace. The workforce requirement is already clear: AI skills need to develop alongside the technology.

CX leaders have a strong starting position. They already understand customer needs, sources of friction and the relationship between customer experience and business performance. The immediate task is to apply that expertise to AI-enabled workflows and determine where the technology improves speed, quality or decision-making.

Practical opportunities are readily available. Teams can use AI to synthesize customer comments, connect sentiment with behavioral and financial data, automate repetitive work, strengthen executive reporting and surface patterns that previously required hours of manual analysis. Each use case creates an opportunity to improve both the workflow and the employee’s AI skills.

The objective should be disciplined experimentation. Teams can begin with defined tasks, establish quality criteria and compare results over repeated use. Employees learn how much context an AI system needs, which instructions produce reliable outputs and where additional validation is required.

Executives also need measurable outcomes. Time saved is useful, but it captures only one dimension of value. CX teams can assess whether AI improves the quality of insights, speeds issue identification, supports better decisions or contributes to stronger customer and financial outcomes. Clear measures help leaders decide which experiments deserve wider deployment.

Continuous learning also prepares organizations for changing roles. With 79% of employees expecting to need new skills, traditional one-time training will have limited value in a fast-moving technology environment. Learning should become part of normal work through repeated use, review and refinement.

The goal is sustained capability. CX leaders who develop AI skills through real business problems can build confidence from evidence and improve their methods over time. That creates a stronger basis for scaling AI as customer expectations, employee roles and the technology itself continue to change.

Adaptability is now a core requirement for CX leadership

AI changes quickly. CX operating models will need to change with it. For executives, the central challenge is building an organization that can learn and adjust while maintaining service quality, governance and accountability.

CX leaders already operate in an environment shaped by economic cycles, digital transformation and changing customer expectations. AI increases the speed of these changes. New capabilities can alter how teams analyze feedback, identify customer problems, automate routine work and communicate insights to senior management.

This creates a leadership issue as much as a technology issue. Executives need to decide where AI should enter workflows, which decisions require human oversight, what outcomes teams should measure and when a successful experiment is ready to scale. These decisions require strong knowledge of customers and business economics alongside growing AI competence.

Adaptability also requires an operating process. Teams need permission to test defined use cases, evaluate results and change their approach when evidence supports it. Leaders can set clear boundaries around customer data, security, accuracy and accountability while giving employees enough scope to learn through practical use.

The pace of AI development makes continuous learning especially important. A fixed set of skills can lose relevance as models, products and workflows improve. CX organizations should regularly review where AI is creating value, where employees need stronger skills and where human expertise remains essential to customer and business outcomes.

This approach also changes how executives should evaluate progress. The number of AI tools deployed is a weak measure of organizational capability. Better indicators include employee proficiency, adoption within useful workflows, improvements in decision speed and quality, customer outcomes, productivity gains and measurable financial impact.

Leadership should keep customer value as the governing objective. AI can increase the speed and scale of analysis, help teams find friction earlier and connect customer signals with operational and financial results. These capabilities matter when they lead to better decisions and stronger experiences.

CX leaders do not need certainty about AI’s final shape to act effectively today. They need a disciplined process for learning, measuring results and adapting. Organizations that build this capability can respond to new AI developments with evidence, clear controls and a stronger understanding of where the technology creates business value.

Key highlights

• Close the AI readiness gap: AI adoption has reached 88% of organizations, yet only 13% of U.S. employees use AI daily. Leaders should align technology deployment with workforce skills, governance and clear operating practices.
• Experiment before practices settle: Salesforce reports that 66% of service organizations use agentic AI in 2026, up from 39% a year earlier. CX leaders should run controlled, measurable experiments now to build practical experience and identify valuable use cases.
• Make AI literacy an operating capability: While 72% of leaders consider AI literacy important for daily work, only 35% have mature organization-wide programs. Build AI skills around real workflows, with clear standards for data, quality, human review and accountability.
• Build skills through continuous use: Forty-two percent of employees expect AI to significantly change their roles within a year, while 79% expect to need new skills. Give CX teams recurring opportunities to apply AI to customer analysis, automation and decision support, then measure and refine the results.
• Design CX organizations for continuous adaptation: AI capabilities and workflows will keep changing. Leaders should measure progress through employee proficiency, useful adoption, decision quality, customer outcomes, productivity and financial impact.