How CX leaders are planning for specialized agents

For many customer service teams, the first chapter of AI was about getting started. Piloting a chatbot. Automating routine questions. Learning where AI could help. The next chapter looks different.

As AI becomes more capable, CX leaders are beginning to plan for specialized agents with different roles across customer service. According to Zendesk’s recent research, 83% of business leaders expect to expand their use of specialized AI agents over the next 12–24 months. That raises a new set of questions—not just about technology, but about where AI creates the most value, how different agents work together, and how customer service teams evolve alongside them.

Here are six questions forward-thinking service teams are beginning to ask:

1. Where will specialized AI create the biggest impact?

The temptation with any new technology is to look for as many use cases as possible. A better place to start is with the work itself.

Think about the customer journeys that already rely on specialist knowledge, repeatable decisions, or well-defined processes. Technical support. Billing. Returns. Employee onboarding. It's also worth looking beyond the customer conversation itself. Resolving an issue often involves approvals, business systems, internal teams, or behind-the-scenes processes.

Rather than asking, "Where can we use AI?" start by asking, "Where would specialized expertise make the biggest difference?" The answer will help you identify the first opportunities worth investing in.

2. How many AI agents do you actually need?

For many organizations, one AI agent is the right place to start. But as your service operation grows, different parts of the customer journeys begin to rely on different knowledge, business rules, and workflows. Trying to make one AI agent handle every situation can quickly become difficult to manage.

3. How will your AI agents work together?

Adding more AI agents introduces a new planning challenge. Imagine a customer troubleshooting a product issue that turns out to need a replacement. Different agents might handle each step, but the experience should feel seamless, with context moving between agents, systems, and people.

Customers don't think about individual AI agents. They think about whether their problem was solved. That's why it's worth planning how work moves between agents, systems, and people just as carefully as you plan what each agent is responsible for.

Customers don't think about individual AI agents. They think about whether their problem was solved.

4. How much control should your CX team have?

Customer service teams know where customers experience friction better than anyone. As AI tools become easier to build and improve, the people closest to the customer have more opportunity to shape those experiences. Giving CX teams the ability to test new ideas, refine workflows, and improve the customer experience means they can respond more quickly as customer needs change.

The opportunity is to put AI improvement into the hands of the people who understand customers best.

5. What new skills will your team need?

Your team probably doesn't need everyone to become an AI expert. They do need people who understand customer journeys well enough to identify where AI creates value, test new workflows, and improve them over time.

The skills becoming more valuable are familiar: designing workflows, evaluating what's working, improving customer journeys, and understanding where people add the most value. They're familiar skills applied in new ways.

For service leaders, the role is expanding too. Alongside leading people, they'll increasingly be responsible for designing how people and AI work together—deciding where AI should step in, where human judgment creates the most value, and how both continue to improve over time. New roles like AI Service Architects and AI specialists will emerge, helping teams build, manage, and improve these experiences.

Your team probably doesn't need everyone to become an AI expert. They do need people who understand customer journeys well enough to identify where AI creates value, test new workflows, and improve them over time.

6. How will you keep improving over time?

Planning for specialized AI agents doesn't stop once the first one goes live. Build regular opportunities to review what's working, identify where AI agents create more value, and decide what comes next. Some ideas will prove more valuable than others, and new opportunities will continue to emerge.

These questions don't need perfect answers today. They give you a framework for making better decisions as AI evolves and new opportunities emerge.