The CX and AI industry has made remarkable progress in a very short period of time. Models are becoming more capable, more accurate, and more accessible.
Every CX|AI demo looks amazing these days.
The virtual agent answers questions instantly. The AI assistant summarizes conversations. The chatbot retrieves customer information, updates records, and appears capable of handling tasks without human intervention.
Then the real work begins.
Over the last year, I’ve participated in dozens of conversations with organizations evaluating new technologies designed to improve customer experience. Whether the goal is increasing self-service containment, lowering average handle time (AHT), improving quality management, or reducing operational costs, the discussion almost always reaches the same point:
“Can the platform access the systems it needs to do its job?”
It’s a simple question that often uncovers a much larger challenge.
The platform may need to access the CRM, a knowledge base, an order management system, a ticketing platform, workforce management software, an ERP, an EHR, or a server-based application that has been running the business for years. The technology itself may be capable, but connecting it to the systems that contain customer information and business processes can quickly become the biggest obstacle to success.
This is where a growing concept called Model Context Protocol, or MCP, is beginning to attract attention.
While it may sound like another technical acronym, MCP has the potential to significantly influence how organizations deploy and scale CX|AI across customer experience operations.
The Hidden Problem Behind Most AI Initiatives
Most organizations don’t have a technology problem.
They have an integration problem.
The CX and AI industry has made remarkable progress in a very short period of time. Models are becoming more capable, more accurate, and more accessible. Vendors are adding new AI-powered features to their platforms at a pace that would have been difficult to imagine just a few years ago.
Yet many organizations struggle to move beyond pilot programs.
Why?
Because answering a customer’s question is only part of the challenge.
To create meaningful business value, AI needs context.
It needs access to customer records. It needs order history. It needs knowledge articles. It needs account information. It needs the ability to complete actions, not simply provide information.
The moment AI needs to interact with multiple business systems, complexity increases dramatically.
Historically, that complexity has been addressed through APIs.
APIs Built the Modern Customer Experience Stack
APIs have been the foundation of customer experience technology for decades.
They allow systems to exchange information and execute actions. They enable contact centers to connect with CRM platforms, workforce management solutions, analytics tools, billing systems, and countless other applications.
Without APIs, many of the customer experiences we take for granted today would not exist.
When a customer record appears automatically on an agent’s screen, an API is likely involved.
When a virtual agent checks an order status, an API is involved.
When a quality management platform retrieves call recordings for evaluation, an API is involved.
APIs are not going away.
In fact, they will remain one of the most important components of modern CX architecture.
However, APIs were largely designed for developers and applications, not autonomous AI systems.
That distinction matters.
Why AI Requires a Different Approach
Traditional integrations are typically designed with a specific workflow in mind.
Developers determine which systems need to communicate, what information should be exchanged, and how the process should operate.
AI-powered systems introduce a different dynamic.
Instead of executing a predefined workflow, AI systems are increasingly expected to reason, make decisions, retrieve information, and perform tasks dynamically based on the situation presented.
For example, a customer may contact a business and say:
“I need to change my shipping address, check the status of my order, and speak with someone about my invoice.”
A human agent can easily determine which systems need to be accessed and in what order.
For CX|AI, that process is significantly more complex.
The system must understand what tools are available, what actions those tools can perform, what information is required, and what permissions are necessary to complete the task.
This is the challenge MCP is attempting to address.
Think of MCP as a Universal Translator for AI
One of the simplest ways to think about MCP is as a standard framework that helps AI systems discover and interact with business applications.
If APIs provide the doors into business systems, MCP helps the AI understand which doors exist, what is behind them, and how they can be used.
Instead of creating a custom integration every time a platform needs to interact with a business application, MCP seeks to create a more consistent and standardized approach.
We are not talking about replacing APIs.
We are talking about making APIs easier for AI systems to understand and use.
Some industry observers have compared MCP to USB-C. Before USB-C, every device seemed to require its own cable and adapter. USB-C didn’t replace electricity. It standardized how devices connected to it. MCP aims to do something similar for AI and enterprise applications by creating a common way for AI systems to discover and interact with business tools and data sources.
For readers interested in learning more about the analogy, IBM published a helpful overview of MCP and why many organizations are referring to it as the “USB-C for AI.”
See: IBM: What is Model Context Protocol (MCP)?
Why CX Leaders Should Pay Attention
At this point, some customer experience leaders may be wondering why any of this matters.
After all, isn’t this simply another technical discussion?
Not really.
The organizations that successfully deploy AI over the next several years will likely be the ones that can connect AI to business systems quickly, securely, and at scale.
Today, many AI projects require significant integration work before they deliver value.
As organizations expand from a single use case to dozens of use cases, that effort can become difficult to manage.
Consider the number of systems involved in a typical customer interaction:
- Contact center platform
- CRM, ERP, or EHR
- Knowledge management
- Workforce management
- Ticketing
- Billing
- Order management
- Shipping and logistics
- Scheduling
- Identity verification
- Business intelligence
Now imagine deploying AI across all of them.
The ability to standardize connectivity could become just as important as the intelligence of the AI itself.
The Questions Every CX Leader Should Start Asking
You do not need to become an MCP expert.
However, as vendors increasingly incorporate AI into their platforms, there are several questions worth asking:
- What is your MCP strategy?
- Is MCP supported today or planned for the future?
- What systems can be exposed through MCP?
- How are permissions, guardrails and security managed?
- How does MCP fit into your broader AI roadmap?
The answers may not impact a decision being made this quarter.
They may significantly impact your flexibility over the next three to five years.
The Bigger Question: Will Connectivity Become More Important Than the Model?
For the last two years, much of the industry conversation has focused on AI models.
Which model is best?
Which model is fastest?
Which model is most accurate?
Those questions are important.
But as models continue to improve, another question may become even more important:
Can the AI access the information and systems required to create value?
A brilliant model that cannot interact with the business is of limited use.
A capable model that can securely access customer information, execute workflows, and complete tasks may ultimately deliver far greater business outcomes.
In other words, the future competitive advantage may not come from the model itself.
It may come from the architecture that allows the model to operate effectively.
Final Thoughts
We are still in the early stages of understanding how CX|AI will transform customer experience.
Some technologies will fade. Others will become foundational.
Whether MCP becomes the dominant standard remains to be seen. However, the problem it is attempting to solve is very real.
Organizations want systems that can do more than answer questions.
They want systems that can take action.
Making that possible requires a better way to connect intelligence with business systems.
For customer experience leaders evaluating technology investments today, that may be one of the most important trends to watch.
The next generation of CX automation will not be defined solely by smarter AI. It will be defined by how effectively that AI connects to the rest of the business.