Rearchitecting enterprise customer experience for the MCP generation

The rules of AI write themselves anew almost daily. People and brands are rushing to define what AI means and how to make it work, but a single question stands above the rest: Who controls the future of your customer and employee experiences—you, or your vendors? At Avaya, we believe the answer should be you. That’s why we are embracing the open future of AI, building the Avaya Infinity platform to support Model Context Protocol (MCP). It’s the right move—for our customers, for the market, and for the kind of trust and innovation that will define what’s next for enterprises and public sector agencies. 

Read the press release

For too long, integrating AI into customer experience has meant choosing the lesser of two evils: 

• Lock-in with a single provider’s closed ecosystem, sacrificing flexibility and future optionality 
• Chaos through brittle, costly, one-off integrations across multiple tools and data silos 

This false choice limits innovation and burdens IT teams. It creates fragmented experiences that erode trust—with both customers and employees. In a world that moves at the speed of AI, this approach is simply unsustainable. 

Enter MCP—the open, vendor-agnostic standard changing everything. With MCP, AI models can interact seamlessly with tools, data, and operational logic—securely and at scale. A favorite metaphor for the few opining on this topic is to think of it as the USB-C for AI: one universal connector that works across models, platforms, and enterprise systems.   

Created and released by Anthropic in late 2024. The rapid adoption of MCP by major technology players such as OpenAI, Google DeepMind, and Microsoft underscores its strategic importance. This alignment indicates a collective realization that proprietary integrations hinder market growth more than any advantage gained by closed ecosystems. We are currently witnessing and co-creating a foundational shift toward interoperability, transparency, and user control. 

MCP enables AI systems to operate with dynamic awareness of user, session, and business context. For Avaya Infinity, this means models can consistently deliver responses that are not only accurate but also relevant, personalized, and on-brand — even across fragmented ecosystems and at enterprise scale. This level of contextual intelligence is critical for organizations managing thousands of workflows, users, and integration points across operations. 

In a typical enterprise, there are 'N' number of tools and data sources (e.g., Salesforce, a product database, a knowledge base) and 'M' number of AI models or applications that need to access them. Before MCP, connecting each tool to each AI model required a unique, point-to-point integration. This created a combinatorial explosion of development work, resulting in an architecture that was expensive to build, difficult to maintain, and inherently unscalable.  

MCP transforms this paradigm by introducing a standardized intermediary layer, converting the complex N x M problem into a simple and scalable "N+M" solution. Each of the 'N' tools only needs to be wrapped with a single MCP server, and each of the 'M' AI models only needs a single MCP client. Once a component can "speak" MCP, it can communicate with any other MCP-compliant component in the ecosystem.   This dramatically reduces development overhead, eliminates redundant work, and eradicates the technical debt associated with maintaining dozens of fragile, custom connectors. Investments in new technologies, channels, and modalities meant to unify customer interactions often fracture them. Customers end up feeling less known, less cared for, and less connected to the businesses that serve them than ever before. For many organizations, this pain is felt most deeply in the contact center.

Avaya is turning this tide, unifying the fragmented contact center experience. By connecting the channels, insights, technologies, and workflows that help deliver best-in-class interactions.

With MCP, Avaya Infinity users will realize even deeper business value: faster resolution times, more natural conversations, stronger brand consistency, and higher satisfaction across customer and employee interactions. It unlocks a new tier of performance in generative AI, where outputs aren’t just intelligent — they’re situationally aware and strategically aligned.  

Early research suggests MCP will yield dramatic improvements for enterprises, saving time and money in a number of areas: