Orchestration is the new challenge for CX in the age of AI agents

Enterprises are rolling out AI agents, voice AI and automation across messaging, voice and digital channels faster than the systems built to support them. Gaurav Anand, who leads Tata Communications’ Customer Interaction Suite, says many companies have simply bolted conversational AI onto legacy setups that were never designed for it.

That creates a mess for the people still doing the fixing. Human agents have to stitch together context from disconnected tools, while the company itself lacks a shared view of customer identity, conversations, transactions, policies and workflows. The old CX stack was built for linear routing by humans. It was not built for autonomous AI systems, data lakes and people passing work back and forth in real time.

Anand’s bigger point is that the priority is shifting. Automation handles isolated tasks. Orchestration connects those tasks into outcomes, with AI agents, applications and human workers all using the same understanding of the customer and the business. He says the real advantage now is not in piling on more tools, but in deciding how those tools hand off, collaborate and escalate.

He also says the industry keeps repeating the same mistake: putting a voice AI layer on top of an old system and calling it progress. That just recreates the rigid phone menus AI was supposed to replace. Tata Communications’ answer is the Interaction Fabric, which ties together contact center, messaging, collaboration, AI and customer data so conversations can move across voice, WhatsApp, chat, email and CRM without losing context.

The network matters too. Anand argues that if the underlying infrastructure can’t keep up, latency and inconsistent journeys creep in as customers switch channels. His broader forecast is a CX model built around shared context, real-time intelligence and AI agents that can increasingly resolve interactions on their own, with humans stepping in where judgment and empathy still matter.