As companies race to automate more customer interactions, California lawmakers are asking a deceptively simple question: What happens when a customer just wants to talk to a person?
A recent SFGATE article highlighted Assembly Bill 1609, proposed legislation that would require companies with at least $500 million in annual revenue to give callers the option to speak with a human customer service representative. Under the current proposal, companies would need to connect customers who request a human within 15 minutes and make their customer service phone numbers clearly accessible online.
The proposal doesn't attempt to eliminate AI from customer service. In fact, its author acknowledges that automated systems can be perfectly effective for simple requests. The concern is what happens when the customer's problem becomes too complicated for automation to handle.
That distinction points to a much bigger question for contact centers. As AI becomes the first line of customer service, how good is the experience when AI is no longer enough?
The Hardest Conversations Are the Ones Humans Will Inherit
AI is getting increasingly good at resolving routine customer needs. Checking an order status, answering a basic policy question, updating account information, or providing business hours may not require a human agent at all.
That's good news for contact centers. But it also changes the nature of the conversations that eventually reach their agents.
If automation successfully handles the easy interactions, human agents increasingly inherit the exceptions: the unusual billing issue, the frustrated customer, the complicated insurance claim, the potential fraud case, or the healthcare question that requires judgment and empathy.
In other words, AI may reduce the number of human conversations while simultaneously increasing their complexity.
That makes the transition from AI to human one of the most important moments in the entire customer journey.
A Handoff Shouldn't Mean Starting Over
Most customers have experienced the bad version of an AI handoff.
You explain the problem to a chatbot. You answer several questions. You provide identifying information. The system eventually determines that it can't help and transfers you to an agent.
Then the agent asks, "How can I help you today?"
Everything the customer just did disappears.
This is where an automation strategy designed to reduce customer effort can unintentionally create more of it. The customer doesn't experience the chatbot and the agent as two separate systems. They experience one company that just made them explain the same problem twice.
A great handoff should feel completely different. The agent should already understand why the customer is calling, what they've attempted, what information they've provided, and where the automated experience broke down. The conversation should continue, not restart.
The Agent Needs More Than Context
Preserving the history of the interaction is only part of the solution. If the conversations reaching humans are becoming more complicated, agents also need better support once they take over.
This is where real-time agent assist becomes especially valuable.
An agent dealing with a complex customer situation shouldn't have to simultaneously listen, empathize, search multiple systems, remember changing policies, navigate compliance requirements, and determine the next best action. AI can support the agent in the background by surfacing relevant guidance at exactly the right moment.
The role of AI changes at the handoff. Instead of talking directly to the customer, it begins helping the human deliver a better conversation.
That's an important distinction because some of the most valuable applications of AI in customer service may ultimately be invisible to the customer.
AI Should Make the Human Experience Better
At MosaicVoice, we believe the future isn't about choosing between AI and human agents. It's about designing an experience where each does what it does best.
Automation can efficiently handle predictable interactions. But when judgment, empathy, reassurance, or complex problem-solving are required, human agents remain incredibly valuable. MosaicVoice helps make those conversations stronger by providing real-time agent guidance, accurate transcription, automated QA, and conversation intelligence that helps organizations understand what is actually happening across customer interactions.
That creates a continuous feedback loop. Organizations can identify why customers are escalating from automation, understand which issues agents struggle with after the handoff, improve coaching, update guidance, and ultimately make both the automated and human portions of the experience better.
The goal isn't simply a successful transfer. It's a system that learns from every transfer.
The Handoff Is Becoming a Moment of Truth
For years, contact centers focused heavily on reducing transfers because customers disliked being passed from one person to another. AI introduces a new version of the same challenge.
A customer who asks for a human is telling the organization something important: whatever happened before this moment wasn't enough.
That makes the next few minutes disproportionately important. A well-prepared agent can restore confidence, solve the problem, and potentially turn a frustrating experience into a positive one. A poorly executed handoff can confirm everything the customer already feared about the company's automation strategy.
California's proposed legislation may ultimately change as it moves through the legislative process. But the customer frustration behind it shouldn't be ignored. The fact that lawmakers are debating whether consumers need guaranteed access to a human is itself a signal about how customers are experiencing the current wave of automation.
The Bottom Line
The future of customer service isn't going to be entirely automated or entirely human. It's going to be a combination of both.
The organizations that succeed will be the ones that make the transition between them almost invisible. Customers won't have to repeat themselves. Agents won't enter complicated conversations without context. And AI won't disappear when the human joins the interaction. It will shift into the background, helping that agent deliver the accurate, confident, empathetic experience the customer needed in the first place.
As AI gets better at handling the easy conversations, the human ones will matter even more.
And that means the handoff may become one of the most important moments in customer service.