In Verizon CEO Dan Schulman's view, AI will take over “a large percentage” of customer service roles. 

Disruption to the workforce, he said last week, was a near given.

AI is rapidly changing the customer service workforce, but it isn’t eliminating customer service representatives completely. Humans will still be needed, but their roles will change.

Forrester predicts that AI will halve the customer service workforce by 2030, mostly impacting representatives handling customers’ easy-to-answer questions. The remaining roles will become more technical, requiring both soft skills and the ability to work with and train AI. 

“There will be a decent number of customer-facing roles that continue,” Max Ball, principal analyst at Forrester, told CX Dive. “Some of them will be brutally hard, some of them will be more relationship building, friendly and fun interactions, and then there's going to be a whole lot of AI-focused work that's going to come along.”

CX Dive spoke to Ball about how representatives' jobs will change as AI spreads throughout the contact center. 

Frontline reps will become ‘bot unblockers’

Many customer service representatives, particularly those currently serving customers with easy questions, will take on the “human in the loop” role. 

Sometimes, these roles are referred to as “a bot unblocker,” according to Ball.

Instead of talking to customers themselves, these representatives will manage teams of AI agents. When the AI agent runs into an issue that requires a human judgment call, the human in the loop will unblock the AI and provide feedback to optimize its response in the future. 

“That is a job that I think a fair number of frontline reps will transition into,” Ball said. 

While very few companies currently have bot unblockers at the moment, it's only a matter of time, according to Ball. 

“When we look at it, we see it as something that grows over a couple of years and then starts to shrink because, eventually, the AI learns enough,” Ball said. “It's not going to go to zero, but it's gonna go down pretty significantly once it gets enough signal that it knows the answers anyway.”

It may also attract a different cohort of applicants as it takes away the main component of the current role: talking to people. 

“You'll hear people talk about, ‘Oh, it's perfect, it's such a similar skill set, and people will love doing that context,’” Ball said. “One thing I know about most contact center agents is the thing they like is they get to talk to people all day, and yeah, some of them are yelling at them, but other ones they have fun conversations with. This would take that completely out of the equation.”

Higher-tiered representatives will become “judges” and “experts” 

Representatives that handle Tier 2 and Tier 3 inquiries will specialize. While some will take on more technical subject matters, others will take on policy-expert roles that address issues that are too nuanced for AI to handle. 

Many of these roles may be “brutally hard,” according to Ball, as they take only complex and sensitive matters. Instead of occasionally helping a customer out with an easy question like a lost password, these representatives may constantly face difficult cases and frustrated customers.

Ball provided an example. When trying to transfer money from one bank to another, the transferring bank confirmed that his other bank had received the money. When he went to pay his taxes from the bank account with the recent deposit, the money still wasn't there, and he was charged a fine. 

“There's no bot in the world that's going to satisfy me at that point,” Ball said. “

Some representatives will be charged with adding value

As AI handles more customer service inquiries, some businesses will look to turn the contact center from a cost center into a revenue stream. Select customer service talent will be charged with adding value and driving revenue. 

“There is so much buzz about upsell in the contact center, some of it is legitimate,” Ball said. “If I give you a really good experience, if I am proactive and help you through something before you even knew it was a problem, that's nice. That's good. I'm gonna feel pretty warm and fuzzy towards you, so I do think there's something real there.”

Some businesses have already started doing this. 

Forrester cites Glossier and Thirdlove, two retail brands with service teams that “function more like stylists or fit consultants, blending sales and service into a single, relationship-driven interaction.”