Will AI mean the end of call centres?

Ask ChatGPT if AI will replace humans in the customer service industry and he will offer you a diplomatic answer; The summary of this is “they will work side by side”.

But people are not so optimistic.

Last year, K Krithivasan, chief executive of Indian technology firm Tata Consultancy Services, told the Financial Times that artificial intelligence could soon mean: “Minimal need” for call centers in Asia.

Meanwhile, AI will autonomously solve 80% of common customer service issues Business and technology research firm Gartner predicts it by 2029.

There’s a lot of hype around “AI agents” right now. This is the term given to artificial intelligence systems that can operate and make decisions more autonomously.

They can speed up existing non-AI chatbots, known as “rule-based chatbots,” that can only respond to a specific list of questions.

My recent experience with package delivery company Evri’s chatbot illustrates the current state of affairs without AI.

My package hadn’t arrived and Ezra (the chatbot’s name) offered to “get this sorted right away.”

He asked for a tracking reference and said my package had been delivered after I wrote it.

I was able to request proof of delivery and when I did I was shown a photo of the package at the wrong front door. And once this “evidence” was shown there was no option to move the conversation forward.

In response, Evri told the BBC it would invest £57 million to further develop the service.

“Our smart chat facility uses tracking data to suggest the most helpful responses and ensure the customer’s package is delivered as quickly as possible if it hasn’t happened as planned,” he says.

“Our data confirms that the vast majority of people get the answers they need from our chat facility for the first time, within seconds. We’re always reviewing feedback and continuing to make improvements to ensure our services are as helpful as possible.”

Meanwhile, rival parcel delivery company DPD was forced to disable its less rule-abiding AI chatbot after it criticized the company and cursed at users.

Striking the balance between staying true to the brand and truly helping customers is a difficult task for businesses to grapple with when switching to AI.

Some 85% of customer service leaders According to Gartner, they are researching, piloting or deploying AI chatbots. But it also turns out that only 20% of such projects fully meet expectations.

“You can have a much more natural conversation with AI,” says Garner analyst Emily Potosky.

“But the downside to this is that the chatbot can hallucinate, give you outdated information, or tell you things that are completely wrong. I would say rules-based agents for package delivery are great because there are so many permutations of questions about someone’s package.”

Resources and money are among the key reasons why businesses are considering switching from human to AI-based customer service. But Ms. Potosky notes that it’s not clear that AI will be cheaper than human agents.

“This is a very expensive technology,” he says.

The first thing any business looking to replace humans with AI needs to do is make sure they have comprehensive training data.

“There is this idea that knowledge management is becoming less important because generative AI can work around the fact that its knowledge is not particularly well organized, but the opposite is the case,” Ms. Potosky said. he adds.

“Knowledge management is more important when deploying generative AI.”

Joe Inzerillo, chief digital officer at software giant Salesforce, told the BBC that call centers provide fertile training grounds for AIs, especially those moving to low-cost regions such as the Philippines and India.

This is because a lot of staff training has been done so that artificial intelligence can also learn.

“You have a lot of documentation, and these are all great things for AI to have when it’s going to take over the first line of defense,” he says.

Salesforce’s AI-powered customer service platform, AgentForce, is currently used by a wide range of clients, from Formula 1 to insurance company Prudential, from restaurant booking website Open Table to social media site Reddit.

Mr. Inzerillo says Salesforce learned some valuable lessons about how to make AI more human-like when it first tested the platform.

“A person might say, ‘I’m sorry to hear that,’ but the agent just opened a ticket,” Mr. Inzerillo says.

So the AI ​​is trained to be more sympathetic, especially when the customer has a problem.

Salesforce also found that not allowing the rep to talk about competitors was problematic.

“This backfired when customers asked legitimate questions about Microsoft Teams integration with Salesforce,” Mr. Inzerillo says. “The representative refused to help because Microsoft appeared on our competitor list.”

The company later changed this strict rule.

Salesforce has ambitious plans to roll out AI agents, and so far it claims they’re gaining traction with its customers. It is also stated that the majority of customers (94 percent) prefer to interact with AI representatives when given the option.

“We found that customer satisfaction rates are much higher than what we get with humans, then AI can unlock the next level of customer service,” Mr. Inzerillo says.

It also meant the firm cut customer service costs by $100 million, but he was willing to downplay recent headlines suggesting it had led to 4,000 job cuts.

“A very large percentage of those people were redeployed to other areas around customer service.”

Fiona Coleman runs QStory, a firm that uses artificial intelligence to offer human call center workers more flexibility in their shift patterns. Its clients include eBay and NatWest.

While he sees the value of AI in improving working conditions, he’s not sure the technology will ever fully replace humans.

“There are times when I don’t want to have a digital interaction and I want to talk to a human,” he says.

“Let’s see what it looks like in five years; can an AI apply for a mortgage or talk through a debt problem? Let’s see if the AI ​​is empathetic enough.”

The use of AI in customer service may actually already be facing a backlash.

Legislation currently being proposed in the US to move offshore call centers back to America also requires businesses to disclose their use of AI and transfer the caller to a human if desired.

Meanwhile, Gartner predicts that the EU will can represent What’s called the ‘right to talk to people’ as part of consumer protection rules.