Will AI Replace Humans in Customer Service? The Realities Behind the Hype
Ask ChatGPT if artificial intelligence will replace humans in the customer service industry, and it will likely give you a balanced response: “Humans and AI will work side by side.”
But among business leaders and workers, optimism is harder to find. With rapid advances in automation, many fear that AI may soon dominate an industry that employs millions worldwide.
AI’s Growing Role in Customer Service
The CEO of Indian tech giant Tata Consultancy Services (TCS), K. Krithivasan, predicted that AI could soon make traditional call centers “minimally necessary” across Asia. That sentiment is echoed by research firm Gartner, which estimates that by 2029, AI will autonomously resolve up to 80% of common customer service inquiries.
The current wave of enthusiasm centers on “AI agents” — intelligent systems capable of acting autonomously and making decisions in real time. These agents promise to outperform existing “rule-based chatbots,” which can only answer a limited set of pre-programmed questions.
While AI-powered agents promise efficiency and cost savings, recent experiences suggest they are still far from perfect.
The Limits of Current Chatbots
Consider a recent interaction with delivery firm Evri’s automated chat assistant, “Ezra.” When a customer reported a missing parcel, the chatbot confidently declared the item had already been delivered — accompanied by a photo of the wrong address.
After that, the chat offered no further options. The customer was stuck.
In response, Evri announced a £57 million investment to improve its customer service technology. “Our intelligent chat facility uses tracking data to suggest the most helpful responses,” a company spokesperson explained, adding that most users “get answers within seconds.”
But not every company has found the AI transition smooth. Rival courier DPD was forced to disable its AI chatbot after it began criticizing the company and using profanity toward customers — a cautionary tale about giving AI too much freedom.
Businesses Race to Integrate AI
Despite these challenges, AI integration in customer service is accelerating. According to Gartner, about 85% of customer service leaders are now exploring or deploying AI chatbots. However, only 20% of these projects are currently meeting expectations.
“AI allows a much more natural conversation,” says Emily Potosky, a Gartner analyst. “But the downside is that AI can hallucinate, give out-of-date information, or just get it wrong. For straightforward issues like parcel tracking, rule-based bots might actually be better.”
In other words, the race toward automation isn’t always smooth — or cheaper.
The Cost and Complexity of AI
Many assume that replacing humans with AI will reduce costs. But experts warn the opposite may be true — at least initially.
“This is a very expensive technology,” Potosky explains. “You need massive amounts of high-quality training data. Knowledge management becomes more important, not less, when deploying generative AI.”
In other words, the system is only as good as the information you feed it. Poorly organized data can lead to costly mistakes, especially when AI systems are expected to make decisions without human oversight.
Training AI on Human Experience
Joe Inzerillo, Chief Digital Officer at Salesforce, believes that traditional call centers — especially in countries like India and the Philippines — have become ideal training grounds for AI systems.
“You have a huge amount of documentation, scripts, and training material,” he says. “That’s invaluable for teaching AI how to handle the first line of customer interaction.”
Salesforce’s AgentForce platform is already used by companies such as Formula 1, Prudential, OpenTable, and Reddit. It combines automation with conversational intelligence, learning from human customer service practices.
But Salesforce’s journey wasn’t without hiccups. During early testing, the AI system failed to show empathy. “While a human might say ‘I’m sorry to hear that,’ the AI would just open a support ticket,” Inzerillo recalls. The company later retrained its models to sound more sympathetic.
Another lesson came when Salesforce banned its AI from discussing competitors. This backfired when customers asked about Microsoft Teams integration — the AI refused to help because Microsoft was on the “competitor list.” That rule was quickly revised.
AI Success Stories and Cost Savings
Despite early challenges, Salesforce claims remarkable success. The company reports that 94% of users now choose to interact with AI agents when given the option. According to Inzerillo, “Customer satisfaction rates are higher with AI than with humans in many cases. It unlocks a new level of service.”
The company also reports saving $100 million in customer service costs. However, Inzerillo emphasizes that the transition didn’t lead to widespread layoffs. “Most of those employees were redeployed into other customer experience roles,” he says.
When AI Meets Human Empathy
While AI continues to grow, some experts argue it will never fully replace human empathy — especially in emotionally charged or complex situations.
Fiona Coleman, CEO of QStory, a firm that uses AI to improve scheduling flexibility for human call center staff, believes in a hybrid future. Her company serves major clients such as eBay and NatWest, helping agents manage shifts and workloads using AI-driven data.
“There are times when I just want to speak to a human,” Coleman admits. “AI can make things faster, but when you’re dealing with something sensitive — a debt issue, a mortgage, or a bereavement — empathy can’t be automated.”
She adds that while AI will continue to advance, there’s still a long way to go before it can truly understand emotional nuance. “Let’s see what it looks like in five years — whether AI can genuinely handle conversations that require compassion.”
The Pushback: Regulation and the “Right to Talk to a Human”
The growing presence of AI in customer service is now attracting political and regulatory scrutiny.
In the United States, proposed legislation seeks to bring offshore call centers back onshore — while also requiring companies to disclose when customers are interacting with AI. The law would also grant callers the right to request a human representative at any time.
Meanwhile, Gartner predicts that by 2028, the European Union could introduce similar legislation, enshrining a “right to talk to a human” in its consumer protection framework.
These moves signal an increasing awareness that while AI can improve efficiency, it must remain transparent and accountable.
Striking the Right Balance
As AI continues to evolve, businesses face a delicate balancing act: combining efficiency and cost savings with empathy and trust.
The most successful customer service models may not be purely AI-driven, but hybrid systems — where machines handle routine tasks, freeing human agents to focus on complex or emotionally sensitive issues.
In this vision, AI isn’t a replacement but an enhancement — a digital colleague that makes human work more meaningful.
However, experts agree that success depends on thoughtful deployment, ongoing oversight, and ethical guardrails. Poorly implemented AI can frustrate customers, damage brand reputations, and create more problems than it solves.
The Future of Customer Service
The future of customer service may not be one where humans are obsolete, but one where technology enables better interactions. AI agents can manage repetitive queries instantly, while skilled human staff handle nuanced or high-stakes cases with empathy and understanding.
As Fiona Coleman puts it: “AI can make the experience smoother. But empathy — that’s still human territory.”