“Now the customer doesn’t know where I am located,” Kumar said. “If it makes the caller happy, it makes me happy, too.”

His experience offers a glimpse into how generative AI – turbocharged by the release of ChatGPT – is reshaping India’s massive service industry and is a data point for optimists who believe the technology will complement, rather than replace, human workers.

Sharath Narayana, co-founder of Sanas, the Palo Alto, California-based start-up that built the tool, said AI has actually helped create thousands of new jobs in India, which was overtaken by the Philippines as the world’s largest hub for call centres more than a decade ago, due in part to accent concerns.

“We don’t see AI as taking jobs away,” said MV Prasanth, the chief operating officer for Teleperformance in India. “We see it as easier tasks being moved into self-serve,” allowing Kumar and his colleagues to focus on “more complex tasks”.

Those who use the software are engaging in “digital whitewashing”, critics say, which helps explain why the industry prefers the term “accent translation” over “accent neutralisation”. But companies say it’s delivering results: happier customers, satisfied agents, faster calls.

Many are not convinced. Whatever short-term gains automation may offer to workers, they say, it will ultimately eliminate far more jobs than it creates.

They point to the quality assurance process: when callers hear, “this call may be monitored”, that now usually refers to an AI system, not a human.

Teleperformance says such systems now review all calls for compliance and tone – tasks that workers could previously perform for only a small fraction of calls.

“AI is going to crush entry-level white-collar hiring over the next 24 to 36 months,” said Mark Serdar, who has spent his career helping Fortune 500 companies expand their global workforce. “And it’s happening faster than most people realise.”

‘A rapid wave’

To understand how AI will reshape the future of work, there are few better places to start than India’s US$280 billion ($470b) business process outsourcing, or BPO, sector.

“We will always be at the frontier of this question,” said Brian Johnson, chief executive of the outsourcing firm Transcom. “And there is no match to India for its scale, the youth, the intelligence, the hunger for work.”

The stakes for India couldn’t be higher.

It is the world’s largest destination for offshoring, which has remained among the few drivers of job creation at a time of deepening unemployment.

An estimated three million Indians provide customer support, software development, accounting, marketing and other back-office operations for the likes of Verizon, JPMorgan, Microsoft, Walmart, HSBC and Shell.

Some Indian companies are already bracing for a crisis.

“Areas like BPO and coding are in trouble and will get replaced by generative AI,” said Ajai Chowdhry, co-founder of one of the country’s largest IT consulting firms, Hindustan Computers Limited.

“India needs to altogether shift to designing and making products instead of services,” he said.

Already, chatbots, or “virtual agents”, are handling basic tasks like password resets or balance updates. AI systems are writing code, translating emails, onboarding patients, and analysing applications for credit cards, mortgages and insurance.

The human jobs are changing, too.

AI “co-pilots” are providing call centre agents with instant answers and suggested scripts. At some companies, bots have started handling the calls.

There is no shortage of ominous predictions about the implications for India’s labour force.

Within a year, there will only be a “minimal” need for call centres, K Krithivasan, chief executive of Indian IT company Tata Consultancy Services, recently told the Financial Times.

The Brookings Institution found 86% of customer service tasks have “high automation potential”. More than a quarter of jobs in India have “high exposure” to AI, the International Monetary Fund has warned.

“There is a rapid wave coming,” said Pratyush Kumar, co-founder of Sarvam, a leading Indian AI firm, which recently helped a major insurance provider make 40 million automated phone calls informing enrolees that their insurance programme was expiring. He said corporate clients are all asking him to help reduce headcount.

“You can’t hide from a trend like this,” Kumar said.

The tremors are also being felt at Indian universities, which produce more than a million engineering graduates each year. Placement rates are falling at leading IT firms; salaries have stagnated.

“The entire hope for four years of college that ‘I will get a job in the IT industry’ – that’s dashed, at least for the short term,” said Kris Gopalakrishnan, a co-founder of Infosys. “That story, in India, is life-changing.”

The human edge

While AI may be phasing out certain jobs, its defenders say it is also creating different kinds of opportunities.

Teleperformance, with hundreds of other companies, has hired thousands of data annotators in India – many of them women in small towns and rural areas – to label training images and videos for AI systems.

Prompt engineers, data scientists, AI trainers and speech scientists are all newly in demand.

“People think AI just eliminates the need for people,” said Nick Richards, head of AI at CXponent, a customer service advisory firm. “It happens in pockets, but as an outlier. What we are seeing is that it is adding value to individual people.”

Veteran BPO leader Raman Roy believes India has a formidable buffer against AI: its back-office work is increasingly complex, making it more immune to automation.

“The number of people required for the same number of calls may only marginally reduce due to AI-assisted calls,” he said.

At some firms, those who previously worked in quality assurance have transitioned to performance coaching, said Narayana, whose previous firm, Observe.ai, also built QA software. Still, he admits, 10 to 20% of workers he observed “could not upskill at all” and were probably let go.

Even the most hopeful admit that workers who can’t adapt will fall behind. “It’s like the industrial revolution,” said Prithvijit Roy, Accenture’s former lead for its Global AI Hub. “Some will suffer.”

Industry experts say “re-skilling” must emphasise not just AI training but also human connection – helping workers get better at building trust, communicating clearly and showing empathy.

“AI might stunt the growth rates of jobs that are manual and predictable, but it will not eliminate human-to-human interaction on the whole,” Narayana said.

More complex, revenue-driven interactions – like sales calls, which require relationship building – could increasingly shift to India if those skills are further strengthened, said Richards.

“A customer may still want to talk to somebody and get some reassurance,” said Prasanth. “That human emotion is not going to go away.”

Back at the company’s call centre in Gurgaon, that sentiment was echoed by Sagar Rana, a 23-year-old customer service agent. “AI can never have the human touch,” he said.

And if AI keeps advancing? “I will evolve accordingly,” Rana said. “It’s human nature to evolve.”