Call Center Jobs Surge Despite AI, Defying Automation Predictions

Call center employment climbs as AI spreads

Call center employment in the Philippines has nearly doubled over the past decade, reaching 2 million workers in 2025 despite widespread predictions that artificial intelligence would eliminate these roles. According to data from the IT & Business Process Association of the Philippines cited by Apollo chief economist Torsten Slok, the sector has grown every year from 2016 through 2025.

Unemployment rates in major outsourcing hubs have moved in the opposite direction of what AI displacement would suggest. In the Philippines, joblessness fell from 9% in 2021 to roughly 5% by mid-2026. India saw a similar trend, with unemployment dropping from around 7% to 6% over the same period.

The pattern contradicts high-profile workforce reductions like Salesforce's elimination of 4,000 customer service roles in September 2025, when CEO Marc Benioff declared he needed "less heads." Customer service representative tasks have an 86% automation potential according to Brookings Institution estimates, making them among the most vulnerable to AI replacement.

Why it matters

The growth of offshore call center employment challenges the narrative that AI will rapidly displace white-collar work. Instead, it suggests AI may expand job markets by reducing costs and enabling companies to serve more customers across more channels—a dynamic with implications for how business leaders should plan workforce strategy in other professional sectors.

A 19th-century paradox returns

Slok attributes the phenomenon to Jevons paradox, named for English economist William Stanley Jevons's 1865 observation about coal consumption. When the Watt steam engine made coal more efficient as an energy source, consumption increased rather than decreased because energy became cheaper and more accessible.

"Lower cost per interaction does not mean fewer interactions," Slok wrote. "It means more customers served, more channels opened and more markets worth reaching."

The pattern appears in other fields once predicted to face AI-driven job losses. Geoffrey Hinton, often called the "godfather of AI," declared a decade ago that radiology would no longer require human practitioners. Instead, the number of U.S. radiologists has increased by 10% over that period, according to Christoph Herpfer, an economist at the University of Virginia's Darden School of Business, who noted a significant radiologist shortage.

Productivity gains fuel demand

Research supports the productivity-expansion hypothesis. A 2023 study led by Erik Brynjolfsson, director of the Stanford Digital Economy Lab, found that an AI conversational assistant increased productivity for more than 5,000 customer support agents by an average of 14% per hour. Earlier research by Brynjolfsson showed that AI translation features on eBay boosted international exports by 17.5%.

Emma Harrington, an economics professor at the University of Virginia, sees parallels between those findings and current offshore employment trends. "We can trade labor more easily across countries when language can be traded more seamlessly," she noted.

Some experts point to AI's current limitations as another factor sustaining human employment. Benjamin Shestakofsky, a sociologist and information science professor at Cornell University, argues that AI cannot yet navigate complex customer problems and that humans experience cognitive overload—"AI brain fry"—when handling increased case volumes. Additionally, some companies may deliberately maintain human agents as a brand differentiator in an increasingly automated world.

These details were first reported by Fortune.