Sam Altman said customer support would be "totally gone." So why is the Philippines' outsourcing sector still growing?
For the past two years, a number of boardrooms have worked from a simple assumption: generative AI would do to call centres and back-office outsourcing what container ships did to manufacturing, and do it fast. That assumption traces back to one widely repeated line. In July 2025, OpenAI chief executive Sam Altman told a US Federal Reserve audience that some job categories were "just like totally, totally gone," singling out customer support as the clearest example.
Eleven months later, Altman was telling a different story. Speaking to audiences across Asia in May 2026, he said he was "delighted to be wrong" about how fast and how completely AI would displace jobs. For any HR leader planning workforce strategy around outsourced or offshore teams, that reversal is worth remembering. The loudest predictions about AI and jobs tend to age quickly, and not always in the direction people expect.
What the global data shows
The Philippines remains the world's largest hub for voice-based outsourcing, and its industry body's own figures are a reasonable proxy for the sector's health. The IT and Business Process Association of the Philippines says the industry closed 2025 with roughly US$40 billion in export revenue and about 1.9 million workers, and is forecasting US$42.3 billion in revenue and 1.96 million workers by the end of 2026. That's growth, not collapse, and it matters since the Philippines is a primary outsourcing destination for many businesses.
The longer-range outlook tells a different story, though. The industry's original 2028 roadmap, published in 2022, called for US$59 billion in revenue and 2.5 million jobs. It now expects somewhere between US$43.3 billion and US$50.5 billion, and between 1.85 million and 2.14 million jobs, a genuine cut to the growth trajectory even if the near term still looks solid.
IBPAP president and chief executive Jack Madrid attributed the revision to AI adoption, shifting buyer behaviour and rising global competition. Speaking to reporters at a briefing in July, he put it plainly: "We need to review where we are and be honest about what we can achieve realistically, and those are the numbers today."
Anyone following broader coverage of how AI is reshaping jobs rather than simply eliminating them will recognise the pattern. It also aligns with independent research from analyst firm Cavell, which found demand for human contact-centre agents globally is still projected to grow, from 15.3 million in 2025 to 16.8 million by 2029. That headline number hides something worth flagging: Cavell's own modelling suggests automation will suppress roughly 1.9 million additional agent roles that would otherwise have been created over the same period. Growth and displacement are happening at the same time.
Why the economics aren't as simple as they sound
Gartner published one of the more useful correctives for HR and procurement teams weighing an AI-versus-offshore business case back in January. The firm predicts that by 2030, the cost per resolution for generative AI in customer service will exceed US$3, more than many business-to-consumer offshore human agents cost today.
"Customer service leaders are determined to use AI to reduce costs, but return on those investments is far from guaranteed," said Patrick Quinlan, a senior director analyst at Gartner. "Full automation will be prohibitively expensive for most organisations; instead, leading organisations will use AI to drive customer engagement rather than to cut costs."
The pressure to automate isn't easing because of that, though. A separate Gartner survey of 321 customer service leaders found 91% report pressure from their own executives to implement AI in 2026, up sharply on prior years. HR leaders are stuck between that executive pressure and a cost picture that's more complicated than most business cases assume.
The part that's actually disappearing
Offshoring versus onshoring isn't really the question. What's eroding is the entry-level rung specifically. Basic data entry, routine account queries and simple document processing (the "tier one" work that has traditionally been how school leavers and career-changers got a foot in the door) is the part of the job most exposed to automation. There's already reporting on this pattern showing up in hiring data for younger workers more broadly, and MIT-linked research has estimated roughly 11.7% of total US wage value is already technically automatable with current AI systems, concentrated in exactly these administrative and customer-facing functions.
If the entry rung disappears faster than new junior roles appear, organisations lose a pipeline they have relied on for years, whether that talent sits on their own payroll or a vendor's. Outsourced headcount also often sits outside formal consultation and redundancy processes.
A regulatory wildcard worth watching from a distance
The most concrete regulatory threat to offshoring so far has emerged in the United States, but it's relevant to any organisation with US-facing operations or US clients. The bipartisan Keep Call Centers in America Act and a companion House bill, the HIRE Act, would require employers to notify the US Department of Labor before moving call-centre work offshore, publish a public register of companies that do, and strip federal grants and contracts from those on it.
Both bills were introduced in mid-2025 and remain stuck in committee, with no vote recorded as of mid-2026. Neither is close to becoming law, but the debate tends to travel: onshoring pressure in the US has previously prompted similar political conversations in Australia and Canada, even without matching legislation.
Where this leaves HR
The outsourcing industry isn't being shredded. It's being sorted, increasingly by task complexity rather than geography. Simple, scriptable work is genuinely at risk wherever it sits. More complex work, judgement calls, exception handling, anything that requires empathy, is proving stickier than the 2025-era predictions assumed, and in some cases it's moving to offshore hubs precisely because AI tools let less-experienced staff there handle it competently.
For HR leaders managing outsourced or offshore relationships, the useful questions aren't "should we reshore?" or "should we automate?" They're smaller and more concrete: which specific tasks in our outsourced contracts are entry-level and script-based, what happens to the people doing them, and does our vendor governance account for AI-driven headcount changes the way it would account for a redundancy on our own books? Recruitment process outsourcing providers are already rethinking their value proposition around exactly this question. The rest of the outsourcing industry is likely to follow.
