AI Reshapes Philippine Outsourcing as Workers Help Train Their Replacements

A writing career interrupted by the system she helped train

Lisa, a single mother with 15 years of experience in content writing after college, lost her outsourcing job eight months into the role, just one month before it was due to become permanent. Before her dismissal, she and her colleagues had been editing material generated by artificial intelligence for a public relations agency. Their knowledge helped train the system to follow the company's style.

Lisa had built her career around meeting deadlines without relying on AI. She described feeling as though she had dug her own grave: the team's work had helped train the technology that replaced them. Her experience captures a difficult turn in the Philippines' outsourcing industry, where workers can contribute to improving an AI system while their own employment becomes less secure.

As of August 5, 2026, the sector employs roughly 1.9 million people, generates about $40 billion annually and accounts for around 10 percent of the economy. Its future matters both to employees and to the businesses and cities that depend on their spending.

The International Labour Organization estimates that 12.7 million Filipinos work in roles exposed to generative AI, with the Philippines having the highest share in Southeast Asia. Generative AI produces material such as text. In Lisa's workplace, employees supplied the editing and knowledge of company style that made its output more useful.

Other former outsourcing employees have described similar experiences while remaining anonymous because of confidentiality agreements attached to their severance packages. Those accounts sit alongside industry assurances that AI will change many jobs and that most affected employees have been moved into other roles.

When a productivity tool creates more work

Mary, another former content writer, said her employer introduced AI as a way to improve productivity. Her experience was different: the tool increased the workload because employees had to edit its output and check inaccurate statements. She later faced redundancy too.

Mary's account shows what the change required from the people doing the work. Producing a draft was only part of the task. Employees still had to correct it and make sure it was accurate, even as the company moved toward greater use of automation.

Paul Quintos of the University of the Philippines Diliman identifies pressure from foreign clients as a key driver. Clients want outsourcing providers to adopt AI as a major way to reduce expenses. For workers, that pressure can mean having to use a new system, correct its mistakes and face uncertainty about whether their role will continue.

AI pilots spread while hiring slows

Jack Madrid, president of the IT and Business Process Association of the Philippines, said more than two thirds of the association's members are running AI pilots. These are trials in which companies test how the technology fits into their operations before deciding how widely to use it.

Madrid said agentic AI can automate work at an accelerated pace. Such systems can carry out tasks rather than simply produce a response for a person to use. He also said most affected workers have been reassigned within their organizations.

Slower hiring has more than one cause, according to Madrid. Reduced global demand and uncertainty about how companies will deploy technology are also holding back recruitment. The sector therefore faces pressure from both its customers' business conditions and changes in how services are delivered.

Major employers promise training and human oversight

Teleperformance, the largest call center operator, says AI will handle routine tasks while people take on more complex roles. The company plans to retrain employees as that division of work changes.

Accenture expects generative AI to reshape jobs rather than eliminate them and has pledged billions toward capabilities and training. Concentrix, the biggest outsourcing firm in the Philippines, says human oversight remains necessary and has committed to training staff.

The companies' position is that employees will continue to provide the judgment and empathy that technology requires. Their plans place retraining at the center of the transition: workers handling routine requests would need the skills to take on more demanding work.

Lisa's and Mary's experiences show the personal stakes alongside those commitments. Both worked with AI before losing their positions. For employees facing redundancy, the immediate concern is whether another role is available and whether they can obtain the training needed to move into it.

The government's response focuses on 300,000 workers

Leandro Aguirre of the Philippines' Department of Information and Communications Technology acknowledged that the country is slightly behind competitors and must catch up. The government aims to upskill more than 300,000 outsourcing workers, helping them develop the abilities needed for changing jobs.

Aguirre also called for attention to local talent and AI companies developed in the Philippines. The aim is to create jobs with greater value as the country adapts to technology that is changing the services it sells to foreign clients.

Against an outsourcing workforce of roughly 1.9 million, a target of 300,000 represents about 16 percent. The government's stated ambition exceeds that number, but current retraining programs reach far fewer workers than are needed annually.

Aguirre said existing labor, privacy and consumer protection laws can be adapted through regulatory guidance without sweeping new legislation. Under that approach, authorities would explain how current rules apply to AI instead of replacing them with an entirely new legal framework.

Workers have limited input into AI decisions

Labor groups say current employment laws cover redundancy but offer limited guidance on introducing AI or consulting workers. Redundancy rules address the loss of a position; the concern raised by these groups is how employees participate in decisions before technology changes their work.

The outsourcing sector remains nonunionized, limiting formal worker input into decisions about deploying technology. Employees who help improve a system may therefore have little structured involvement in deciding when it is introduced or how their responsibilities will change.

The International Labour Organization projects that many roles will be transformed rather than disappear. The Philippines is nevertheless particularly exposed because service jobs vulnerable to change are common in its economy.

For workers, transformation can involve learning new tasks, moving to another position or spending more time checking automated output. Lisa and Mary experienced job losses, while Madrid says reassignment has been the outcome for most affected employees. Both experiences are part of the sector's transition.

India and Southeast Asia face uneven changes

India faces parallel pressure in an outsourcing industry that employs nearly 6 million people and generates about 7 percent of gross domestic product. Tata Consultancy Services cut 12,000 jobs last year, the largest reduction in its history. Oracle eliminated 12,000 positions in India amid increased spending on AI.

World Bank analysis indicates that multinational companies are reducing hiring faster than domestic peers as generative AI reshapes the labor market. That pattern matters for an industry serving foreign clients and operating across national borders.

Across the Association of Southeast Asian Nations, Singapore leads in AI talent density, while the Philippines, Malaysia, Indonesia and Vietnam trail. Talent density describes how concentrated the relevant skills are within a workforce.

Regional projections suggest AI could add nearly $1 trillion to economic output by 2030. The gains may concentrate in economies that are already more digitally advanced, widening the divide between countries better equipped to use the technology and those still building their capabilities.

A Sourcefit survey presents a more optimistic view among offshore professionals: fewer than 7 percent fear AI will harm their roles, while 68 percent report productivity gains. Those results show substantial confidence among respondents, alongside the job losses and uncertainty experienced by other workers.

Fewer jobs affect businesses beyond the office

Quezon City, Makati and Bonifacio Global City have built infrastructure around outsourcing operations running 24 hours a day. Employees working night shifts support restaurants, gyms and other services, creating an economy around the hours and needs of the industry.

Those businesses now face contraction as employment changes. Major providers have reduced headcount by 15 to 20 percent over 18 months while maintaining service levels. A smaller workforce can therefore affect businesses outside the outsourcing companies even when customers continue receiving the same volume of service.

The cost comparison gives clients a strong reason to consider automation. An AI agent costs roughly 10 cents per interaction, compared with $8 to $12 for a human agent. Using those figures, the human interaction costs about 80 to 120 times as much.

Some firms are moving toward more complex services, including fraud analysis and enterprise relationship management, which involves managing relationships with business customers. Industry observers estimate that roughly two positions requiring greater skills emerge for every 10 customer service roles eliminated.

That estimate puts the scale of retraining in perspective. New types of work are emerging, but workers leaving routine service roles face a smaller number of those new positions and a need to develop different skills.

Employees face immediate uncertainty despite growth

Ivan Peregrina, a former quality analyst, is seeking part time work after being replaced by an AI system that audits call recordings. His former task involved reviewing calls; the system now performs that checking function.

Paul Ponce, a call center agent who paid for his children's education through business process outsourcing earnings, worries that chat support roles will disappear first. His concern reflects how closely family finances can depend on continued employment in the sector.

Forrester projects that AI could handle half of customer service by 2030. Microsoft and Commonwealth Bank of Australia have already reduced support staff, adding to the examples of changing demand for people in service roles.

The Philippines' industry also continues to add jobs. Industry data showed 80,000 jobs added last year, and forecasts predict further growth. Companies say AI supports employees rather than replaces them and that human oversight remains necessary for empathy and judgment.

Growth targets meet a changing mix of jobs

The Philippines' IT Business Process Management road map targets 2.5 million jobs by 2028. Compared with the current workforce of roughly 1.9 million, that would mean about 600,000 additional positions, an increase of approximately 32 percent. Analysts say the targets require recalibration.

The central challenge is the changing mix of work. The industry can add jobs while individual employees lose theirs, and companies can maintain services with fewer people while developing new roles that require different abilities. Government training plans and employers' retraining commitments address that shift, but current programs reach fewer workers than needed each year.

The main developments span several periods:

• Over an 18 month period, major providers reduced headcount by 15 to 20 percent while maintaining service levels.
• Last year, the Philippines' outsourcing industry added 80,000 jobs, while Tata Consultancy Services cut 12,000 positions.
• As of August 5, 2026, more than two thirds of the Philippine industry association's members are running AI pilots, and the government aims to upskill more than 300,000 workers.
• By 2028, the industry's road map targets 2.5 million jobs.
• By 2030, Forrester projects AI could handle half of customer service, while regional projections suggest AI could add nearly $1 trillion to ASEAN economic output.

For Lisa, Mary, Peregrina and Ponce, the transition is already personal. The sector's ability to keep growing will matter alongside its ability to help existing workers move into the jobs that remain and the new ones being created.

The Bottom Line

• Philippine outsourcing employs roughly 1.9 million people, generates about $40 billion annually and accounts for around 10 percent of the economy.
• The International Labour Organization estimates that 12.7 million Filipinos work in roles exposed to generative AI, with the highest share in Southeast Asia.
• Former employees describe editing or training AI systems before losing their jobs.
• More than two thirds of industry association members are testing AI; industry leaders say most affected workers have been reassigned.
• The government aims to upskill more than 300,000 outsourcing workers, while current programs reach fewer people than needed annually.
• India faces similar employment pressure, and projected regional gains may favor economies with stronger digital capabilities.
• The Philippines added 80,000 outsourcing jobs last year and targets 2.5 million by 2028, even as automation changes the work available.