Sourcefit, a global outsourcing company, surveyed close to 2,000 employees in client-managed teams across its Philippines, South Africa, the Dominican Republic, and Madagascar offices between February and July 2026. Fewer than 7% said they are concerned AI will hurt their role over time. Everyone else lands far more confident: most feel cautiously optimistic, many expect AI to make their jobs better outright, and a smaller group have simply moved on.
They work as accountants, software engineers, healthcare administrators, customer service representatives, data analysts and more, spanning more than 50 roles and 20 industries, embedded daily inside client operations. The respondents answered eight questions about their AI use, trust, and where it falls short.
The adoption numbers explain some of that confidence. 52% use AI tools daily or several times a week, and another 22% use them occasionally, putting total adoption at 74%. The top use case is writing and documentation at 37%, often for employees working in a second language, followed by research and analysis at 30% and customer response support at 14%.
68% report a moderate or significant improvement in their own productivity and output quality, while only half a percent report a negative impact. These numbers are self-reported, but the same direction shows up independently across almost 2,000 people in four countries and more than 50 roles.
The data breakdown by country highlights some interesting variation as well. The Dominican Republic reports 70% regular AI use, 73% citing productivity gains, and zero respondents concerned about their role. South Africa shows adoption similar to the Philippines at 52%, with a lower reported productivity gain, 59% against 69%. Adoption and impact do not move together everywhere in this data.
Asked what limits their ability to use AI more effectively, 30% named limited access to tools, the largest single response, followed by data security and compliance concerns at 20%, then unclear policy and lack of training. Concern about AI reducing or eliminating parts of their role came in at 7%, consistent with the topline number. Of the remainder, 28% expect AI to mainly enhance their work, 37% describe themselves as cautiously optimistic, and 17% have not thought much about it yet.
This data lands amid shaky market sentiment around outsourcing, particularly within its customer experience and call center segment. Concentrix and Teleperformance, two of the largest providers in that segment, both saw sharp share price declines this year after Concentrix cut its revenue guidance and pointed to clients pulling back on customer support spending. Analysts at RBC Capital Markets and Bloomberg Intelligence both pointed to AI as a factor, with Bloomberg Intelligence noting that AI may be reducing demand for traditional customer experience outsourcing faster than newer AI-related services can offset it.
Outsourcing has built value for decades by giving companies access to specialized talent, infrastructure, and operational scale that would be costly to build internally on their own. Investors are now pricing a future where AI absorbs a slice of that work directly, without a labor layer involved at all.
That model treats the workforce as a fixed input waiting to be replaced. But the survey data says something different is already happening inside it. Adoption sits at 74%. Employees most often cite access as the barrier standing between them and using AI more, ahead of fear or resistance to the technology. The second-largest barrier, cited directly by the people doing the work, is the same regulated environment that slows full automation in healthcare records, financial data, and other compliance-bound processes, a constraint the market narrative tends to skip, but not a permanent shield against disruption.
Andy Schachtel, founder and CEO of Sourcefit, oversees the company’s operations across all four countries surveyed. His read on the data: “When your offshore team and your onshore team have access to the same AI tools, the gap in output quality narrows. AI is actually an equalizer between local and offshore staff. The offshore team becomes more competitive, not less.”
That reframes the market’s core assumption. The prevailing thesis treats AI as a substitute for outsourced labor. The data suggests it currently functions as a multiplier of it instead, provided access keeps pace with willingness to use it. Shared tool access closes the output gap that used to require extra client-side oversight, the same friction that made some companies hesitant about offshore models in the first place. When both teams work from the same baseline, offshore talent keeps its cost advantage and strengthens its case on quality too.
This is also where outsourcing providers have an opening to strengthen their position inside this narrative. As AI moves into these workflows, healthcare records, financial data, customer accounts, someone must build governance around how it is used, function by function. Providers already running these operations at this scale are the ones doing that work now, building AI oversight directly into the offshore workflow as adoption happens.
Of course, a workforce assessing its own relevance is biased. People predict their own obsolescence poorly, and confidence in one’s own value proves little on its own.
But proximity to the work also surfaces what a repeatable-task model tends to miss from the outside: judgment calls, context, the friction points that do not show up cleanly in an automation spreadsheet. The employees closest to this shift are also the ones already living inside it, using the same tools the market assumes will eventually replace them, and reporting that the work has gotten better because of it.
Sourcefit plans to keep running this survey. For now, the workforce doing this work every day is already using the tools everyone else is racing toward, and says the shift is making them more valuable to the clients they work for. That is a specific, testable claim, and it holds up better under nearly 2,000 responses than a stock chart does on its own.
More on how Sourcefit builds AI-ready global teams is available at sourcefit.com.
