AI is changing what companies pay for in outsourced services, while India’s technology firms are finding new revenue in the same tools reshaping their business. The important question is not just where work goes, but who captures the productivity gains.
Watch Desk analysis
What happened
An analysis by Burak Oktenli in Eurasia Review points to a shift from billing for hours towards contracts tied to results. It cites Tata Consultancy Services saying about 80 per cent of contracts in its finance, HR and related business-services segment are outcome-based. That is not a figure for all TCS contracts, but it illustrates the changing model.
The article also notes that TCS reported annualised AI revenue above $3.1 billion, while Infosys said AI-first services accounted for 10.5 per cent of revenue and had grown 29 per cent sequentially. These figures suggest Indian providers are selling AI services even as clients press for more productivity and fewer billable hours.
The analysis distinguishes bringing work in-house from bringing it back to a company’s home country. It describes multinationals expanding their own capability centres in India, as well as Best Buy agreeing to transfer its India centre to TCS. In either case, work can remain in India while its ownership or supplier changes.
Why it matters
AI may reduce the number of human hours needed for some tasks, but that does not automatically favour domestic teams over offshore ones. The analysis cites a study of 5,172 customer-support agents in which AI assistance increased successfully resolved issues per hour by 15 per cent on average. That result came from a particular company and occupation, not every service job, but it undercuts the tidy claim that AI inherently returns work to the West.
The consequences reach beyond company accounts. The article cites an International Labour Organization estimate that 12.7 million jobs in the Philippines have some potential exposure to generative AI, while stressing that exposure is not the same as displacement. Whether workers are assisted, displaced or moved into new roles depends on how employers and governments manage the change.
Our read
“AI will bring the work home” is a better slogan than forecast. The more useful test is total cost per completed task, including errors, rework, integration and the expertise needed to keep a workflow running. That accounting could favour a local team, an offshore provider or a company’s own overseas centre. The address on the invoice does not settle the case.
What to watch
- Whether providers and clients disclose results, quality and rework alongside productivity claims.
- How outcome-based contracts change hiring and entry-level opportunities.
- Whether service-exporting economies build expertise in integration and specialist work, rather than relying on lower labour costs alone.
Discussion spark: When AI makes a service task cheaper, should the productivity gains go mainly to clients, workers, or the companies that build and run the tools?
Sources and evidence
- Source update (2 November 2026, 01:53 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.