
...

By Ravi Kumar S
CEO, Cognizant
0
Share
Newsweek is a Trust Project member
See more of our trusted coverage when you search.Prefer Newsweek on Googleto see more of our trusted coverage when you search.
For decades, IT services companies were built on the simple production function of human effort, delivered through billable hours and the pyramid structure. The real craft of the industry extended beyond labor arbitrage to the ability to size, structure, and reliably deliver mission-critical work at scale. Through specialized expertise and predictable execution, service providers built institutional confidence for clients to outsource parts of their value chain. The ability to translate human expertise and effort into consistent business outcomes became the defining moat of the industry.
The production model is now evolving with the diffusion of AI across enterprise operations and delivery. The services provider for the AI era will be built on an integrated system of human and digital labor. As model interactions become more embedded into workflows, tokens emerge as the new production input reshaping the foundation of the services model.
As access to model intelligence scales, differentiation will shift from who has intelligence to how effectively it is deployed. The asymmetry will be created by the ability to integrate human and digital labor, and drive production value for businesses in a predictable, reliable and cost-efficient way. In a system where outputs are inherently contextual and variable, consistency of outcomes becomes the advantage.
Read More on Opinion
Rising demand will place pressure on both cost and capacity, forcing more deliberate allocation of tokenized intelligence to create real enterprise value. If token consumption continues to be treated as the primary metric, costs will scale linearly with demand without a corresponding return in business outcomes.
The accelerated push for adoption and usage continues to widen the AI velocity gap between investment and value realization. For example, Uber shared that it exhausted its FY2026 budget within four months due to surging token usage from tools like Claude Code. As supply-and-demand constraints emerge, tokens will increasingly flow toward use cases that demonstrate clear value, forcing a shift from measuring usage to measuring outcomes.
Tokens are an input into delivery rather than a measure of value and cannot become the equivalent of billable hours. While many companies race to optimize access to model intelligence, the larger opportunity is to systematically reduce the unit cost of delivering outcomes. This requires building compounding systems where human expertise and token usage reinforce each other over time, improving reliability, efficiency and quality of outputs.
Every token interaction leaves a digital trace, which can be captured, contextualized and integrated into a broader knowledge fabric for the organization. Over time, this creates a compounding system of context graphs, methodologies and execution patterns that improve outcome predictability. Learning accumulates through people, teams and client engagements, turning human mobility into structured, reusable patterns of delivery. Digital labor introduces an institutional knowledge layer that reinforces and refines these patterns across contexts.
This system shifts the focus toward how tokens are best deployed for specific tasks and outcomes, developing a craft around their effective use. This enables a transition from input-based pricing to outcome-based pricing, where accountability extends across both human and machine labor.
The IT services companies that can manage and optimize the flow of tokens through guardrails, context and accumulated experience will achieve higher throughput and greater predictability. More importantly, they will be able to deliver outcomes better, faster and cheaper than their clients can internally. As experience is continuously captured into the institutional fabric, the advantage compounds and becomes difficult to replicate.
Tokenization introduces a new form of traceability that can be harnessed to build institutional muscle at scale. As tokens flow through these optimized systems, they increase stickiness and reinforce learning loops. This ultimately underpins the confidence to underwrite outcomes, moving from selling effort to guaranteeing results. The moat for the future of IT services lies in integrating human and machine labor into a compounding system that reliably, consistently and predictably delivers business outcomes.
Ravi Kumar S is the chief executive officer of Cognizant leading 350,000 associates and $21 billion in annual revenues, partnering with Global 2000 Companies in their AI journeys and reimagining enterprise and workforce landscapes.
View 0 Comments
Submit Correction
View Editorial & AI Guidelines