Cognizant Is Trying to Name the Jobs That Come After AI Disruption

On Sept. 7, Cognizant announced that it plans to hire 1,500 U.S. college graduates in 2026 while scaling two relatively new professional job families, called Frontier Certified Engineers and Frontier Business Operators, to a combined workforce of 15,000 people. The announcement folds together university recruiting, apprenticeships, employee training and partnerships with major artificial-intelligence platforms, but its larger proposition is straightforward: AI should change the structure of work without making human workers incidental to it. Cognizant chief executive Ravi Kumar S said the company believes artificial intelligence will create more jobs than it displaces and will move more value, wages and accountability toward workers who know how to use the technology.

That is a corporate position, not a settled economic fact, and the distinction matters. The future of AI and employment remains uncertain enough that the U.S. Bureau of Labor Statistics now explicitly incorporates artificial-intelligence exposure into its employment projections while warning that the effects of emerging technologies are difficult to estimate. The International Labour Organization has reached a similarly measured conclusion: generative AI is likely to transform far more jobs than it simply replaces because most occupations contain a mixture of tasks, many of which still require human judgment, interaction or accountability. In other words, the evidence increasingly suggests that the useful question is not only whether a job survives, but what remains inside the job after automation removes some of its traditional tasks.

Cognizant's new titles are an attempt to answer that second question before the labor market has developed common language for it. The company first announced the Frontier Certified Engineer and Frontier Business Operator categories in June, describing them as roles created from scratch for an AI-enabled business environment. Frontier Certified Engineers are meant to examine business processes and redesign them around artificial-intelligence systems, identifying where automation, AI agents and other tools can remove friction or change how work is performed. Frontier Business Operators are then supposed to own the day-to-day results of those redesigned systems, managing a blended workforce that Cognizant describes as a combination of human labor, digital labor, automation and judgment.

The titles sound futuristic because they are supposed to. 'Frontier' is doing branding work as well as organizational work, signaling that these employees are not simply software engineers with an updated certification or operations managers with access to a chatbot. Cognizant is attempting to establish a distinction between people who build or redesign AI-enabled systems and people who are accountable for running the business processes those systems produce. Whether the titles ultimately spread beyond Cognizant is almost beside the point. The important thing is that a major employer has reached the stage where adopting AI is no longer only a software-purchasing decision; it is becoming an organizational-design problem that requires new names, training paths and lines of responsibility.

That naming process has happened repeatedly during periods of technological change. Entire occupations become ordinary only after businesses decide that a new category of work has become durable enough to deserve a title. 'Webmaster,' 'social media manager,' 'cloud architect' and 'data scientist' each sounded unusually specific before the activities behind them became normal parts of the economy. Some titles survive, some disappear, and others are absorbed into broader occupations, but naming the role is one way an organization signals that a temporary experiment has become something it expects to budget, hire and promote around.

Cognizant is making that signal unusually explicit. In its July earnings materials, the company said it intends to build a workforce of 5,000 Frontier Certified Engineers and 10,000 Frontier Business Operators. The Sept. 7 announcement describes those as real, funded positions and says university hiring will be built into the model. The company is also training employees through its SkillSpring platform and says its workers hold more than 15,000 certifications for Anthropic's Claude, 5,000 OpenAI Codex certifications, and additional partnerships around Google's Gemini Enterprise platform. Those figures come from Cognizant and should be understood as part of the company's own account of its workforce strategy, but they show the scale at which it wants AI competence to move from specialist knowledge into ordinary professional expectations.

The practical example Cognizant chose to highlight is revealing because it is not a humanoid robot replacing a worker or an algorithm eliminating an entire department. According to the company, a combined pod of Frontier Certified Engineers and Frontier Business Operators redesigned an account-management workflow for a large food-service company into 17 production AI agents, reclaiming roughly 11 hours per account manager each week. Cognizant does not identify the client in the release, and the performance claim is not independently verified there, but the example illustrates the workforce model the company is selling. The account manager remains. A portion of the manager's work does not.

That distinction is where the labor story becomes more complicated than the familiar argument over whether AI is good or bad for employment. If eleven hours of repetitive or administrative work disappear from a forty-hour week, the outcome depends on what happens to those eleven hours. They could become time for deeper client relationships, strategic analysis, problem solving or higher-value work. They could also become a justification for larger account loads, fewer employees, tighter performance expectations or a quiet reduction in entry-level opportunities. Productivity gains do not determine their own distribution. Management does.

The ILO made a related point in a 2026 review of empirical research on generative AI at work. The organization found that productivity improvements are real but uneven, while broad job displacement has so far remained limited. At the same time, it warned that younger workers and the organization of work deserve particular attention because the tasks most easily automated are often the same tasks through which less-experienced employees traditionally learn a profession. Removing low-level work can make a job more interesting for an experienced worker while also removing the ladder that allowed the next worker to become experienced in the first place.

That makes Cognizant's plan to hire 1,500 U.S. college graduates especially relevant. The company had already announced in July that it expected to reach that figure by the end of 2026, drawing graduates from its core technology-services business, its Belcan engineering subsidiary and its Frontier talent programs. Cognizant says it recruits across more than 40 U.S. universities and maintains deeper partnerships with institutions including the University of Georgia, Arizona State University and the University of Kentucky. The Sept. 7 announcement packages those hires into a broader argument that the entry-level workforce should be trained directly into AI-era work rather than treated as the first layer to be automated away.

That is a significant bet because entry-level work has always done more than produce entry-level output. Junior employees perform routine tasks, but they also learn the systems, language, politics and tacit knowledge of an organization while doing them. If AI absorbs much of that task layer, employers will have to decide whether to reinvest the resulting efficiency into training people more deliberately or simply assume that future workers will somehow arrive with experience they were never given a chance to acquire. Cognizant's Frontier model is, in part, an attempt to formalize that missing apprenticeship by training workers to supervise, validate and orchestrate AI systems from the beginning of their careers.

The company has some self-interest in that conclusion. Cognizant sells technology and professional services to enterprises trying to implement AI, so a world in which companies need thousands of trained people to redesign processes, govern AI agents and integrate human and machine work is also a world in which Cognizant has more to sell. Its estimate that AI could unlock $4.5 trillion in U.S. labor productivity and generate an additional $1 trillion in economic value over the coming decade comes from Cognizant research, and the company frames the gap between technological capability and realized business value as a workforce problem it is positioned to help solve. That does not invalidate the argument, but it does mean the argument arrives with a business model attached.

Independent labor projections provide a useful counterweight to the promotional language without producing a simple contradiction. In July, the Bureau of Labor Statistics projected strong growth through 2034 in several occupations closely connected to AI and data work, including data scientists, information security analysts, operations research analysts and computer and information research scientists. The agency also expects AI to reduce demand in some occupations or change the tasks performed inside them. Its most recent guidance emphasizes uncertainty rather than declaring either mass unemployment or universal job creation. That uncertainty is precisely why Cognizant's experiment is worth watching: the company is making organizational choices now in territory where the final occupational map does not yet exist.

There is also something revealing about the fact that this experiment is being announced from Teaneck. Cognizant's principal executive offices and global headquarters sit in Bergen County, even as the company operates across a global workforce and serves some of the world's largest enterprises. New Jersey is therefore not merely the dateline on a national technology press release. A company headquartered in the state is attempting to design and name a workforce model that it hopes will be deployed across industries far beyond it.

The story will not be settled by whether 'Frontier Certified Engineer' becomes a familiar phrase. Corporate history is full of titles that sounded essential for a few years and then vanished into the next reorganization. The more consequential question is whether the responsibilities inside those titles become ordinary. Someone will have to decide which business tasks should be handed to AI agents, someone will have to test whether those systems are producing reliable work, someone will have to intervene when automated workflows fail, and someone will still have to be accountable when the outcome matters. Those responsibilities exist regardless of what the people performing them are eventually called.

For the last several years, artificial intelligence has been discussed as though the future of work would arrive as a verdict: jobs lost or jobs saved, humans replaced or humans protected. Cognizant's announcement suggests the transition may look much messier and much more familiar. Companies will redesign workflows, carve old occupations into new combinations of tasks, invent titles, build training programs, discover which parts work, and eventually normalize whatever survives. The future of work may not begin when an AI system takes someone's job. It may begin when the organization chart changes because nobody can quite use the old job description anymore.

That is what makes the Teaneck announcement more than a hiring release. Cognizant is not simply adding graduates to a payroll or promoting another AI partnership. It is trying to put names around a category of labor that is still being invented in real time.

We have spent years asking what jobs artificial intelligence will eliminate. One of the more useful things to watch now is who gets to name the jobs that come next.

SOURCE NOTES

• Cognizant — Sept. 7, 2026: "Cognizant Invests in America's AI-Era Workforce"
• Cognizant — June 1, 2026: Frontier Certified Engineer and Frontier Business Operator announcement
• Cognizant — July 15, 2026: U.S. college graduate hiring announcement
• U.S. Bureau of Labor Statistics — Artificial Intelligence impacts on employment projections
• U.S. Bureau of Labor Statistics — AI, information technology and employment, 2024–34
• International Labour Organization — Generative AI and Jobs: A Refined Global Index of Occupational Exposure
• International Labour Organization — The impact of GenAI on jobs, productivity and work organization
• Cognizant Careers — U.S. locations / Teaneck global headquarters