There are two stories you have probably heard about entry-level work in the age of AI. One says the bottom rung is being sawed off, that the junior analyst and the first-year associate are about to vanish. The other says nothing much is really changing, that this is overblown and the org chart will look the same in five years.
A new study from Cognizant and Pearson says both stories miss the point. The entry-level job is being rebuilt right now, and the rebuild has already started inside your company whether you planned for it or not.
What the study actually found
Cognizant and Pearson announced The AI Workforce Pulse on June 18, 2026, out of Teaneck, New Jersey and London. The research was conducted by Wakefield Research, which surveyed 750 HR professionals at the director level and above, all at companies with at least 1,000 employees, across the United States, the United Kingdom, and India. Fieldwork ran from March 23 to April 3, 2026.
The headline number: 94% of HR leaders expect AI to generate new entry-level roles within the next five years, roles that did not exist before. At the same time, 85% still call entry-level positions essential in the AI era. The bottom of the org chart is being redrawn.
of HR leaders expect AI to generate new entry-level roles within the next five years, while 85% still call entry-level positions essential.
Source: Cognizant and Pearson, The AI Workforce Pulse, 2026Pay attention to this one as a leader. AI already completes roughly one in three entry-level tasks today, and 18% of HR professionals say AI handles half or more of the work in first-job roles. This is happening in your current operations right now.
From doing the task to directing the tool
The study offers an image that does the heavy lifting: the emerging entry-level worker looks "less like a data entry associate and more like an air traffic controller."
The new junior role is about directing the tool, checking its output, catching what it got wrong, and deciding what happens next.
That single line redefines what early-career work is for. The value of a junior employee used to sit in executing the task: pulling the data, drafting the first version, running the routine analysis. AI is now fast and cheap at exactly that layer. So the work moves up. The new junior role is about directing the tool, checking its output, catching what it got wrong, and deciding what happens next. Ninety-six percent of HR leaders expect entry-level roles to evolve within five years into positions where people supervise or manage AI systems.
For anyone who owns a P&L, this is the strategic read. You are not replacing your entry-level talent with software. You are changing what you hire them to do, and the change is from production to judgment.
The skills leaders now reward
If the job is moving toward supervising AI, the hiring criteria move with it. The study is direct about what HR leaders now value:
- 97% say soft skills are more important in the AI era.
- 98% emphasize AI fluency when hiring for non-technical roles, not just technical ones.
- 69% now prefer broad, interdisciplinary backgrounds over narrow specialist skillsets.
- 67% value liberal arts degrees more than they used to, rising to 72% in the UK.
Read those together and a pattern emerges. The premium is moving toward people who can think across domains, communicate, judge quality, and work alongside a tool rather than compete with it. Those are learnable proficiencies, not fixed credentials. That distinction is the whole game for a leader trying to build an AI-capable team, because it means the capability you need can be developed in the people you already have.
This is the climb that The 7 Levels of AI Proficiency framework describes. The lower levels are about awareness and basic use. The higher levels are about judgment, building, and orchestrating AI output across a workflow. The air traffic controller the study describes is a person climbing from doing the task to directing the system, which is precisely the movement the framework measures.
The shortfall every executive should see
The study also surfaces an uncomfortable disconnect, and it is the most actionable finding for a leadership team.
Demand for AI training is up. Ninety-one percent of HR professionals report that employee demand for AI training increased over the past 12 months. Yet 60% say their learning and development programs cannot keep pace with AI transformation. And only 54% of organizations proactively arrange AI upskilling, which means 46% are still reacting to skills shortages after they appear instead of getting ahead of them.
Add the hiring pressure: 64% report difficulty finding AI-skilled talent because the skills they need keep changing.
So the picture is a workforce asking to be trained, a training function that cannot keep up, and a talent market that cannot supply the skills fast enough either. For a CEO or a VP of operations, that is the opening. The companies that build a deliberate, measurable path to proficiency will pull ahead of the ones still hoping to hire their way out of the problem.
Middle managers are where adoption succeeds or stalls
One more finding deserves a place in your next leadership meeting. The study identifies middle managers as the people who make AI adoption work. Ninety-five percent of HR leaders say middle managers are essential for effective AI use, and 92% say they are essential to redefining job roles. Ninety-seven percent of organizations are actively planning to redefine roles. The translation for an executive: your AI strategy does not succeed or fail in a vendor contract or a model choice. It succeeds or fails at the manager level, where roles get redrawn, where junior people learn to supervise the tools, and where the new way of working either becomes the norm or quietly dies. If your managers are not equipped to lead that change, the investment above them does not reach the work below them.
This is the same direction the broader data points to
The Cognizant and Pearson study fits a larger pattern Cognizant laid out earlier this year. Its "New Work, New World 2026" report, released January 15, 2026, found that AI is already capable of affecting 93% of U.S. jobs and could move up to $4.5 trillion in U.S. labor tasks. That figure came from a reassessment of 18,000 tasks and 1,000 jobs in the O*NET labor database.
The throughline across both is that AI changes the task before it changes the headcount. Most jobs are bundles of tasks, and AI is rebalancing which tasks belong to people and which belong to the tool. Entry-level work is simply where that rebalancing is most visible, because so much of it was routine to begin with.
For its part, Cognizant is not treating this as a reason to stop hiring early-career talent. The company hired 20,000 fresh graduates in 2025 and expects to exceed that in 2026. The bet is on developing junior people into the new kind of role, not removing the rung.
What the people closest to the data are saying
Two quotes from the study leaders capture the takeaway well.
Kathy Diaz, Chief People Officer at Cognizant, put the pressure plainly: "AI is reshaping the talent landscape and exposing the limits of traditional talent and learning models." As entry-level tasks and skill requirements change rapidly, she said, "organizations must rethink how they hire and develop talent at pace."
Ali Bebo, Chief Human Resources Officer at Pearson, pointed to where the advantage will sit: "As work evolves, the most successful organizations will focus less on replacing tasks and more on building the capabilities that help humans and AI work together. That starts with early-career talent. The future belongs to organizations that combine AI innovation with a deep understanding of how people learn, develop, and apply new skills in the real world."
A next step worth taking
If the entry-level rung is being redrawn around judgment instead of rote output, the question for your team is how to measure where your people stand today and build a deliberate path forward. A clear read on proficiency, role by role, turns a vague worry about being behind into a plan you can act on. That is what The 7 Levels of AI Proficiency assessment is built to give you: a measurable starting point for the climb the data is already describing.
Sources
- Entry-Level Work Remains Essential: 94% of HR Leaders Expect AI to Create New Entry-Level Roles, Cognizant and Pearson Study Reveals
- Adapting Roles for an AI Workforce (Cognizant Insights)
- Cognizant, New Work, New World 2026: How AI Is Reshaping Work
Related reading: Level 3: The Lieutenant (Critical Thinker).
Frequently Asked Questions
Does this study say entry-level jobs are going away?
No. It says the opposite about the roles themselves: 85% of HR leaders still call entry-level positions essential, and 94% expect AI to create new entry-level roles within five years. What is changing is the work inside those roles, with AI already doing about a third of entry-level tasks.
What should a leader actually do with this?
Start by mapping where AI is already absorbing routine tasks in your junior roles, then redefine those roles toward supervising and judging AI output. Equip your middle managers to lead the change, since the study identifies them as the people who make adoption stick. And move your training from reactive to proactive, because most organizations have not yet.
Is AI fluency only a concern for technical hiring?
The study says no. Ninety-eight percent of HR leaders emphasize AI fluency when hiring for non-technical roles, alongside 97% who now weight soft skills more heavily.
How was the study conducted?
Wakefield Research surveyed 750 HR professionals at director level and above, at companies with 1,000 or more employees, across the U.S., U.K., and India, between March 23 and April 3, 2026. Cognizant and Pearson released the findings as The AI Workforce Pulse on June 18, 2026.
Find your AI Proficiency level
The free 7 Levels assessment places you across seven stages of AI capability. Under ten minutes. Research-backed scoring.