AI is redefining entry-level tech roles — here’s what IT leaders need to change now

We’ve all seen the headlines announcing the end of entry-level jobs, especially in tech. Given my role as President of Per Scholas, a nonprofit that provides no-cost training and then connects individuals to tech careers, it’s something that comes up often.

A common question: “What’s Per Scholas going to do when all the jobs you train for are taken over by AI?”

My response: “Keep doing what we’re doing. Train for the roles where early-career technologists work with AI instead of competing against it.”

The fact is, the CIOs I’ve spoken to in the last several months aren’t just anxious about entry-level hiring. They’re worried that AI adoption is accelerating faster than their workforce strategies can keep up.

These tech leaders realize that rather than eliminating entry-level roles, AI is redefining them — and that a successful AI strategy needs to include a thoughtful analysis of how those roles should be re-engineered to pair human potential with emerging technology from day one.

The most forward-thinking CIOs are identifying how entry-level work is changing, where their organization’s talent practices are misaligned with this reality, and the changes that they can make this year to move from awareness to action.

The myth of ‘AI replacing entry-level talent’

In a recent survey of global CEOs, 67% reported that AI is increasing entry-level headcount. According to the World Economic Forum’s Future of Jobs Report 2025, the global economy will see a net increase of 78 million jobs by 2030.

And in 2025, Per Scholas recorded more initial job attainments for graduates of our training program than in any other year in our 30-year history.

In other words, demand for entry-level tech talent remains strong. The issue isn’t fewer roles — it’s misaligned expectations.

Take this analysis of 2,000 “entry-level” job postings on LinkedIn last summer. It found that in industries like software and IT, more than 60% of jobs required 3 or more years of experience. Which of course leads to the classic catch-22 — how can someone gain the experience to get hired when no one will hire them without that experience?

And if they are hired, how can they excel in their current role while upskilling to keep pace with the rapidly changing tech landscape?

How AI is actually changing entry-level work

Five years ago, a Per Scholas graduate trained in Software Engineering would spend the bulk of their time creating user interfaces and styling websites line by line. Today, an employer would expect them to be a capable Prompt Engineer, to assist in the creation of client-facing aspects from their first day on the job.

AI is an amazing tool that can drastically improve productivity. What it can’t provide, however, is judgment and accountability.

And there’s a price to that increasing productivity. Without the “grunt work” that enables junior employees to learn by doing, companies could be facing what Gartner analyst Nate Suda terms “Experience Starvation” — higher-level workers find it easier and faster to use AI to help complete a new task rather than enlist a junior employee. That means no one is mentoring entry-level talent so they can develop the necessary experience and expertise to become those higher-level workers in the long-term.

Where CIOs are seeing a breakdown

Three big areas of concern come up when I speak with CIOs: hiring criteria, onboarding and expectations around working with AI.

Job descriptions written even two years ago no longer resemble the actual work today. A post for a junior developer role that lists “write and debug code” doesn’t reflect that the job now most likely involves validating AI-generated outputs, reviewing code at unprecedented volume and troubleshooting edge cases AI can’t handle.

Then there are the postings seeking a “Junior Data Analyst” requiring 3 years of SQL experience, proficiency in Python and R and experience with Tableau and Power BI. These postings screen out actual junior data analysts who could thrive with proper mentorship and support.

Which raises the issue of what that support should be. When new hires no longer have the opportunity to learn by performing routine tasks, what should their onboarding involve? How can they judge AI output effectively with the necessary domain knowledge?

What forward-looking CIOs are doing differently

Don’t worry; I’m not just asking questions without having some answers. I’ve learned several lessons from tech leaders I admire, and can also draw on my own experience implementing AI at Per Scholas.

First: Lean into skills-first hiring. Identify the specific technical and durable skills a role actually requires, and distinguish what candidates must know on day one versus what they can learn on the job. (Try to be flexible on that front — the ability to learn and adapt is often more important than knowledge of a specific tool, especially when that tool could change quickly.)

Second: Invest in career development. Accenture — one of the largest AI services providers — sources 20% of its entry-level hires from its apprenticeship program. Barclays Apprenticeships program offers entry-level talent the opportunity to develop targeted skills working side by side with more experienced workers. And then there are Per Scholas’ employer partners (including Accenture and Barclays), who not only hire our graduates but help design our curriculum, ensuring that our training reflects their real-time needs.

Finally: Design AI-supported workflows that preserve or accelerate learning instead of replacing it. Taking the time to do this has a demonstrated ROI. McKinsey’s State of AI 2025 report found that “intentional redesigning of workflows has one of the strongest contributions to achieving meaningful business impact [of AI use] of all the factors tested.”

At Per Scholas, that means putting people before platforms. Clear communication, role clarity and training drive change. We also test workflows with small groups and refine with staff input, as they’re the ones closest to the work.

3 changes CIOs can make this year

Without a workforce strategy that incorporates AI, organizations run the risk of failing to generate a return on their AI investments –– and could face a huge talent shortage in the years ahead. Bain & Company estimates that in the US alone, 1 in 2 AI jobs could be unfilled by 2027 unless 700,000 workers are reskilled.

There are three concrete steps that technology leaders can implement today to set their organization up for success:

1. Audit “entry-level” job descriptions against actual role requirements. Remove credential and experience requirements that effectively exclude new tech talent without improving candidate quality.
2. Redesign onboarding to replace lost learning opportunities. Map the tasks handled by AI that previously taught new hires essential skills. Create structured alternatives — shadowing programs, rotational assignments — that build the same competencies. Measure whether new hires are reaching proficiency benchmarks, and adjust programs based on the data.
3. For AI fluency, consider investing in targeted AI training by partnering with an organization that specializes in it. Create internal pathways for tech talent to progress from independent AI-augmented work to AI workflow design. Technological innovations are happening every day, and to keep your talent on the cutting edge, look into short-format upskilling programs that fit the lives of working professionals.

The organizations that will lead in the AI era recognize a fundamental truth: AI transformation is workforce transformation.

Entry-level tech roles in 2025 looked vastly different from those in 2020 and will look vastly different from the ones we’ll see in 2030. CIOs who treat this evolution as a workforce design challenge — not just a hiring problem — will build the adaptive, AI-capable teams that define the next decade of competitive advantage.

This article is published as part of the Foundry Expert Contributor Network.
Want to join?