Professional Growth in 2026: How to Build a Career When the Bottom Rungs Are Being Automated

For generations, careers began with small work. Newcomers prepared documents, checked records, answered routine questions and watched experienced colleagues handle the exceptions. These tasks could be repetitive, but they were also an informal apprenticeship: a steady stream of low-stakes practice that helped people build context, accuracy and judgment.

That apprenticeship was rarely designed with precision. Much of it came with the job. Now, as generative AI and automation take on more routine work, the informal learning system is under pressure. The result is a missing-ladder problem: organizations may expect junior employees to perform at a higher level while removing some of the early assignments that once helped them get there.

This shift makes professional growth a shared challenge. Individuals need ways to create the learning and feedback they might not receive automatically. Employers need to consider whether short-term efficiency is weakening the talent pipeline they will depend on later. The issue is not simply whether AI replaces jobs. It is what happens to expertise when people lose opportunities to practice.

The disruption is real, but uneven

A January 2026 survey of 546 U.S. HR leaders, commissioned by D2L and conducted by Morning Consult, found signs of a changing early-career landscape. More than half of respondents said GenAI had reduced the basic tasks delegated to early-career professionals. Thirty per cent said their talent strategy was shifting toward hiring fewer entry-level workers and more mid-level talent using AI for those tasks.

The survey also points to a structural risk. Fifty-eight per cent of respondents worried that reducing entry-level roles because of AI could contribute to a shortage of qualified senior leaders within five years. Meanwhile, 74% said they did not yet have active upskilling or employee development programs to replace on-the-job learning lost to automation.

These figures reflect employers’ perceptions, not a universal forecast. Other evidence offers a more mixed picture. NACE’s 2026 Spring Update, based on 185 employer respondents, found that more than one-third of entry-level jobs now require AI skills, nearly triple the fall 2025 figure. Nearly 60% of employers said they assign interns projects involving AI tools. Yet more than half reported that AI was not reducing the tasks entry-level workers perform, and just 11% were discussing replacing positions.

Taken together, the reports suggest that the ladder is being redesigned unevenly. In some workplaces, routine tasks are shrinking; in others, the work remains but now includes AI. Either way, expectations are shifting. Nearly half of the D2L survey respondents said AI was raising productivity expectations for entry-level roles, even where hiring was unchanged.

The challenge is not speed alone. It is whether people are gaining the experience to know when a fast answer is a good one.

How small assignments build judgment

Basic work can be a training ground when it includes real feedback and increasing responsibility. Reviewing a report repeatedly helps someone notice unusual patterns. Handling customer questions builds the ability to distinguish a clear request from an underlying problem. Drafting an analysis, receiving critique and revising it teaches more than producing a polished document once.

This development depends on three connected elements: repetition, feedback and range. Repetition lets a person see enough examples to recognise patterns. Feedback explains what was missed and why it matters. Range exposes them to different situations, so they learn where a familiar rule applies, and where it does not.

Automation can remove steps from that process. A tool may draft a response or summarise a dataset, but it does not automatically teach a junior colleague how to judge whether the result is relevant, fair or safe to use. Nor does faster output guarantee the person behind it understands the decision.

D2L’s Sandy Rezendes, Head of Corporate Learning and Development, warns that “foundational on-the-job learning” can be lost when tasks are automated. Michael Rochelle, Chief Strategy Officer at Brandon Hall Group, describes the moment as an inflection point: “AI is accelerating productivity, but it’s also disrupting the developmental pathways that have historically built expertise.”

That is the central tension. If the first drafts and routine analyses disappear, people must find deliberate ways to practice the decisions around them.

Build the reps that used to come free

Professional growth in this environment is less about collecting a generic list of skills and more about designing a personal practice system. The aim is to move toward work that combines tools with responsibility, context and learning.

1. Seek roles with stakes, not just tasks

When evaluating a role or project, look beyond its task list. Ask what decisions you will be expected to make, who uses the work, what happens when something goes wrong, and whether you will see the outcome. A role that includes a smaller share of consequential work can provide more useful learning than one dominated by high-volume production.

In interviews and internal conversations, ask: “What will I learn to judge independently by the end of six months?” The answer can reveal whether the position offers a progression path or simply a queue of deliverables.

2. Volunteer for the messy work AI cannot finish

AI can generate a first pass; many business problems still require someone to define the right question, reconcile conflicting information, understand stakeholder needs and handle exceptions. Seek projects where the problem is not fully specified or where the output must be checked against real-world constraints.

For example, offer to map a workflow before automating it, investigate why a process has recurring errors, or gather feedback from users after a new tool is introduced. These assignments help turn automation into a learning opportunity rather than a way to skip understanding.

3. Create a feedback loop with a senior reviewer

A finished deliverable is not always a learning experience. Ask a more experienced colleague to review not just the result, but the reasoning behind it. What assumptions did you make? What evidence changed the recommendation? What risk deserved more attention?

Keep the review focused and regular, perhaps a short debrief after each project milestone. Feedback is most valuable when it arrives close to the decision, while the reasoning is still fresh.

4. Keep a record of your judgment

A decision journal can make invisible progress visible. For important choices, record the problem, the evidence available, the risks considered, your decision and what you expected to happen. Later, compare the outcome with your prediction.

Include notes such as “I changed my mind because…” This captures how your thinking develops, not only what you produced. Over time, the record can show patterns in your judgment: where you assess evidence well, where you make assumptions too quickly, and which situations require a second opinion.

5. Teach and review to practice leadership

Early leadership does not require a title. Explaining a process to a newer colleague, reviewing a peer’s work or documenting a recurring decision forces you to make your own reasoning clear. It also builds interpersonal and communication skills, two areas where D2L respondents perceived declines among recent entry-level hires compared with cohorts from three to five years earlier.

Teaching is not a substitute for formal management experience. But it is a practical way to practice clarity, patience and accountability while helping a team preserve knowledge that might otherwise remain informal.

6. Negotiate learning into the role, and use AI deliberately

At a performance review, discuss development alongside output. Propose a stretch assignment, rotation, supervised client exposure or scheduled review time. Frame the request in business terms: what capability it will build, what work it will improve and how progress can be assessed.

Use AI to accelerate low-risk first drafts, organise information or test alternatives. But protect the parts of the work where judgment is formed: defining the problem, checking evidence, handling exceptions and explaining the recommendation. The goal is not to avoid assistance. It is to remain accountable for the thinking.

A practical 90-day plan

Days 1–30: Map the missing reps. List the recurring tasks in your role and identify which ones are automated, augmented or still handled manually. Then note the experience each task used to provide: pattern recognition, customer context, quality control or decision-making. Choose one capability you want to develop and ask a manager or mentor where you can practice it in current work.

Days 31–60: Take on a supervised stretch assignment. Find a bounded project that includes a real problem, a decision and a measurable outcome. Use AI where it helps, but agree with your reviewer on which parts you will own and how the work will be evaluated. Keep a brief decision journal.

Days 61–90: Review the evidence and agree on the next rung. Compare your initial assumptions with the project’s outcome. Ask for specific feedback, document what changed in your approach and discuss a next assignment that builds on it. The output of the 90 days should be more than a completed project. It should be evidence of improved judgment.

Employers must rebuild the ladder

Individuals can manufacture more opportunities to learn, but they cannot replace a development system that organizations have removed. If companies reduce junior hiring or compress work without providing new routes to expertise, they risk turning a short-term productivity gain into a longer-term capability gap.

Rebuilding the ladder does not mean preserving every repetitive task. It means preserving the learning those tasks once made possible. Employers can do this through structured apprenticeships and rotations, supervised stretch assignments, and reviews that treat AI-generated output as a teaching opportunity. A junior employee who checks a model’s response should learn how to trace its assumptions and validate its sources, not simply approve or reject it.

NACE’s findings underline the value of integrating AI into early-career work rather than treating it only as a replacement question: nearly 60% of surveyed employers said interns were assigned AI-related projects. The next step is to connect that exposure to mentorship, clear responsibility and opportunities to learn from mistakes safely.

Managers should also examine promotion criteria. If entry-level employees are expected to deliver faster, their path to broader responsibility needs to be explicit. Define what “ready for the next level” looks like, provide feedback against that standard and make room for supervised practice. Efficiency matters. So does the supply of people capable of exercising sound judgment.

The next career ladder is built intentionally

The apprenticeship model was never only about doing simple work. It was a way to encounter real problems in manageable stages, with people nearby who could explain what mattered. AI can change the shape of those stages, but it cannot make the developmental need disappear.

For professionals, the task is to seek responsibility, feedback and repeated practice, not merely more output. For employers, the task is to redesign entry-level work so that productivity does not come at the cost of future expertise. McKinsey Global Institute’s September 2026 research on workforce movement and pathways adds to the wider examination of how people and jobs may adapt as AI changes work.

The most durable advantage will not be speed alone. It will be the capacity to interpret uncertain information, understand people and take responsibility for decisions. Those abilities are built over time, through practice and reflection. When the old ladder loses its rungs, the work of building the next one becomes deliberate.

Sources

• D2L and Morning Consult, The Future of Work and Learning: GenAI Impact on Entry-Level Work
• D2L press release: Survey findings on AI, entry-level work and the talent pipeline
• NACE, Demand for AI Skills in Entry-level Jobs Nearly Triples Since Fall 2025
• McKinsey Global Institute, Workforce in Motion: Skills and Pathways to Future Jobs in the United States
• Explore more career development and professional growth and artificial intelligence articles from Our Success Journey.