Kicking Away the Ladder: AI, Entry-Level PR and a Profession Eating Its Own Seed Corn

3 September 2026

By Gerard F. Corbett. 

Somewhere right now, a communications major with a 3.7 GPA, two internships and a portfolio of clips is sending out her fortieth application of the semester. She may hear back from three agencies. One will ghost her after the first interview. The other two will tell her, politely, that they are “not hiring at the entry level right now.”

What they increasingly mean is that they no longer believe they need to.

We should be alarmed by that sentence. We are not nearly alarmed enough.

For decades, the entry-level job, the assistant account executive, the coordinator, the intern who never quite went home, was where PR professionals were made, not born. You learned to write a press release by writing forty bad ones. You developed news judgment by having a pitch rejected because it missed the point. You learned client counsel by watching a senior practitioner do it for three years before anyone trusted you to try.

The process was inefficient and occasionally humiliating. But it worked. More important, it created the pipeline that produced the senior counselors on whom the profession depends.

AI is now very good at many of the tasks that once filled those formative years: media monitoring, research synthesis, first-draft press releases, boilerplate social copy and background briefs. A generative model can produce in ninety seconds what once occupied a junior staffer for an afternoon. It does not need mentoring, managing, benefits or a desk.

Stephen Waddington and Ben Verinder have been examining this shift within public relations. Verinder put the danger more bluntly than I might have dared: The profession, he warns, “is going to see practice cutting off its own blood supply.”

The broader employment numbers reinforce that concern. A 2026 Resume.org survey of nearly 1,000 U.S. business leaders found that 21 percent of their companies had stopped hiring entry-level workers because of AI. Twelve percent said AI had already eliminated entry-level positions, while another 21 percent expected such positions to disappear before the end of 2026.

These are the reported expectations of business leaders, not independently measured employment figures, and they should be read accordingly. Even so, they reveal how a significant share of employers are thinking about the bottom rung of the career ladder.

Indeed data reported by Fortune document a related phenomenon in technology hiring. The share of postings open to candidates with two to four years of experience fell from 46 percent in mid-2022 to 40 percent in mid-2025. During the same period, the share requiring at least five years of experience rose from 37 percent to 42 percent.

Indeed economic research director Laura Ullrich calls this “experience creep”: Employers are demanding more experience from applicants even as the opportunities to acquire that experience are becoming harder to find.

Communications is showing its own warning signs. An analysis by Jobs in PR found that vacancies advertised by the United Kingdom’s 50 largest PR agencies fell 27 percent between the fourth quarter of 2024 and the first quarter of 2025, reaching their lowest level in a year. At the same time, the analysis found growing demand for AI- and data-related product, strategy and technical positions.

One quarter does not establish a permanent trend, and a hiring decline cannot automatically be attributed to AI. But the pattern raises a question the profession cannot afford to dismiss: Will employers continue to seek experienced practitioners while eliminating the jobs through which practitioners become experienced?

I know the rebuttal, and it deserves a fair hearing. Plenty of smart people in this field argue that AI will not eliminate junior roles so much as upgrade them. If a machine prepares the clip report, a 24-year-old can devote that time to strategy, message development and client-facing work she might not otherwise have encountered until her third year.

Some agencies are genuinely trying to redesign junior positions around judgment rather than routine production. I believe them when they say it is possible.

In fact, PwC’s 2026 Global AI Jobs Barometer offers evidence for that more hopeful interpretation. It found that entry-level positions highly exposed to AI were seven times more likely than less-exposed junior roles to require traditionally senior capabilities such as leadership and strategic thinking. PwC also found that these “seniorised” entry-level roles had grown 35 percent since 2019.

That sounds promising. It also exposes the problem.

Employers may be raising the level of judgment they expect from junior practitioners without rebuilding the training systems through which such judgment is acquired. Calling an entry-level position “strategic” does not make a new graduate a strategist. Giving someone access to a client meeting is not the same as teaching that person how to read the room.

Rebuilding a career path around supervised judgment requires deliberate investment. It demands new training models, more senior time devoted to teaching and a willingness to carry junior headcount while the long-term benefits accumulate.

Freezing a requisition takes an email.

Every agency and corporate communications department will find the second option easier than the first. In a margin-squeezed industry, what is easiest has a way of becoming policy. Productivity gains do not automatically become investments in people. Unless leaders make that choice deliberately, efficiency will be harvested now and the consequences deferred to someone else’s budget.

That is what should keep us up at night.

This is not only a story about young people losing jobs, serious as that is. It is a story about where the next generation of senior counsel will come from.

The judgment a good chief communications officer exercises in a crisis is not downloaded. Neither is an instinct for what a reporter actually wants, the ability to recognize a client’s blind spot before it becomes a headline, or the nerve to tell a CEO something she/he does not want to hear.

Those qualities are developed slowly, through thousands of small assignments, corrections, observations and decisions. They are built through the repetitions that used to constitute someone’s first job.

Remove those repetitions and, ten years from now, we will not merely face a shortage of entry-level practitioners. We will face a shortage of senior counselors, people standing in rooms they were never adequately prepared to enter.

So what should we do?

First, agencies and corporate communications departments must stop treating “efficiency gain” and “headcount reduction” as synonyms. The junior years should be recognized as a professional-development investment, not simply as a collection of tasks that AI conveniently allows employers to eliminate.

Second, employers must redesign entry-level positions intentionally. If AI takes over monitoring, drafting and routine research, junior practitioners need structured opportunities to develop judgment: participating in client meetings, analyzing failed pitches, working through crisis simulations, defending strategic recommendations and receiving serious feedback from experienced professionals.

Third, professional associations and educators, and I say this as someone who serves on exactly this kind of committee, must stop preparing people for a career ladder from which employers are quietly removing the bottom two rungs. We need explicit apprenticeship models, shared training standards and employer commitments that can survive the automation of routine work.

Finally, those of us further along in our careers must say loudly and publicly that a profession unable to explain how it will produce its successors does not have an AI strategy. It has a countdown clock.

The entry-level PR job is not merely someone’s first paycheck. It has long been this profession’s principal means of manufacturing judgment.

We are dismantling that system quietly, one unfilled requisition at a time, and congratulating ourselves on the efficiency. Unless we build a replacement, we will discover that we did not eliminate the profession’s least important work.

We ate our own seed corn.

Gerard F. Corbett, APR, Fellow PRSA, is CEO, Redphlag.