As corporations rush to deploy artificial intelligence across offices and software systems, a growing number of economists and labor experts are warning that the drive to automate junior roles could create a damaging long-term talent vacuum inside some of the world’s largest companies.

At the center of the debate is a paradox increasingly visible across the technology sector: firms are aggressively investing in AI to improve productivity, yet many are simultaneously scaling back the entry-level positions that traditionally produced future senior talent, technical specialists, and corporate leaders.

According to Fortune, Andrew McAfee, principal research scientist at Massachusetts Institute of Technology and co-leader of its Initiative on the Digital Economy, believes companies may be underestimating the long-term consequences of that strategy.

“How else are people going to learn to do the job except via on-the-job learning and training apprenticeship?” McAfee said in remarks to Harvard Business Review.

“That’s how you learn to do difficult knowledge work is by helping somebody who’s good at that with the routine stuff. And when we put too much automation in that too quickly, we lose that apprenticeship ladder.”

His warning comes as generative AI systems increasingly absorb tasks that once served as foundational training work for graduates and junior staff. Functions such as document drafting, coding assistance, research compilation, financial modelling, customer support, and administrative coordination are now being automated at scale through tools developed by companies including OpenAI, Anthropic, Google, and Microsoft.

That shift is beginning to alter corporate hiring patterns. Recruitment platform Handshake reported that entry-level job postings have fallen below pre-pandemic levels, while the unemployment rate for recent U.S. college graduates aged between 22 and 27 has climbed to 5.6%, according to data from the New York Federal Reserve.

The deterioration in hiring conditions is feeding a broader sense of unease among younger workers entering the labor market during what many economists describe as the earliest large-scale AI disruption cycle.

According to Monster, nearly 90% of graduates in the class of 2026 believe AI could eliminate entry-level jobs, a sharp increase from the previous year.

The concern has been amplified by comments from senior technology executives themselves. Dario Amodei, chief executive of Anthropic, has repeatedly warned that AI systems could eventually remove up to half of entry-level white-collar positions.

Yet labor analysts argue that eliminating junior roles could produce structural weaknesses that become visible only years later.

Entry-level work has historically served as the foundation of corporate succession planning. Junior analysts become managers, associates become executives, and trainees evolve into specialists with institutional memory. Without those early-career layers, companies may eventually struggle to replenish leadership pipelines organically.

McAfee argues that firms are also overlooking another advantage tied to younger workers: AI fluency itself.

A Deloitte survey found Gen Z has the highest adoption rate of standalone AI tools among all generations, with roughly 76% reporting active usage. Analysts say younger employees are often more comfortable experimenting with AI systems, adapting workflows around them, and identifying new commercial applications.

“There is a big demographic falloff,” McAfee said. “As people tend to get older, we tend to be more set in our ways and less willing to try crazy new things like AI.”

In effect, some corporations may be removing precisely the employees most capable of accelerating internal AI adoption.

The contradiction is becoming more apparent across Silicon Valley and corporate America. Even as firms tout AI-driven efficiency gains to investors, many are quietly discovering that replacing junior employees entirely is harder than expected.

Several executives have acknowledged that AI systems still require extensive human supervision, context management, and quality control. In industries such as law, finance, consulting, and software engineering, junior employees often perform the operational groundwork that allows senior professionals to focus on higher-value decisions.

Without that layer, some analysts warn, productivity bottlenecks could simply shift upward rather than disappear.

There is also mounting evidence that companies continuing to invest in graduate recruitment view AI not as a substitute for junior talent, but as a force multiplier. IBM chief executive Arvind Krishna said the company intends to expand college hiring even as it integrates AI more deeply into operations.

“People are talking about either layoffs or freezing hiring, but I actually want to say that we are the opposite,” Krishna said.

Salesforce has also increased graduate recruitment tied to AI development initiatives. Chief executive Marc Benioff recently said the company would hire 1,000 graduates and interns to help build AI systems.

At Amazon, executives have maintained that demand for software engineers remains strong despite rapid AI deployment. AWS chief executive Matt Garman said the company plans to recruit roughly 11,000 software engineering interns this year.

The divergence in hiring strategies reflects a broader uncertainty surrounding the future of white-collar work. Some firms view AI primarily as a labor replacement tool capable of reducing headcount and operating costs. Others increasingly see it as infrastructure that still requires large pools of adaptable human talent to generate commercial value.

Historical precedent offers mixed signals. Previous waves of automation displaced certain categories of work while creating entirely new industries and professions. Economists note that younger workers have generally adapted more successfully to technological disruption because they are more flexible, more mobile, and quicker to acquire emerging skills.

A recent analysis by Goldman Sachs found that younger college-educated workers tend to recover more effectively from displacement shocks and are more likely to transition into technology-complementary roles. Still, the pace of generative AI development is unusually fast, compressing transitions that previously unfolded over decades into just a few years.

That acceleration is forcing companies into a difficult decision with lasting implications. Companies now have to decide whether to treat AI as a replacement for entry-level talent or as a tool that amplifies the capabilities of the next generation entering the workforce.

OpenAI chief executive Sam Altman says artificial intelligence is fundamentally reshaping the startup world, lowering the barriers that once kept non-technical founders from building major companies and altering long-held assumptions about what makes a successful entrepreneur.

Speaking at Stripe Sessions alongside Patrick Collison, Altman argued that the rise of generative AI tools has sharply reduced the premium Silicon Valley traditionally placed on elite engineering talent, creating room for founders whose main strength is a deep understanding of customer problems rather than coding expertise.

“For a long time, I think the most important ingredient that I looked for YC looked for, that kind of this part of our industry looked for on a founding team was technical talent,” Altman said. “And that’s still very important, but now people who just really deeply understand their users and can’t code at all. I want to fund those people.”

The remarks mark a striking shift from the culture that defined Silicon Valley for decades, particularly within startup accelerator Y Combinator, where Altman built his reputation before leading OpenAI.

For years, investors routinely dismissed so-called “idea guys” — founders who claimed to have transformative business concepts but lacked the engineering ability to build products themselves. In the pre-AI era, venture capital firms often viewed technical founders as indispensable because software development required highly specialized coding expertise and large engineering teams.

Altman acknowledged that this mindset is now changing rapidly as AI coding assistants, autonomous software agents, and large language models dramatically compress development timelines.

“All of a sudden it’s like the revenge of the idea guys,” he said.

The comments come at a time when generative AI is beginning to alter the economics of startup formation across industries. Tools from OpenAI, Anthropic, Google, Microsoft, and a growing list of AI firms are allowing small teams to perform work that previously required large engineering departments, from writing software code and debugging systems to building websites, automating customer support, and analyzing data.

That transition is already reshaping venture capital strategies. Investors increasingly believe the next wave of startups may emerge from domain specialists in healthcare, law, logistics, education, and finance who can pair industry expertise with AI systems rather than build large technical teams from scratch.

The shift is also helping fuel an explosion in AI-native startups. According to multiple industry estimates, funding for AI companies has surged to record levels this year as investors race to back firms seen as capable of leveraging foundation models into scalable businesses with lower staffing costs and faster product cycles.

Altman, whose early investment track record includes stakes in Reddit, Stripe, and Airbnb, suggested that founders with sharp product instincts may now hold greater leverage than in previous startup cycles.

He recalled how Silicon Valley once openly mocked entrepreneurs who guarded vague business ideas while searching for programmers to execute them.

“There were these people that wanted to start a company and they’d say like, ‘I have the best idea. I’m not going to tell you what it is. I have the best idea. I just need a coder to build it for me and then I’m going to be in great shape,’” Altman said. “And we would make fun of these people.”

Now, however, AI systems are increasingly acting as those coders.

But the development carries broader implications for the technology labor market. Analysts say AI-assisted programming is already beginning to compress demand for entry-level software engineering tasks while increasing the importance of product design, workflow integration, and industry-specific expertise.

Several technology firms have also started reorganizing teams around AI-enhanced productivity. Companies are increasingly expecting smaller engineering groups to deliver output that previously required significantly larger workforces.

Still, Altman cautioned that AI has not erased all traditional startup fundamentals. He maintained that founder chemistry and long-term trust remain critical ingredients for building enduring companies.

“The teams that came together seven days before applying to YC on a cofounder matching side or whatever, that didn’t work too often,” he said.

Altman pointed to his long-standing relationship with OpenAI cofounder Greg Brockman as a major reason the company survived years of intense pressure, competition, and internal turmoil.

“I think we had this deep mutual respect and complimentary skillset that has just worked really well,” Altman said.

The comments arrive as OpenAI sits at the center of an increasingly fierce global AI race involving rivals such as Google, Microsoft, Amazon, and Anthropic. The company’s rapid ascent has transformed Altman from a well-known Silicon Valley investor into one of the most influential figures in global technology and policy discussions surrounding artificial intelligence.