By Bella Wu

Imagine graduating from college after years of studying, only to discover that many of the entry level jobs you had hoped to pursue can now be performed by artificial intelligence. Despite spending years developing the right skills, employers no longer need as many inexperienced workers to complete that work. In today’s job market, this possibility is becoming increasingly difficult to ignore.

Fears of an AI “job apocalypse” have amplified as generative AI becomes more capable. Dario Amodei, chief executive of Anthropic, has warned that AI could eliminate a significant share of white-collar jobs and push unemployment rates higher. However, predictions of widespread unemployment should not be confused with evidence that had already occurred. In the 2026 analysis, the Stanford Institute for Economic Policy Research found little evidence of widespread economy-wide job displacement proven to be due to AI. The whole picture is more complicated than the binary of an impending catastrophe versus a harmless technological development. AI is changing how work gets done, but its effects are uneven across occupations and experience levels. The greatest danger may not be the disappearance of work altogether, but the unequal losses of opportunities to enter and advance in the workforce. Whether AI ultimately expands or restricts economic opportunities will depend on just how effectively workers, employers, and institutions adapt to it.

One of the utmost concerns in the American labor market is the possibility that AI is weakening the pathways where young workers can begin their careers. Entry level positions often involve research, drafting, data analysis, coding and customer support: all the kinds of tasks that generative AI can now assist with or perform. As a result, occupations such as tech & product support, data & operations engineers, and legal tech consultants are especially vulnerable. When employers can use AI to complete assignments with fewer people, those inexperienced workers may find it harder to secure their first jobs. Erik Brynjolfsson, Bharat Chandar, and Ruyu Chen published a research report through Stanford’s Digital Economy Lab, providing evidence aligning with this concern. In their 2025 paper, Canaries in the Coal Mine, they analyzed payroll data and found that employment among workers ages 22-25 in AI exposed occupations declined 19% relative to their other less exposed peers. Noticeably, older workers in the same occupations did not experience a comparable decline. This distinction is relevant because it suggests that burden may fall on people who have not yet administered a professional establishment. For example, an experienced software developer may have years of specialized knowledge and understand how the company systems operate. A recent graduate applying for the same position may be competing for tasks that AI can already handle. However, if the employer reduces junior hiring, the graduate not only loses a paycheck but also an opportunity to get more experience necessary for more advanced work.

However, the research does not prove that AI alone caused the decline. After ChatGPT’s public release in November 2022, there were changes to remote work and adjustments due to overhiring due to the pandemic.  The extent to which the drop relates to AI influence remains inconclusive. Nevertheless, the possibility that AI is contributing to these weaker entry level hiring deserves attention, especially as AI systems become more capable in day to day life, and as businesses integrate them into everyday operations.

Research also shows that AI exposure does not automatically mean workers will lose their jobs. In the research paper, Generative AI at Work, by Erik Brynjolfsson, Danielle Li, and Lindsey Raymond, they found more than 5,000 customer service agents using an AI assistant. Access to it increased productivity overall by about 14%, with especially noticeable gains among less experienced and lower skilled workers. The AI assistant helped them resolve customer issues more efficiently and effectively, and allowed newer employees to benefit from better performances. This undermines the idea that inexperienced workers will always be the first to lose out on jobs. AI can replace some tasks previously assigned to junior employees, but it can also help those employees learn and perform better. It depends on how employers choose to use AI, whether to reduce the entry level workforce or train new employees to expand their capabilities.

Other occupations showcase different challenges for AI automation. Jobs like Nursing, teaching, counseling, and personal care involve physical interaction, communication, and relationships that require built trust. AI is able to support workers by organizing information, preparing materials and reducing workload, but it does not replicate human relationships that these jobs require. Consider nursing. Would you feel comfortable trusting a robot to care for you when you are sick, and to comfort you during a frightening medical situation? Students rely on teachers not only to explain academic concepts but to recognize their struggles and offer encouragement. Counselors must build trust to understand complex human emotions. In professions such as these, AI cannot replace empathy and personal connections that make these jobs meaningful. Although not immune to technological changes, these tasks present strong barriers to automation due to their reliance on human interaction.

Just last year, my friend’s father lost his job as a financial advisor after AI replaced much of his work. Since then, he has struggled to find a new job, leaving his family still troubled with the financial and emotional insecurity that comes with losing their primary source of income. His experience should help us understand how AI consequences extend beyond just the workplace, but affects entire families and their financial stability. Was their research role no longer needed, being accommodated by AI? Ultimately, the vulnerability of a job depends more on whether AI can perform these tasks. A teacher may use AI to prepare practice questions but still need to explain the concepts and respond to student needs. However, in the present example, AI instead substituted the worker, allowing employees to cut down the workforce.

The possibility of declining opportunities for early workers is a serious concern, even if data of employment rates can not fully show it. The labor market can seem to remain relatively stable for now, while it becomes harder to enter for certain positions. If employers increasingly rely on AI to complete work that is traditionally assigned to new employees, those workers may struggle to gain the experience needed. Consequences may also extend beyond someone’s first job, affecting their future and access to more professional opportunities. At the same time, AI can help people learn, improve productivity, and support work that still depends mostly on human  expertise. Whether these benefits can offset the disruption will depend on how businesses use the technology and whether workers can adapt to it meaningfully. Therefore, the American Labor market is not facing a problem that can be dismissed but nor is disaster inevitable. The greatest challenge is ensuring that progress does not come at the expense of other people trying to build their futures. The true obstacles will not be of how much human work can be replaced, but of whether society can use its capabilities for the benefit of the people.

Leave a Reply

Trending