AI is wiping out entry-level jobs. Here’s how to surf the wave and not get crushed by it

Not long ago, I was speaking at an event when a recent college graduate approached me. He’d studied neuroscience and, like a lot of STEM generalists, had set his sights on consulting—firms, like Deloitte or Accenture, that have long hired armies of junior associates for data gathering and analysis. He’d earned top grades at a great school. But all of his outreach—his informational interviews, his applications and follow-ups—had come to nothing.

His story is not unusual. If entry-level consulting or finance jobs have always been difficult to land, they’re even harder to get now. This generation grew up believing that developing key skills such as coding and data analysis, research and writing would get them a foot in the door for a fulfilling white-collar career. And now so many of those assumptions are wrong. Anthropic CEO Dario Amodei said AI could wipe out roughly 50% of all entry-level white-collar jobs within five years—absorbing the data entry, basic analysis, and research synthesis that used to be a new graduate’s first rung on the ladder.

A global study by the British Standards Institution seemed to confirm his prediction: Polling 850 business leaders in Australia, China, France, Germany, Japan, the UK, and the US, it found that business leaders have been prioritizing AI automation over training junior employees, with 39% saying they’d already reduced or cut entry-level roles due to AI and 43% saying they expect to do so in 2026.

The dominant narrative is doomsday: “AI is taking all the jobs!” And the data behind that narrative is real enough to generate legitimate concern. But the story is both more hopeful, and more complicated, than the data suggest. What we’re actually witnessing is a compression of the traditional career timeline that, navigated intentionally, can accelerate professional growth in ways no previous generation has experienced.

For first-job seekers willing to adapt their approach, and rethink how they envision their early-career arc, success remains within reach. Here are some new strategies to keep in mind.

Look For Companies That Are Bucking The Trend

Over the recent weeks, there have been signals that some employers are recognizing the danger of choking off their talent pipelines entirely. Reddit CEO Steve Huffman said that the company will “go heavy” on hiring new college graduates, because they are “AI native.” IBM likewise announced plans to substantially increase entry-level hiring. And Dropbox, Cloudflare, and LinkedIn have all signaled significant expansion of internships, new graduate, and entry-level programs. PWC, which partially rolled back entry-level hiring last year, recommitted itself to it in about 20% of its office locations. The firm now advises clients and others to continue hiring early-career workers or risk setting “starving [their] organization of its future.”

What’s happening in my view is this: companies that experimented aggressively with AI are realizing that young, adaptable people are critical to investing in growth and accelerating transformation. A generational age-out is coming. Succession and progression can’t happen if you’re only hiring into mid-career roles. And the mid-career people you might bring in don’t have the neuroplasticity or the openness—the unwritten chapters—that make early-career workers so valuable. Organizations are flattening, yes, but they still need humans who can both grow with the business and serve as more agile catalysts for organizational change.

Work At The Skills That AI Cannot Replicate

What should new-workforce hopefuls focus on? At the risk of stating the obvious, they should focus on the capabilities AI cannot replicate.

This includes relationship-building and its many subsidiary skills: The ability to listen deeply and synthesize what you’ve heard in multiple different offline conversations into something actionable. The ability to negotiate, facilitate a difficult conversation, or tell a compelling story. The ability to exercise judgment when the data is ambiguous.  The ability to read a room or tell an (appropriate) joke.

For human jobs, human skills. Such capabilities have always mattered in leadership, but they typically didn’t develop until mid-career, because early-career professionals were too focused on the kind of work AI now handles.

So this career-timeline compression is actually an opportunity. If the rote, lower-level work is being done by AI, new entrants can accelerate their development of these distinctly human skills, learning them much earlier than previous generations did. It’s about getting in the door and then, once you’re there, knowing that the first trimester of your career will not be consumed by the tactical as you’d always assumed it would. You must build the skills that create real leverage—storytelling, communication, intellectual and emotional agility, critical thinking, relationship building—from day one.

Make a point of describing your search strategy and objectives in these terms to job recruiters. It will demonstrate the self-awareness that is often lacking in even much more experienced job seekers.

Remember: This is Your Superhero Origin Story

I spend a lot of time talking to people in their early twenties, many of whom range from neutral to deeply unhappy in their first jobs. They are lucky to have landed a role over the past year, before many doors began to close. Some are on Wall Street making good money. Some are in solid companies with real training programs. And the most common answer I’ve gotten from this important cohort of people about how they feel about their jobs is “meh.”

When I reflect on my own early career, happiness was never part of the early equation. The goal was to satisfy basic needs: become financially independent, find an environment where I could learn and add value, meet some cool people. The idea of being happy at work would have struck me as secondary.

I don’t say this to dismiss the concerns of today’s early-career workers. I say it because I believe the framing is wrong. Instead of asking “am I happy?” try a more useful question, like “am I growing?” Look for satisfaction in your increasing competence, in mastering something difficult, in developing abilities in dealing with a wide variety of people. Importantly, there will be great satisfaction in knowing that a year from now you’ll be able to see how far you’ve come, and where you want to go will become clearer. That’s not the same as happiness but can lead there. My best career moments have been more about learning, advancement, and fulfillment in uncomfortable struggle than about my initial personal feelings which did actually flourish once I accomplished what I set out to do.

Keep An Ikigai Career Journal

Scott Galloway, the entrepreneur turned NYU business-school professor, podcaster, and author, often says that “follow your passion” is the worst advice you can give a young person. I agree. Most people in their twenties don’t know what their passion is. But they can pay attention. They can notice what piques their curiosity. They can track which parts of a meeting make them lean forward and which make them zone out. They can become attuned to what types of people they find more engaging, and importantly, what type of leadership attributes are admirable, perhaps even aspirational. As you wade deeper into a new role, write down where you excel, where you struggle; what energizes you, what drains you. This goes for the content and type of work as well as the people with whom you spend your business days. That kind of deliberate self-observation, accumulated over time, is how you find the intersection of what you’re good at, what the business needs, and what you actually enjoy. The Japanese concept of ikigai describes this convergence, and you can’t find it without experimenting.

Become a Great Mentee

One of the most underrated career skills is learning how to be mentored well. Experienced professionals want to help. But the person being mentored has to bring something to the relationship: respect, curiosity, vulnerability, a genuine willingness to build a connection that goes beyond transactional advice-seeking. Bonus: always bring an observation about the business, even suggestions for improvement, whether in processes, policies, or even team culture.

If someone takes the time to share their experience with you and you show up with gratitude, follow-through, and a willingness to be honest about what you’re experiencing, that person will go to bat for you. They’ll make introductions, advocate for your progression, and think of you when opportunities arise. Note also: Mentoring is one of the most human dynamics in a professional environment. And it’s one that AI will never replace.

Chart The Multi-Lane, Mad Scientist Career

The side gig was already a fixture of working life long before ChatGPT arrived. And technology, including AI, has continued to lower the barriers to entrepreneurship. You can build a website armed only with a two-paragraph description and an AI tool. Be a mad scientist. You can run a side business while holding a full-time job. You can operate in multiple lanes simultaneously.

For early-career workers who can’t find the entry-level job they want, this is worth exploring. Start something. Experiment. You’re building your own entry-level position. If a hiring manager sees someone who launched a business—even a small one—they’re looking at a person who understands initiative, risk, and execution, not to mention the subject-matter knowledge you will have developed. Career paths are no longer a single-lane highways. You can pursue several directions at once, in ways that, paradoxically, AI and other technologies have made possible.

Maintain AI Literacy As Table Stakes

If you’re entering the workforce in 2026, you must be able to use AI effectively to the same degree you once needed to be fluent in the Microsoft and Google office suites. There will be a transition period in which you’ll need competency across both AI and the various legacy toolsets. But AI is not optional. It is a baseline skill, like knowing how to use a spreadsheet twenty years ago. The candidates who understand how to use these tools to solve business problems large and small will have an advantage over those who don’t.

The good news for those just starting out is that they likely already use AI flexibly, in contrast to seasoned professionals who have approached it with a bit more bias and resistance, given our poor experiences with prior tech transformations. New capabilities have come online, like Claude Code and Cowork, OpenClaw and OpenAI’s Codex. Those who can integrate disparate systems in creative ways, those confident enough in their ideas to improvise and experiment, will be in demand.

Dropbox’s chief people officer Melanie Rosenwasser told Bloomberg that, when it comes to early career workers using AI, “It’s like they’re biking in the Tour de France, and the rest of us still have training wheels.”

But AI proficiency alone isn’t enough. The experienced professional who combines deep business acumen, strong relationships, and AI fluency is nearly uncatchable. What that means for new entrants and aspirants is that basic AI skills will be expected and your differentiator will be the human capabilities you develop alongside it.

The Case for Optimism

I’m optimistic about this generation. Gen Z is more socially aware, more globally connected, and more principled than perhaps any generation before them. They won’t sacrifice themselves for a broken system. There’s something powerful in that.

The world they’re entering is turbulent. The rules are changing. But they have a chance to build careers that are more varied, more self-directed, and more human than anything my generation has experienced. What such a career asks of us, however, is a willingness to be curious, to invest in the skills that matter most, and to ride this wave rather than let it wash us away.

The entry-level job you imagined may not exist anymore. But the opportunity has never been bigger. And this requires bigger thinking, bigger doing, and bigger leadership.

View the full article