AI Is Making You Faster. It May Also Be Making You Worse At Your Job

1. Competence: When the Scaffolding Becomes a Crutch

Genuine competence requires knowing that you produced the work, that your judgment and skill made the difference. That internal sense of ownership matters enormously, and AI is quietly eroding it for a generation of early-career professionals.

Before AI, skill development had a predictable shape. You drafted a memo badly, got feedback, drafted it again, and slowly internalized what good writing actually looked like. You built a financial model from scratch, made errors, and developed real intuition about where models break. The difficulty was part of the process.

Now, a 24-year-old analyst can prompt their way to a polished memo in four minutes. The output is impressive, but the underlying skill is often underdeveloped, and the person knows it. This creates a specific kind of discomfort: a gap between what you appear capable of and what you actually feel capable of. When the AI-assisted output gets challenged in a live setting, that gap becomes visible to everyone in the room.

The young professionals who thrive here treat AI as a sparring partner rather than a ghostwriter. Generate a first draft, then interrogate it. Build the model yourself first, then use AI to check your work. The goal is to develop judgment about when AI is wrong, and that only comes from having done the underlying work yourself enough times to notice.

2. Relatedness: The Loneliness Inside the Productivity

Relatedness has always been built in the margins of work: the ten-minute conversation after a meeting, the lunch where a senior colleague tells you how they actually navigated a political situation, the moment when someone trusts you enough to share what they couldn’t say in the formal debrief. Remote and hybrid work already compressed these moments, and AI is compressing them further.

When your primary collaborator is a chat interface, you get faster answers and fewer real relationships. You solve the immediate problem but miss the apprenticeship: the unwritten rules, the institutional history, the interpersonal dynamics that separate someone technically competent from someone who actually knows how to operate inside an organization. The young professional who can’t navigate a difficult client conversation because they’ve never watched a seasoned colleague handle one is missing experiences that previous generations accumulated almost by accident.

Organizations that treat human connection as optional will pay for it in turnover and quiet disengagement.

3. Autonomy: Choice, or the Illusion of It

Autonomy doesn’t require total independence, but it does require that your choices feel genuinely self-authored, reflecting your own values and judgment rather than algorithmic nudges.

The more you rely on AI to prioritize your inbox, structure your arguments, and flag your risks, the more your own judgment quietly atrophies. You start deferring rather than deciding. And the longer it goes on, the harder it is to tell the difference.

Picture a 25-year-old account manager who lets AI triage her inbox every morning for six months. One day, a client email flagged as low-priority turns out to be an early warning sign of a relationship in trouble, something a more experienced colleague would have caught immediately. She missed it because she’d stopped exercising the judgment that notices those things.

Professional identity is built through choices: making calls, living with the consequences, and iterating. If too many of those choices are quietly outsourced to AI before you’ve developed the foundation to evaluate them, you end up technically productive but professionally underdeveloped in ways that don’t show up until they really matter.

What Managers and Organizations Need to Do Differently

Most organizations are letting AI reshape early-career work without thinking through what gets lost in the process. Research tracking 138 million U.S. workers found that starting wages in AI-exposed companies fell 4.5% after ChatGPT’s launch, with junior positions absorbing a 6.3% pay drop while senior compensation held steady or rose. That’s a signal of how quickly organizations are devaluing the development of early-career talent.

Build in assignments where AI is off the table and developing judgment is the explicit goal, not speed. UCSB researcher Matt Beane frames the risk clearly: “The way you make a senior employee is not through school. It’s by doing the job alongside someone who knows more, and you learn by doing.” When juniors are treated as optional because AI is faster, you’re not just cutting costs today. You’re dismantling the pipeline that produces experienced senior talent in five years.

Then engineer the informal moments that used to happen by accident. Structured mentoring and visible senior modeling don’t materialize on their own in hybrid environments. The firms that understand this are putting real money behind it: Achieve Partners recently closed a $450 million fund to build apprenticeship programs in direct response to AI’s disruption of the labor market, backed by JP Morgan Asset Management and Prudential.

Finally, audit where AI is quietly making decisions that junior employees should be making themselves. Sampsa Samila, director of IESE’s Artificial Intelligence and the Future of Management Initiative, captures the long-term danger simply: “One of the key challenges is how junior people learn. Where do you get the experts?” Organizations that design AI-assisted workflows with human development in mind will build workforces that are genuinely capable.

If you’re early in your career, don’t wait for your organization to figure this out. The professionals who come out of this moment with real capability will be the ones who treated AI as a tool they mastered instead of a crutch they leaned on before they learned to walk.