A few days ago, while hiking in Mallorca with my friend Bea, our conversation turned to our children and the world they’re preparing to inherit. Bea is a real estate lawyer, and her daughter is studying law in college. Naturally, we began talking about artificial intelligence and what it might mean for the legal profession, particularly for young people just entering it.
Bea considers herself old school. She believes in people, in bringing young professionals into the workplace, giving them responsibilities, and allowing them to learn their craft from those who have practiced it for years. Yet she’s also acutely aware of the economic reality confronting her profession. Increasingly, the work traditionally assigned to junior lawyers can be performed by artificial intelligence, often faster and at a fraction of the cost. She may continue insisting on hiring people, but she knows that many firms will make a different calculation.
We talked about what that might mean for her daughter. And inevitably, I found myself thinking about mine.
There’s a unique discomfort in being a parent today. We spend years encouraging our children to work hard, pursue their interests, develop their abilities, and prepare themselves for adulthood. We offer advice shaped by our own experiences, even as we recognize that many of the assumptions underlying those experiences may no longer hold. We want them to become independent, capable adults, yet we cannot say with much confidence what the world will expect of them when they get there.
And somewhere along that hike, I found myself wondering about something more fundamental. In our enthusiasm to build machines that can do increasingly sophisticated things, have we given enough thought to what happens to the humans who haven’t yet learned how to do them?
For generations, the path into most professions followed a familiar progression. You studied, graduated, found someone willing to give you an opportunity, and began learning what your education hadn’t prepared you for. You started with modest responsibilities. You observed people who knew more than you did. You made mistakes, occasionally embarrassed yourself, and gradually developed the confidence and judgment that come from experience.
Much of that early work was mundane. Junior lawyers researched cases and reviewed contracts. Young accountants reconciled accounts and checked financial statements. Entry-level analysts gathered information and assembled reports. Aspiring journalists covered routine assignments before being entrusted with consequential stories.
Some of those tasks were tedious, even pointless. But buried within them were opportunities to observe, question, experiment, and discover that knowing the answer and understanding the problem were two very different things. Experience was something you accumulated through countless encounters with situations for which you weren’t entirely prepared.
And someone, somewhere, had to be willing to tolerate your inexperience.
Now consider what happens when the economic value of those early assignments begins to disappear.
If an AI system can review a contract, identify potential risks, summarize relevant precedents, and prepare a reasonable first draft in minutes, what incentive does a law firm have to employ a young graduate to spend hours doing the same work? If software can produce a market analysis or generate a financial report, why should a company hire someone who needs months of training before delivering comparable results?
From a business perspective, these are entirely reasonable questions. Every organization faces pressure to become more efficient, and technology has always allowed us to accomplish more with fewer resources.
But when every organization independently makes the economically rational decision to reduce its reliance on inexperienced workers, we may collectively create a problem that no individual organization has an incentive to solve.
Who develops the next generation of experts?
And who pays for their development?
I find this question especially uncomfortable because I occupy both sides of the argument.
As the founder of a company working at the intersection of education, technology, and human development, I’ve spent considerable time building AI-powered tools that improve efficiency and expand what people can accomplish. I believe deeply in the potential of technology to eliminate repetitive tasks and free people to focus on work requiring judgment, creativity, and empathy.
I also run a business. I understand the financial realities of hiring, training, managing, and retaining people. When technology offers a more efficient way to accomplish something, ignoring it out of nostalgia would be irresponsible. I cannot claim to be concerned about the disappearance of entry-level opportunities while pretending that I am immune to the very economic incentives driving that disappearance.
And yet, as a father, I find myself looking at the same transformation through an entirely different lens.
What happens when my daughters enter a world in which employers increasingly expect them to arrive with capabilities they haven’t yet had the opportunity to develop? What happens when the positions that once provided that experience are no longer available, or are so few that competition for them becomes ferocious?
I want them to be resourceful, adaptable, and willing to learn. But I also recognize that no amount of encouragement from a parent can substitute for a society willing to give young people the opportunity to prove themselves.
There is a contradiction here that I cannot easily reconcile. I want my business to be more productive, but I want my children to inherit a world that still has room for people who aren’t productive yet.
Perhaps we’ve underestimated how much of human development depends on someone else’s willingness to tolerate our inexperience.
Of course, there’s another side to this argument, and it deserves serious consideration.
We shouldn’t romanticize the old apprenticeship model. Many entry-level jobs involved tedious work with limited educational value. Young professionals were often underpaid, overworked, and expected to demonstrate gratitude for the privilege. Access to prestigious opportunities frequently depended as much on family connections and social networks as on ability.
And there’s no inherent virtue in spending three hours performing a task that technology can complete in thirty seconds.
In fact, AI could become one of the most powerful learning tools we’ve ever created. Imagine a young lawyer using it to examine complex legal questions, explore competing arguments, simulate negotiations, and receive immediate feedback. Perhaps young professionals will be able to develop sophisticated capabilities earlier than previous generations ever imagined, without enduring years of repetitive work simply to earn the right to do something meaningful.
I would welcome that future. It would be a mistake to assume that every traditional pathway to expertise deserves preservation.
But it would be an equally serious mistake to assume that because technology makes learning easier, employers will necessarily invest in learners.
The benefits of AI-assisted apprenticeship require intention. The elimination of positions can happen through a spreadsheet.
And I wonder whether we’re sufficiently honest about which of those outcomes our current incentives favor.
There’s another dimension to this that troubles me. Even if meaningful opportunities for beginners become scarcer, they probably won’t disappear equally for everyone.
Young people from affluent families may still have access to internships, professional networks, mentors, and opportunities to experiment without immediately earning a living. Their parents may be able to support them through extended periods of exploration, additional qualifications, or unpaid work.
But what about the young person whose family cannot afford another year of uncertainty? Someone who needs a paycheck immediately after graduation? Someone whose parents don’t know a lawyer, an executive, or a business owner who can open a door?
We already live in a world where opportunity is distributed unevenly. Could AI deepen that divide by making the opportunity to gain experience itself a privilege?
Maybe the next great inequality won’t simply concern who has access to technology. It may concern who has access to the human experiences that technology makes economically unnecessary.
And the consequences may take years to become apparent.
A law firm that replaces several junior associates with AI might become remarkably profitable. Its remaining lawyers might accomplish more than ever before. Clients might benefit from faster service and lower costs. By every conventional measure of productivity, the decision could be considered a success.
But what happens ten or fifteen years later, when those junior associates who were never hired would otherwise have become seasoned professionals? Who will possess the judgment that comes from years of navigating difficult clients, ambiguous situations, ethical dilemmas, and decisions whose consequences cannot be reduced to an algorithm?
Perhaps AI will have developed sufficiently to perform much of that work too. That’s certainly possible. But even then, I wonder what it would mean for a society to become increasingly dependent on sophisticated systems while fewer humans have the experience necessary to understand, question, and challenge their decisions.
We’ve become remarkably impatient with the process of becoming good at something. We celebrate speed, scale, optimization, and immediate results. We want students to demonstrate competence, employees to contribute from their first day, and investments to produce returns as quickly as possible.
Yet human development has never been particularly efficient.
People need time to discover what they’re good at. They need opportunities to make mistakes without those mistakes defining them. They need mentors who recognize potential before it becomes performance. And they need someone willing to invest in who they might become, rather than judging them exclusively by what they can already do.
I wonder whether we’ve begun to see inexperience as a defect to be corrected rather than a necessary stage of human development.
And whether, in our pursuit of efficiency, we’re forgetting that every accomplished person was once someone who didn’t know what they were doing.
I’ve thought about my conversation with Bea several times since that hike. I wonder what her daughter’s first years in the legal profession will look like. I wonder what opportunities my daughters will encounter, which doors will remain open, and which ones may have disappeared before they ever reach them.
Like most parents, I want to believe they’ll find their way. But belief is no substitute for examining the world we’re creating for them. And as someone actively building and adopting the technologies reshaping that world, I cannot pretend the responsibility belongs entirely to someone else.
Every expert was once a beginner. Every accomplished professional was once someone who needed a chance, a little patience, and another human being willing to invest in their potential.
We have spent generations telling young people that experience is the best teacher. Now we’re building a world that may no longer have much patience for the inexperienced.
And perhaps, years from now, when we wonder where all the capable people have gone, we’ll have to confront a rather uncomfortable possibility.
We became so good at doing things without beginners that we forgot how to make them.
Ex Cogitatione, Progressus
Girish
