The Experience Crisis No One’s Talking About (Yet)

March 9, 2026

Last week, we watched markets reprice services businesses in real time. Intuit, Thomson Reuters, Raymond James, CBRE — established firms punished not for bad earnings, but because investors questioned the durability of business models built on structured expertise.

This week, I want to go one layer deeper. Because inside every organization that watched its sector take a hit, there’s a quieter crisis unfolding. One that won’t show up in a stock ticker for years — but will determine which firms survive the next decade and which ones hollow out from the inside.

It’s the experience crisis.

A Renaissance Word. A Permanent Truth.

“Apprenticeship” sounds ancient — because it is. The word traces to the Latin root apprehendre — “someone learning” — and was formalized during the medieval guild system in Renaissance Europe. Guilds structured knowledge transfer around a simple premise: mastery isn’t taught in a classroom. It’s built through reps. Through proximity. Through doing the work badly enough times that you start to feel when it’s right.

The word is six hundred years old. The concept is unchanged.

In every professional services firm — legal, financial advisory, consulting, accounting, wealth management — expertise has been built the same way. Junior professionals learn by doing. They research. They draft. They model. They sit in rooms where decisions get made and absorb the patterns that turn knowledge into judgment.

A first-year analyst doesn’t understand risk because someone explained it on a whiteboard. They understand risk because they built 200 models and watched 50 of them get torn apart by a partner who’d seen it go wrong before.

That’s not training. That’s apprenticeship. And it’s the invisible engine behind every firm’s long-term value.

Why Experience Builds Resilience

My doctoral research on leadership resilience keeps returning to the same finding: resilience isn’t a personality trait you either have or don’t. It’s a capacity built through experience. Specifically, through the accumulation of challenge, recovery, and recalibration that happens when people are exposed to real problems over time.

This is well-documented in the psychology literature. Research on post-traumatic growth and organizational resilience frameworks consistently show that the people and organizations best equipped to handle disruption are those with the deepest repositories of navigated difficulty. Not theory. Not training. Actual lived experience with hard things.

In leadership terms: resilience is built from experience. Experience is built from reps.

And AI just took the reps away.

The Engine Just Got Disrupted

Here’s what’s happening: AI is absorbing exactly the work that the apprenticeship model depends on.

The research that took a junior associate three days? AI drafts it in minutes. The financial model that took a first-year analyst a week? AI generates the first version before lunch. The contract review, the competitive analysis, the scenario planning — the entire layer of cognitive labor that built expertise through repetition is being compressed.

And most firms are celebrating.

Efficiency is up. Turnaround times are down. Clients are getting deliverables faster. From a quarterly metrics standpoint, this looks like progress.

But if AI handles the reps, how do your people develop the judgment that made those reps valuable in the first place?

The data is starting to confirm the concern. A Harvard study analyzing 62 million U.S. worker profiles found that firms actively adopting generative AI have significantly reduced junior hiring while continuing to expand senior ranks — what researchers call “seniority-biased technological change.” The decline is driven by slower hiring, not layoffs. Firms aren’t firing juniors. They’re just not creating the entry points that build the next generation.

The Efficiency Trap

I call this the efficiency trap — and it’s the most dangerous pattern I’m seeing in organizations at inflection points right now.

It works like this: a firm adopts AI tools that accelerate junior-level work. Productivity metrics improve immediately. Leadership sees faster output, lower cost per deliverable, happier clients in the short term. They double down.

But underneath those metrics, something structural is shifting. Junior professionals are reviewing AI outputs instead of creating original work. They’re editing instead of building. Quality-checking instead of pattern-recognizing.

The difference matters enormously. Building a model from scratch and discovering where it breaks is a fundamentally different cognitive activity than reviewing one an AI generated. Spending three days immersed in a dataset and developing an instinct for what’s missing is not the same as scanning an AI-generated research brief.

One builds resilience. The other builds familiarity.

This isn’t theoretical. PwC confirmed in 2025 that it plans to cut U.S. entry-level hiring by one-third by 2028, citing AI and automation as primary drivers. Their AI Assurance Leader told Business Insider that in three years, new hires will be performing today’s manager-level work — with no clear plan for how they develop the judgment to match.

Play This Forward Five Years

A firm spends the next two to three years aggressively leveraging AI to boost junior productivity. Their mid-level professionals — the ones who should be developing into senior advisors, partners, practice leaders — arrived at that career stage having done fundamentally less of the work that builds expertise.

They’ve reviewed more. Built less. Checked more. Struggled less.

Now that firm needs those mid-level professionals to exercise judgment in a high-stakes situation — a complex deal, a volatile market, an ambiguous client scenario. And the judgment isn’t there. Not because those individuals aren’t smart. Because the system that was supposed to build their judgment was optimized out of existence.

As Harvard Business Review researchers Edmondson and Chamorro-Premuzic warned, eliminating entry-level work isn’t just short-sighted — it threatens the development of future leaders, organizational innovation, and the long-term culture of firms that depend on human judgment.

This is a capability gap. Unlike a technology gap, you can’t close it by buying something. It takes years to develop. It takes intentional design to create. And once it’s lost, it’s extraordinarily expensive to rebuild.

Markets always price capability gaps eventually.

AI Didn’t Just Change the Work. It Changed How Experience Gets Made.

Here’s the part that isn’t getting enough attention: this isn’t just a talent problem. It’s a human capital architecture problem.

For generations, organizations have had a built-in experience engine. It was called the org chart. You hired people at the bottom, gave them volume work, and let time and repetition do the development. You didn’t have to design it. The structure did it for you.

AI broke that structure. Not maliciously — it just made the volume work obsolete. And in doing so, it eliminated the default mechanism by which organizations built resilient, judgment-capable people.

The McKinsey Global Institute confirms that unlocking AI productivity gains requires reimagining workflows rather than layering AI onto existing structures. Nearly 90% of companies have invested in AI, but fewer than 40% report measurable gains — because most are automating tasks without redesigning how people develop alongside them.

This means the experience that used to happen organically now has to be engineered deliberately. That’s a fundamentally new organizational design challenge — and most companies don’t have the infrastructure to address it.

Engineer the Experience

The solution isn’t to reject AI or slow down adoption. That’s not realistic and it’s not the point.

The solution is to engineer the experience deliberately. If the reps aren’t happening organically anymore, organizations need to create them on purpose — and they need to do it with haste. Not next fiscal year. Not when the capability gap starts showing up in performance reviews. Now.

That means asking questions most firms haven’t considered:

Where does struggle still matter? Not all cognitive labor is equal. Some tasks are genuinely commodity — automate them, pure gain. But others are developmental. Building a model from scratch, wrestling with messy data, writing a first draft that gets rejected — these are where pattern recognition forms. Organizations need to map the distinction and protect the developmental layer.

What does the junior role look like now? If AI handles the baseline work, the junior role needs to shift from “do the work” to “interrogate the work.” That’s a different skill set — critical evaluation, assumption testing, stress-testing scenarios. It requires deliberate design, not just handing someone an AI output and saying “check this.”

How do you create structured exposure to high-stakes judgment earlier? If juniors aren’t building expertise through volume, they need earlier access to the rooms where judgment happens — as participants, not observers.

Are you measuring development or just output? Most performance systems measure what gets produced. Almost none measure whether the person producing it is growing. Firms that want to protect their pipeline need development metrics alongside productivity metrics.

The Renaissance guilds understood something we’re forgetting: mastery isn’t transferred. It’s constructed. Through repetition, through failure, through proximity to people who’ve already built it. If the repetition is gone, you have to engineer it back in — or accept that you’re building an organization without a future bench.

What This Means in Plain English for Real Leaders

If you run an organization that’s been celebrating AI-driven efficiency gains, sit with one uncomfortable question: Are we building judgment as fast as we’re automating the work that used to build it?

If the answer is no — or if you’re not sure — you don’t have a technology problem. You have a human capital architecture problem. Your systems are optimized for today’s output at the expense of tomorrow’s capability.

The organizations that come out ahead won’t be the ones that adopted AI fastest. They’ll be the ones that redesigned how their people develop alongside it. That means building internal operating systems for human capital development — structured pathways that engineer experience, measure growth, and protect the judgment pipeline even as AI handles the volume.

Because when your product is human judgment, your talent pipeline isn’t HR’s problem. It’s your competitive moat. And right now, most organizations are draining it without realizing it.

Resilience is built from experience. Experience is built from reps. If the reps are gone, you’d better be engineering them back in — deliberately, urgently, and at scale.

Where Alchemy Comes In

At Alchemy Leadership Lab, this is exactly the problem we’re built to address.

We partner with organizations navigating inflection points — growth, succession, AI integration, transformation — to build the internal operating systems for human capital development that most firms don’t have. That means mapping where judgment actually develops inside your organization, identifying where AI adoption has inadvertently disrupted that development, and designing structured experience pathways that build capability intentionally.

Not a training program. Not a coaching engagement. An operating system for how your people grow — aligned to your business model, your growth stage, and the leadership capability you’ll need two and five years from now.

The price just went up on humanity. The firms that invest in building it deliberately will own the premium.

If you’re ready for a conversation: calendly.com/brynn-alchemyleadershiplab