How AI Erodes or Enhances Expertise

This post is part of our series on expertise and AI. If you're concerned about preserving and enhancing the value of your expertise as AI advances, consider joining our short course in August to learn the psychological strategies needed to make your expertise even more valuable with AI.

Microsoft announces plans to cut 9,000 jobs while investing heavily in AI. Amazon's CEO says AI will lead to "fewer people doing some of the jobs." Headlines scream about "white-collar bloodbath" and warn that "AI will wipe out jobs."

But here's a contradiction: PwC's 2025 Global AI Jobs Barometer found that AI is linked to a fourfold increase in productivity growth and 56% wage premiums, with jobs growing even in the most easily automated roles. AI is "amplifying and democratizing expertise," the report concludes, helping companies emerge as leaders in their fields.

So which is it? Are we facing mass unemployment or an expertise renaissance? The answer depends on how AI multiplies expertise and how professionals choose to respond.

The Pattern Behind the Headlines

The "white-collar carnage" hitting professionals right now may be less about AI capability and more about timing and broader economic uncertainty. Perhaps companies are cutting expensive, experienced workers now—before AI productivity gains translate into higher wages they'd have to pay later—a preemptive move to lock in cost savings. Whatever it is, it's hitting professionals who have the most to lose: those with decades of experience, mortgages, and career identities built around expertise that suddenly feels fragile.

But this strategy misses something crucial about how AI actually affects expertise. AI acts as a force multiplier. And it works differently depending on what you're multiplying. When AI automates the core tasks that define expertise, it reduces demand for human experts. When AI automates the routine work around expertise, it makes the remaining expert work more valuable and complex.

This creates two completely different paths for every profession.

Two Paths

Path 1: The Automation Track When AI can replicate expert-level work reliably, those roles become candidates for automation or de-skilling. Tasks that once required years of training—financial analysis, basic legal research, routine medical interpretation—become accessible to less experienced workers with AI assistance. Result: fewer experts needed, lower skill requirements, wage pressure.

Path 2: The Collaboration Track
But when AI handles routine expert tasks, it often makes the remaining work more complex and strategic. Take radiology, which was supposed to be one of the first professions replaced by AI.

In 2016, Geoffrey Hinton famously declared, "We should stop training radiologists now." He was completely wrong. The opposite happened. AI didn't replace radiologists—it made their jobs harder. Collaboration, not replacement. Shortages increased. Wages went up. Why? Partly demographics—aging populations need more scans. But also because AI handles routine scans, radiologists now focus on complex cases, treatment planning, and patient consultation. The profession became more expert-demanding, not less.

The Individual Challenge

Which path you end up on depends on your profession but also on how you navigate AI collaboration. Within the same field, some may become AI-enhanced super-experts while others get relegated to managing "last mile" outputs they don't fully understand.

What determines this split gets lost in much of the noise around AI. Studies showing AI closes the proficiency gap between experts and novices typically measure "mostly good enough" performance on isolated tasks—idea generation, basic analysis, routine task execution. But real professional work requires "always good enough and occasionally perfect" performance in complex, interconnected systems where context matters enormously.

Just as a stethoscope is invaluable to a cardiac surgeon but useless to a tree surgeon—and a chainsaw is essential for a tree surgeon but terrifying in the hands of a cardiac surgeon—AI's value depends entirely on the expertise of the person wielding it. The more you understand the domain, the better you can direct AI's capabilities and catch its limitations.

How Experienced Professionals Can Become More Valuable

The professionals who thrive with AI share three critical abilities that turn decades of experience into competitive advantage rather than liability:

Cognitive Permeability: The ability to use AI without losing yourself in it. Experienced professionals have something novices lack—deep pattern recognition that lets them spot when AI suggestions don't fit the broader context. A senior lawyer can immediately tell when an AI-generated brief misses crucial precedent. A veteran engineer knows when an AI solution violates unwritten design principles. They didn't reject AI nor unthinkingly embrace it. They focused on strategies for having the judgment to direct it effectively.

Identity Coupling: The wisdom to evolve your role rather than defend your tasks. The most successful experienced professionals don't cling to doing everything they used to do. Instead, they leverage their years of experience to move up the value chain. They become the people who know which problems are worth solving, who can spot the edge cases AI misses, and who can translate between technical possibilities and business reality. Those prepared to take accountability for both the end result and the process of decisions.

Symbolic Plasticity: The ability to redefine expertise itself. Experienced professionals have seen multiple technology shifts—they know that value moves, it doesn't disappear. The accountants who survived spreadsheets became financial strategists. The photographers who embraced digital became creative directors. Today's challenge is similar: redefining your expertise around uniquely human capabilities like judgment, relationship-building, and complex problem-solving that becomes more valuable when routine work is automated.

Collaborative Intentionality

The key is developing what we might call “collaborative intentionality”—the ability to work with AI while still owning your thinking and decisions. This requires a new kind of professional discipline. You need to be intentional about cognitive boundaries, deliberately tracking which insights come from your experience versus AI suggestions. You need to be intentional about role evolution, actively choosing which responsibilities to delegate and which to claim as uniquely yours. You need to be intentional about value creation, consciously redefining what expertise means in your field.

The most successful professionals are asking: How can I direct AI to amplify what I already know while preserving what makes my judgment irreplaceable? How do I architect a professional identity that gets stronger when machines can think?

This approach treats the AI transition as a design challenge. You're designing a new form of expertise that combines human judgment with machine capability. You're designing workflows that leverage AI's speed while preserving your contextual understanding. You're designing a professional identity that evolves with technology rather than being displaced by it.

Collaborative intentionality is about staying conscious of your own thinking while working with AI. It’s a new kind of discipline, and the people who practice it will set the standard for expertise.