Why Expertise Has Changed—but Still Matters in the AI Age
Woman and man characters putting light bulb, books and gear into human head to improve work skills. ... More
If you’re an expert, generative AI might feel like an existential threat. If you’re not, it probably feels like a shortcut. Both are wrong.
The truth is more nuanced—and more optimistic. AI doesn’t kill expertise. It simply redefines it. And in doing so, it makes real expertise not only more valuable, but more necessary than ever before.
AI Levels the Playing Field—Until It Doesn’t
There’s no doubt that tools like ChatGPT, Claude, and Perplexity shrink the gap between experts and novices. With a well-crafted prompt, an amateur can now generate content that looks like it came from someone with years of experience. Reports, presentations, code snippets, pitch decks, analyses, even legal memos—they’re all just a prompt away. The performative aspect of expertise—the ability to sound polished, structured, and fluent—has been automated.
But while AI can manufacture fluency, it struggles with depth. It doesn’t really understand what it says. In fact, AI is really good at explaining everything without truly understanding anything (which, alas, has never stopped many humans from mansplaning things, let alone advancing their careers!). AI can hallucinate fact, which is an elegant euphemism for bullshit, and something humans should care about: for many centuries if not millennia, we were pretty much convinced of being the only species capable of BS, but now technology rivals that ability, implying that we are not even masters of our own BS. Still, AI often misses context. And crucially, it can’t tell you whether the answer it gave you is correct—or persuasive, or ethical, or useful.
That’s where the real experts come in. Not to compete with the machine, but to collaborate with it. To interrogate, refine, edit, and add judgment. To know what to ask, what to keep, what to discard, and what to turn into action. In this sense, expertise hasn’t disappeared. It’s been redefined as the ability to get more out of AI than your peers.
If AI Augments Human Intelligence, Human Intelligence Must Augment AI
Most discussions of AI treat it as an end-user tool. But in reality, it is a cognitive amplifier. What it enhances depends entirely on the input. If you’re a brilliant strategist, AI helps you scale faster. If you’re a mediocre thinker, AI just helps you generate more mediocrity.
This is a point I’ve made repeatedly. The real differentiator today isn’t whether you use AI, but how you use it.
The best experts don’t just prompt—they prompt with precision. They vet outputs with skepticism. They edit for nuance. They know when the machine is bluffing (which, let’s face it, is often). In short, they don’t outsource their thinking—they sharpen and broaden it.
If generative AI is a microwave for ideas, then true expertise is the master chef who knows when to use it, when to trust it, and when to turn it off and cook from scratch. Yes, AI speeds things up. But only if you already know where you’re going. OK to defrost the vegetables if you are improvising a recipe… but your creativity, impact, and value are not shown via the microwave!
From Having the Answer to Knowing the Question
Traditionally, expertise was defined by knowing the answer—or at least knowing how to appear as if you did. It was about credentials, confidence, and (in many cases) a bit of intellectual theater. The expert was the person at the front of the room, the one with the bullet points, the frameworks, the final say.
AI breaks that model. When anyone can summon an answer instantly, knowing the answer is no longer impressive. What matters is the ability to:
- Frame the right question
- Spot when an answer is wrong
- Improve what the AI suggests
- Turn bland insights into bold actions
- And perhaps most importantly, convince others to trust your version over theirs
That last point is crucial. Expertise is no longer just epistemological—it’s relational. You need to persuade people not just that you’re right, but that you’re more right than the AI they also consulted. That takes more than knowledge. It takes trust. And trust, today, is built not on IQ but on EQ: emotional intelligence, communication skills, self-awareness, and reputation—topics I’ve long explored, including in "Why Do So Many Incompetent Men Become Leaders (And How to Fix It)".
In other words, the expert of the AI era is not just a brain—but a brand.
The End of the Bullshitter’s Advantage
One of AI’s most useful side effects is that it exposes intellectual impostors. In a world where confident nonsense was once enough to get ahead, the bar has shifted. Anyone who knows just enough to fake it will now be automated into irrelevance. Why listen to a human’s vague, verbose, Wikipedia-derived answer when you can get the same (or better) from a chatbot?
This theme echoes what I’ve argued in "Confidence is Overrated": many people mistake confidence for competence, especially in leadership and consulting. But AI is making it harder to hide. We are entering the era of “knowing enough to know when the machine is dangerous."
AI punishes those who confuse fluency for competence. But it rewards those who bring judgment to the table. Who don’t just repeat what’s known but improve upon it. Who can detect hallucination, correct course, and apply ideas in a way that is actionable, contextual, and human-centered.
In short: AI is a threat to the overconfident amateur—but a gift to the deeply competent expert.
Expertise Is Now About Making AI Safe, Useful, and Persuasive
Let’s be clear: using AI well is not about hiding it. It’s about integrating it so seamlessly into your thinking that people prefer your output—even knowing that AI was involved. Why? Because they trust that you’ve filtered the noise. That you’ve added meaning. That you’ve translated something synthetic into something strategic.
That’s the new function of expertise:
- Not to compete with AI, but to go beyond it
- Not to pretend you don’t use it, but to use it better than everyone else
- Not to hoard information, but to create insight and impact
These ideas are consistent with the emerging talent trends I described in "What AI Reveals About the People Who Use It": the best performers aren’t afraid of AI—they’re better because of how they use it.
This makes deep thinking and deep learning—ironically terms now mostly associated with artificial intelligence—the most important human differentiators of the moment.
Why Now Is the Best Time to Invest in Expertise
We live in an era of infinite content. The ability to produce is no longer impressive. The ability to discern, distill, and decide is.
This is why I believe we are entering a golden age of real expertise. Not the performance of knowledge, but the practice of it. Not the regurgitation of answers, but the thoughtful application of questions.
Yes, AI changes everything. But it also re-centers something timeless: judgment. And judgment is not trained overnight. It is built over years—through knowledge, experience, reflection, and (yes) mistakes. I’ve seen this repeatedly in leadership science and talent analytics, as outlined in my book The Talent Delusion, which argues that potential is about more than performance—it’s about curiosity, adaptability, and learning agility.
The future belongs to those who are not just curious, but credible. Not just fast, but thoughtful. Not just AI users, but AI integrators.
If you’re an expert, don’t panic. If you’re not, don’t pretend.
Because now more than ever, deep expertise is the ultimate competitive advantage.