At a glance: the value has moved from generating ideas to deciding which ones matter
AI has driven the cost of producing ideas, drafts and analyses close to zero. The expensive part of professional work is no longer making things.
It is deciding which of the options now in front of you is worth acting on, which to discard, and why.
That decision draws on exactly what years of practice build: a feel for how a similar call played out last time, and the confidence to make it.
Two forces are pushing that value up. The more AI produces, the more there is to sort through, and the apprenticeship path that once trained the next generation of people who can do the sorting is being cut off.
Your accumulated experience is not only valuable. It is becoming harder to replace.
The question is no longer whether we can produce this.
It is whether we should act on it. That question draws on exactly what you have spent decades building.
The cost of producing ideas, drafts, analyses and options has collapsed. AI tools turn out in minutes what used to take days. None of that output is worth anything until someone who knows the terrain decides what to act on and what to throw away. That decision, the exercise of professional judgment, is where experienced professionals now earn their keep in the AI era.
The bottleneck in professional work has moved from generating ideas to working out which ones matter.
What changed and why it matters
For most of the last thirty years, the expensive part of professional work was production.
- Researching a market took weeks.
- Drafting a proposal took days.
- Building a competitor analysis meant gathering data, sorting it, and assembling something coherent.
Whoever could do that faster or more thoroughly had the advantage.
AI has compressed that production cost to almost nothing. A capable model produces a competitor analysis in twenty minutes that used to take three days. It generates a dozen options before lunch. It drafts, summarises, researches and reorganises faster than any person can.
The brand strategist Paul Worthington captures the effect well. AI, he writes, is removing the costume of execution that disguised its absence. When producing the work was slow and expensive, doing it at all looked like competence. Now that anyone can produce it, the producing proves nothing. What is left on show is the quality of the choices behind it.
The natural response is to feel exposed. If AI does the work faster and cheaper, what is left for you?
What is left is the part that was always worth the most, and is now the only part that carries value: deciding what is actually worth doing.
The shift. Production, which is research, drafting, data analysis and generating options, is now fast and cheap. Verification, which is weighing the options, discarding most of them, and deciding which one fits this situation, is where the value sits.
AI handles the first. You do the second.
Why generating ideas was never the hard part
Consider what happens when a consultant walks into a client’s office. The client does not need more ideas. They usually have too many. Their inbox is full of proposals. Their own AI tools have generated a dozen plans. What they need is someone who can look at all of it and say: this one is right, these three are wrong, and here is why.
That call does not come from processing speed, but comes from having worked through many similar situations across different organisations and industries. It comes from recognising a plan that is going to fail before the data confirms it.
It comes from a feel for the situation that only builds through years of repeated practice.
The mathematician Terence Tao described the same shift from inside his own field. AI has made generating options essentially free, so the constraint has moved to verification. The question is no longer whether you can produce something, but whether it is correct and worth pursuing.
David Bessis, a mathematician who now works in AI, gives that human contribution a precise name. A first attempt that is wrong can still be directionally correct, pointed the right way and needing repair.
Andrew Wiles’s first proof of Fermat’s Last Theorem was wrong, and he repaired it into a valid one. A machine cannot tell you whether a wrong first draft is pointed the right way. Knowing which way to push it is the work, and it is the work you have been training for decades without calling it that.
Production is fast and cheap; the bottleneck is deciding what to act on.
The verification gap in practice
This shows up in specific, observable ways, and the Wisepreneurs Podcast is full of people describing it.
Jon Younger built and advised businesses across a long career, and he is blunt about where the value has gone.
You cannot sell knowledge any more, he says, but you can sell wisdom, insight, a deep understanding of something.
Knowledge is now commoditised, available through any model. What a client pays Jon for is the question only decades of experience produce–the one that reframes what they thought their problem was.
Cedric Chin, who studies how experienced operators actually make decisions, is sharper still about the risk of skipping this step. He points to what he calls the Vaughn Tan rule: do not hand your value judgments to an AI.
A model can generate the options and lay out the trade-offs. Deciding which option is right for this client, given what they have told you and the things they have not, is the part you cannot outsource without losing the thing you are being paid for.
Anna Burgess Yang has automated most of the production in her solo practice. AI drafts her posts from her articles, and workflows publish and schedule them without her. Her own test for the automation is telling.
Is it right a hundred percent of the time? Probably not. Is it right? 90% of the time? Probably.
And that’s good enough for what I’m doing and for the time saved and not having to think about it.
Anna Burgess Yang
It is good enough because she is the one reading the output and catching the tenth that is wrong. The machine produces. She decides what ships.
Melisa Liberman coaches consultants on winning work, and her method is verification made visible. Instead of pitching, she runs a diagnostic conversation, asking the questions that help a client see the blind spot they had missed.
The quality of the questions shows what she will be like to work with before any proposal exists.
In every case the AI handles production. The professional does the verification. The verification is where the money is.
The model generates the options. You decide which one fits. The machine drafts the plan. You see why it will fail. That is not a slower version of what AI does.
It is the part AI cannot do.
Why this advantage is growing, not shrinking
Two forces are compounding in favour of experienced professionals.
First, the more AI produces, the harder verification gets. When options were scarce and expensive, the constraint was making them. When options are abundant and cheap, the constraint becomes sorting them: reading a pile of plausible-looking material and knowing which one is actually right.
The writer Dror Poleg makes the point that as everyone gets access to abundant output, the valuable person is not the one who generates more of it, but the one who generates more specific output, the version that fits this situation and no other. There is more to verify now, not less.
As AI output rises, demand to verify it climbs while the supply of people who can narrows
Second, the path that produces people who can do this verification is narrowing. When organisations automate their entry-level tasks, they remove the way the next generation learned. The junior who used to do the research, make the mistakes, and watch a senior practitioner weigh the results is now being replaced by AI doing the research directly.
The apprenticeship that built this capability over decades is quietly being cut. Richard and Daniel Susskind put the consequence plainly in The Future of the Professions: the finest experts are a very scarce resource. That scarcity is about to sharpen.
So your accumulated experience is not only valuable. It is getting harder to replace. Demand for verification is rising while the supply of people who can do it is being constrained.
There is a growth version of this too, and it matters for anyone who worries the ground is shifting under them. Bessis argues that the feel for a situation is not a fixed gift you either have or do not. It is trainable, and it keeps improving when you test it against something external and correct it when it is wrong.
For an experienced professional, AI is now that external thing to test against. Used that way, your experience does not just resist being automated. It compounds, because every round of checking the machine’s output against what you know sharpens the very capability the machine cannot reproduce. The condition is that you stay in the loop. Hand the checking over as well, and you wear away the thing you are selling.
The compounding advantage. Demand for verification is rising, because more AI output means more to weigh. The supply of people who can do it is narrowing, because the apprenticeship path is being cut. Your experience sits where rising demand meets shrinking supply.
What this means for how you position your practice
If the constraint has moved from production to verification, your positioning should show it. Three practical implications follow.
- Lead with how you think, not the steps you follow. The steps are commoditised. The thinking is not. When you describe your work, describe a moment of diagnosis: the pattern you recognised, the assumption you challenged, the wrong turn you helped a client avoid.
- Use AI for what it does well, and protect what it cannot do. Let it handle research, drafting and generating options, where speed matters and the decision does not. Keep your own hands on the substance of the client’s problem. If you start accepting AI output without weighing it yourself, you slowly lose the capability that makes you worth hiring. The professionals who stay sharp use AI to test their thinking, not to replace it.
- Publish your thinking. Every time you write about how you approach a problem, you show your market what you do that AI cannot. Articles, guides and diagnostic tools all demonstrate that you work at the deciding layer, not the producing layer.
The real question your clients are asking
When a prospective client weighs up hiring you, they are no longer asking whether you can produce the work. AI made producing it cheap. The question they are asking, whether they say it aloud or not, is whether you can tell them what is actually worth doing.
That is answered by years of practice, not processing speed. It is answered by the situations you have seen play out, the problems you have tracked to their real cause, and the calls you have made and watched land.
The verification bottleneck is not a threat to experienced professionals. It is the clearest case for their value the market has produced in a generation.
Frequently asked questions
Why does experience become more valuable as AI improves?
AI generates more options, drafts and analyses than ever, but none of that output checks itself. As the volume rises, so does the demand for someone who can read it and decide which version is right for a specific client. The scarce resource in an economy of abundant information is the person who can tell which output fits, drawing on the crystallised intelligence that decades of practice build.
How can experienced professionals position accumulated expertise in the AI era?
Lead with how you think, not the steps you follow. Show a moment of diagnosis: the pattern you saw, the assumption you challenged, the wrong turn you helped a client avoid. Use AI for research and drafting while keeping your own hands on the client’s actual problem.
What does the verification bottleneck mean for pricing?
When your value sits in deciding what is worth doing, pricing reflects the quality of that decision rather than hours worked. A consultant who names the core issue in a first meeting and prevents six months of misdirection has done the expensive part. That commands a premium because it reduces the client’s cost of understanding their own situation.
References
Podcast episodes
Jon Younger Professional Relevance: Positioning Expertise When AI Does the Routine Work. Wisepreneurs Project podcast, Episode 79, on where value sits once AI does the routine work.
Cedric Chin Testing Business Ideas Against Evidence: Pattern Recognition for Independent Professionals. Wisepreneurs Project podcast, Episode 58, on pattern recognition and not handing over your value judgments.
Anna Burgess Yang Business Automation for Independent Solopreneur Practice. Wisepreneurs Project podcast, Episode 77, on automating production while staying the one who decides what ships.
Anna Burgess Yang Solopreneur Workflow: AI for Independent Professionals. Wisepreneurs Project podcast, Episode 65, on building the workflow behind the practice.
Melisa Liberman Solopreneur Business Development for Experienced Consultants. Wisepreneurs Project podcast, Episode 82, on the diagnostic conversation that replaces the pitch.
Other sources
Terence Tao. Dwarkesh Podcast interview on AI and mathematical discovery (March 2026). His observation that AI has driven the cost of generating options to almost zero, moving the constraint to verification, grounds this article’s central argument.
David Bessis. Mathematica: A Secret World of Intuition and Curiosity, and his EconTalk interview (September 2025). Source for the idea of a wrong first attempt being directionally correct, and for the argument that the feel for a situation is trainable.
Dror Poleg. Rise of the 10X Class. Source for the point that value shifts to more specific output, not more output, as abundance rises.
Richard and Daniel Susskind. The Future of the Professions. Source for the observation that the finest experts are a very scarce resource.
Paul Worthington. Off Kilter, "Swimming Naked" (July 2026). Source for the line that AI removes the costume of execution that disguised its absence.
Cedric Chin. Commoncog, on using AI without eroding your own thinking. Source for the Vaughn Tan rule on not outsourcing value judgments.
Every Tuesday I send The Wisepreneur, a short letter for experienced professionals building independent practice: one usable idea a week. Sign up at wisepreneurs.com.au/newsletter
When you want a second pair of eyes on your own practice, the Positioning and Practice Audit is how I work with people: wisepreneurs.com.au/marketing-partnership