The Knowledge Gap Is Closing. Here's What That Means for Your Advisory Team.
AI · Advisory · Integrated Services
The Knowledge Gap Is Closing. Here's What That Means for Your Advisory Team.
In January, the CEO of one of the world's most sophisticated trading firms called AI "garbage" — impressive on the surface, hollow underneath. By May, he had reversed himself completely, telling a Stanford audience that he'd gone home one Friday "fairly depressed" after watching what the technology could actually do. The work his master's- and PhD-level finance professionals would take weeks to complete, he said, was being done by AI agents in hours. His emphasis: these are not mid-tier jobs. This is the top of the skill ladder.
The same week, the CEO of Salesforce noted that almost no one at the company was being hired right now — with one exception. Sales. The most human job in the building.
Put those two data points together and a comfortable story falls apart — the one most of professional services has been telling itself. AI takes the routine work, but real expertise is safe. It isn't. And understanding precisely what is at risk, and what isn't, is the most important strategic question an advisory firm can ask right now.
The part of expertise that can be computed is being commoditized — at every level. The part that can't is the only thing appreciating.
What the Research Actually Says
For two years the consensus was that AI mostly lifted less-experienced workers — closing the gap between novice and expert. The evidence was real: in a large field study of customer-support agents, AI raised output roughly 15% on average, with the biggest gains going to the least-experienced workers and almost none to the most senior. Researchers called it skill compression.
What changed in 2026 is that the frontier moved up the ladder. The newest agentic systems aren't just accelerating junior work; they're absorbing elite analytical work too — the modeling, synthesis, and first-principles analysis that used to be the exclusive domain of the most credentialed people in the room. The honest read today: AI is draining the computable layer of expertise at every tier, not just the bottom.
Notice the through-line. The work that's most exposed is the work that can be specified, retrieved, and computed — at the junior end and the expert end. The work that's holding its value is the work that can't: judgment under genuine uncertainty, and the human relationship that carries it.
What's Actually at Risk — and What Isn't
It's worth being precise, because the panic version of this conversation isn't useful. Deep expertise still matters enormously. But its value is migrating — off the part you can look up or model, and onto the part you exercise. Knowing the answer is becoming cheap. Knowing which question to ask, and being trusted to ask it, is not.
Where the gap still shows up
Business Exit
An owner sells a company and walks into a 1031 conversation with a broker who has no visibility into the executive-benefit structure that could have funded the transition.
Sponsor Portfolio
A financial sponsor builds out a portfolio company and never connects the workers'-comp captive to the refinancing conversation happening in a different room.
Bank CFO
A CFO gets pitched on BOLI by a broker who doesn't grasp the Tier 1 capital concentration limits that make the sizing question critical.
These aren't edge cases. They're the default when specialists work in isolation — and they turn on judgment and cross-domain connection, the exact things AI can surface inputs for but cannot resolve on your behalf.
The Part That's Appreciating
Be direct about what AI does not do. It does not tell you which question to ask. It does not read the room, weigh a client's true risk tolerance, or carry the trust required to sit across the table on the worst day of someone's year. And outside the edge of what it does well, it fails confidently — which means knowing where to rely on it is itself an expert skill.
That Salesforce tell is not a coincidence. The one role still hiring is the one that is nothing but human — reading what isn't said and earning trust. As the computable layer commoditizes at every level, the premium moves to three things no model can disseminate: judgment, cross-domain integration, and trusted presence. That's the asset that appreciates as the tools improve, not the one that erodes.
The moat was never how much you know — or even how senior your analysis is. It's the part that can't be computed.
The integrated model at work
Distressed CRE
Evaluating a significant distressed office opportunity demanded simultaneous fluency in asset underwriting, capital structure, insurance exposure, and market dynamics. That conversation doesn't happen where those disciplines sit in separate rooms with separate incentives.
Sponsor Insurance
We built a corporate-owned life insurance framework for financial sponsors — key-person risk, executive benefits, and tax-advantaged returns from a single structure. The insight lives at the intersection of insurance mechanics and sponsor value-creation logic. No single-discipline firm surfaces it.
Internal Operations
We mapped our own lead-to-cash process — ten workflow stages, a RACI matrix, an AI automation roadmap, a 16-week plan. If we advise clients on operational efficiency, we hold ourselves to the same standard.
Deep Judgment, Connected. That's the Difference.
The advisory firms that win the next decade won't be the ones using AI to simulate expertise they don't have — and they won't be the ones pretending their expertise is immune. They'll be the ones who let AI absorb the computable work, at every level, and redeploy their people onto the part that's appreciating: judgment, integration, and the relationship. Every discipline at Centered Partners is staffed by practitioners who have gone deep in their vertical. What AI does is make those specialists fluent in each other's context in real time — so an insurance implication inside a real estate deal gets surfaced immediately, instead of falling through the cracks between advisors who aren't talking to each other.
There's a practical upside worth naming. The old model billed clients for coordination they should never have had to buy — multiple engagements, multiple timelines, multiple invoices to get one answer that took two disciplines to see clearly. When specialists are integrated and AI connects their context, that overhead disappears. Faster answers, fewer handoffs, one coherent view of the situation.
And a quieter advantage rarely gets named. When disciplines share context, opportunities surface earlier — a capital-markets conversation reveals a distressed asset before it's broadly marketed; an insurance engagement uncovers a transaction two rooms away. Integrated firms generate more proprietary, first-look opportunities for their clients and partners. Not because they work harder. Because they see the whole picture first.
What This Means If You're a Client
The next time you're across from an advisor — any advisor — ask one question: who else is in the room when you're thinking about my situation? If the answer is a referral network, a rolodex of specialists, or a vague promise to loop someone in later, you're experiencing the knowledge gap firsthand.
At Centered Partners, the room is integrated by design — and every person in it has gone deep in their discipline. AI makes that integration faster and more consistent than any team could deliver by hand, and it frees our advisors to spend their time where it actually moves outcomes: understanding your business, anticipating what's next, and being present when it matters. As the tools keep improving, that presence is the last thing that doesn't get commoditized.
One honest caveat. Reasonable people disagree about the pace — some respected analysts argue AI's displacement is still concentrated in narrow tasks rather than whole jobs, and the timing is genuinely uncertain. But the direction is hard to argue with, and we'd rather build for it early than be caught flat.
The knowledge gap is closing — at every level. The human one never will.
Chip Smith
CEO, Centered Partners
Co-founder and former CEO of ATIS, the global leader in elevator inspections, and co-founder of Greenleaf Capital Partners. Chip leads Centered Partners' integrated advisory practice spanning insurance, real estate, capital markets, and M&A.