The most powerful person in a business right now is not the AI expert.
It is the person who really understands the business, now using AI.
I want to be careful here because the framing matters. AI experts are impressive. The work they do is real. The technical depth required to build, fine-tune, and deploy AI systems is genuinely hard. None of what I am about to say is a knock on AI experts. The argument is about leverage, not value.
Who has the real AI superpower?
The real AI superpower belongs to the person who already understands the business and has learned to use AI on top of that knowledge. AI alone is powerful, but AI plus operational judgment is the combination that changes the output.
The leverage in any organization right now sits with the person who already understands the business better than anyone else. The operator. The veteran. The agent who has closed a thousand transactions and knows where the friction lives. The senior nurse who has seen ten thousand patient interactions and knows which signals matter. The marketing leader who has run hundreds of campaigns and knows which framings work in which markets. That person, equipped with AI, is the most powerful person in the organization.
Picture them with AI. Now you have a superpower.
Why does domain expertise change the AI equation?
Domain expertise changes the AI equation because AI recommendations still need context, judgment, and field knowledge to become useful decisions. The person closest to the work knows what the model output means, what it misses, and when it should be trusted.
This is not a hot take. It is rapidly becoming the consensus position among the most rigorous researchers studying how AI delivers value inside companies.
Prasanna Tambe, Professor at the Wharton School and Faculty Co-Director of Wharton Human-AI Research, recently published research in Management Science titled Reskilling the Workforce for AI: Domain Expertise and Algorithmic Literacy. The paper argues that firms capture more value from AI when algorithmic expertise is distributed across domain experts rather than concentrated in IT departments. Tambe puts it directly in his summary of the work. The future belongs to workers who are bilingual, fluent in their field and fluent in AI.
That is the entire thesis. AI plus domain expertise creates a kind of leverage that AI alone cannot create and domain expertise alone cannot create.
What does Wharton research show about AI and domain experts?
Wharton research shows that companies create more AI value when algorithmic skills spread into business-facing roles instead of staying isolated inside IT. The market is asking for people who can combine technical literacy with deep knowledge of the work itself.
The data Tambe pulled from is serious. He analyzed two large-scale datasets, job postings from Lightcast covering 2013 through 2016 and workforce profiles from Revelio Labs covering 2008 through 2021, linked to financial data from Compustat-Capital IQ to study how the market actually values these workforce shifts. The findings are direct.
By 2016, only one-third of job postings requiring algorithmic expertise were for IT roles. Two-thirds were for business-facing jobs that combined technical literacy with domain knowledge. That ratio has continued to shift in the same direction since. The market wants people who can do both.
It gets sharper than that. Tambe's research found that firms that distribute AI expertise to domain experts who already have decision-making power create more value than firms that keep AI specialists locked in a centralized team. Markets reward the decentralized approach with higher valuations. As he frames it, competitors can buy the same AI software, but they cannot easily replicate a workforce where decision-makers are already trained to integrate algorithms into their judgment.
That is the moat. The AI itself is not the moat. The combination of AI and people who deeply understand the business is the moat.
How does this apply to real estate agents?
In real estate, the strongest agent will not be the person with AI tools alone. It will be the agent with years of local market knowledge, transaction judgment, relationships, and pattern recognition who now uses AI to compress the work that used to consume the week.
I want to translate this into the language of real estate, because that is the industry I am operating in every day.
The most powerful real estate agent in the next five years is not going to be the agent with the most advanced AI tools. It is going to be the agent who has 20 years of market knowledge, who knows their neighborhood at the parcel level, who has seen every shape a transaction can take, and who is using AI on top of that knowledge to compress the work that used to consume their week. That agent is going to be operating at a level no AI-only competitor can match, because the AI-only competitor does not have the market intuition that makes the AI useful.
The same logic applies to brokerages. The brokerage that wins the next decade is not the one with the best technology stack on its own. It is the one whose senior agents and operators have learned to integrate AI into their existing operational excellence. The technology amplifies what they already know. A brokerage that buys the same technology but does not have the operational depth to use it will never get the same compounding effect.
Why are veteran operators positioned to win with AI?
Veteran operators are positioned to win because their experience is the rare input. AI access is cheap and widely available, but two decades of judgment, relationships, market reads, and pattern recognition are not.
Think about what that means for the agents who have been around for 15 or 20 years and who feel threatened by the rise of AI. The framing they have been hearing is wrong. They are not the people most at risk in this transition. They are the people most positioned to win it. The expertise they have built over two decades is the rarest and most valuable input into the AI equation. Anyone can buy AI access for fifty dollars a month. Almost nobody can buy two decades of relationships, market reads, and pattern recognition. The veteran operator with AI access has both. The newer operator with AI access has only one of the two.
The same is true for senior people in title, escrow, lending, and every other corner of the real estate transaction. The professional with deep operational knowledge plus AI is going to outperform the professional with AI alone, every time, in every market.
Tambe makes the point even more cleanly than I can. His framing is that AI is becoming like Excel. It is no longer just for specialists. It is a basic productivity tool across many jobs. The professionals who learn to use it are going to outperform the professionals who do not, regardless of the level of expertise they bring on the AI side. The expertise that matters is the operational expertise. The AI is the leverage on top.
This reframes who has power inside a real estate organization, or any organization, right now.
The AI expert who joined the company three months ago and does not know the business is not the most powerful person in the building. The senior operator who has been running transactions for two decades, who is now teaching themselves to use AI tools on real work, is the most powerful person in the building. The first person has tools without context. The second person has context with tools. The math always favors the second person.
Should companies centralize AI in one team?
Companies should not treat AI as a single centralized specialist function if they want broad operational value. AI needs to live close to the work, in the hands of the people making the operational decisions every day.
This is also why the worst possible move an organization can make right now is to centralize all of its AI capability inside a single team. That structure assumes AI is a specialist function, like cybersecurity or compliance. It is not. AI is a horizontal capability that needs to live closer to the work, in the hands of the people who make the operational decisions every day.
Tambe's research found that the firms that distribute this capability the fastest, with no-code and low-code tools that domain experts can adopt without programming knowledge, see the highest adoption rates and the strongest financial outcomes. The friction of having to go through a centralized team kills the adoption curve. The acceleration of giving domain experts direct access to AI tools is what creates the real upside.
What should leaders do next?
Leaders should identify the people with the deepest operational understanding and give them direct access to AI tools. The goal is not to replace their judgment, but to amplify it where documents, communication, pattern recognition, and decision support used to slow them down.
Here is what I would tell anyone reading this, regardless of industry.
Look around your organization. Identify the three or four people who understand the business at the deepest level. The ones who have been there for years, who can tell you why the workflow is the way it is, who know which clients matter and which ones do not, who can read a deal in five minutes that most people cannot read in five hours. Those are your most valuable people right now. Not because of seniority. Because of leverage.
Now picture each of those people becoming fluent in AI tools that can compress, augment, and amplify the work they already do. Picture them running their workflow with the AI doing the heavy lifting on document analysis, communication drafting, pattern recognition, and decision support, while their judgment and relationships and operational knowledge stay at the center of every call they make.
That is what a superpower looks like in 2026.
It is not the AI expert who has never closed a transaction. It is the operator who has closed a thousand transactions and now has AI doing the parts of the job that used to swallow their week.
Real estate, like every other industry that runs on operational expertise, has not yet figured out how to fully unlock this combination. Most agents are still treating AI as something separate from their work. Most brokerages are still treating AI as a technology project owned by a small team. The agents and the brokerages that figure out how to integrate AI into the operational core, with the senior operators driving the integration, are going to define the next decade of this industry.
The most powerful person in your business right now is the person who already understands the business better than anyone else. Add AI to that person. Now you have a superpower.
The technology is not the variable. The judgment is. Find the people in your organization who have the deepest operational understanding, give them direct access to the tools, and let them lead the integration. That is the play. There is no other play that produces the same outcome.
Judd Hoffman is CEO and Co-Founder of Ethica AI, building AI-powered tools for real estate transaction workflows.