The Death of the Wealth Manager

How the Dark Economy Will Create More Wealth in Server Rooms Than Stock Exchanges Ever Did

Somewhere in Changping, China, a factory produces one smartphone per second.

No humans.

No lights.

No breaks.

One phone every heartbeat of every hour of every day.

This is not a manufacturing curiosity.

This is a preview of how wealth will be created for the next decade.

Most wealth managers still operate as if value originates on trading floors and in quarterly earnings calls. After twenty years building technology systems across Fortune 500 companies, founding 110+ startups, and watching billions evaporate in AI projects that promised everything and delivered PowerPoint, I can tell you: the map of wealth creation is being redrawn. The new territory has no lights.

EY’s 2025 survey reveals that 95% of wealth management firms already use generative AI somewhere. Fewer than a quarter report meaningful business impact. That gap between adoption and results is not a technology problem. It is a comprehension problem.

The firms measuring AI success in efficiency gains are optimizing the wrong variable. The real transformation is structural.

THE LIGHTS GO OUT

We need to talk about what “dark” actually means.

In manufacturing, a dark factory is a facility that operates without human presence. No lighting needed. No heating. No cafeterias. Machines building products in perfect blackness, twenty‑four hours a day. China deployed over one million industrial robots by 2025, more than the rest of the world combined, backed by $1.4 billion in state robotics R&D in 2023 alone. Xiaomi’s Changping facility, BYD’s fully automated battery lines in Shenzhen, Foxconn’s 60,000‑worker replacement at Kunshan: these are not pilot programs. They are the new standard.

Gartner estimates that by 2026, 60% of manufacturers will adopt some form of lights‑out manufacturing. The dark factories market was valued at $119 billion in 2024 and is projected to reach $195 billion by 2030, growing at 8.7% annually.

Now apply this pattern beyond hardware.

A dark startup is a company where one founder orchestrates dozens of AI agents. Product managers synthesizing market signals into specifications while the founder sleeps. Architects proposing system designs for morning review. Developers shipping features before breakfast. QA catching regressions before dawn. Marketing adjusting campaigns based on overnight performance data. One human. Forty‑seven machines. Complete operational continuity.

This is not replacing human judgment. This is decoupling human judgment from human presence.

The dark economy is what emerges when you extend this logic across every sector: factories without workers, startups without employees, portfolios without analysts, compliance without auditors. Value creation that runs continuously, autonomously, in the dark.

CAPITAL WITHOUT LABOR: THE UNCOMFORTABLE MATH

Here is the part nobody in wealth management wants to say out loud.

AI will not democratize wealth. AI will be the single greatest accelerator of wealth concentration in recorded history.

The IMF’s Working Paper 2025/068, published in April, confirms what Piketty theorized and what practitioners have observed: capital income and wealth inequality always increase with AI adoption. Those who already hold capital, particularly in AI‑exposed sectors, are positioned to capture disproportionate returns. The wealth Gini coefficient rises by 2 to 7 percentage points depending on the adoption scenario. Not over decades. Over years.

Piketty’s r > g formula acquires a terrifying new dimension. When a single founder commanding AI agents can produce the output of fifty knowledge workers, the return on capital doesn’t just exceed economic growth. It eclipses it.

Three resources will define who accumulates wealth at this new velocity: land for data centers, access to AI compute, and electricity. Whoever controls these inputs controls the means of production in the dark economy. This is not metaphor. This is industrial economics applied to the age of artificial intelligence.

Global wealth stands at $600 trillion. But as Fortune noted at the close of 2025, asset values have risen far faster than GDP since 2000, and only about a quarter of wealth generated came from actual investment. The rest was paper. AI changes this equation because it creates real output with minimal human input. The wealth it generates is not speculative. It is productive. And it flows to those who own the infrastructure.

THE TEN‑YEAR HORIZON: FOUR PHASES OF THE DARK TRANSITION

What happens between now and 2036 will reshape every portfolio your clients hold. Here is how I see the transition unfolding, based on patterns from hundreds of implementations and the trajectory of manufacturing automation.

Phase One: Augmentation (2025–2027). AI agents handle analysis, reporting, routine compliance, and client communication drafts. Team sizes compress 5x to 10x. McKinsey documents 20% to 45% productivity gains. GitHub’s research shows 55% faster task completion. Firms that treat this as a cost‑cutting exercise will miss the structural shift entirely. The real winners are building orchestration capability: the ability to direct AI systems rather than use AI tools.

Phase Two: Autonomy (2027–2030). Dark startups become the dominant formation for new ventures. 60% of US venture capital in 2025 already flowed to rounds of $100 million or more, because capital concentrates where velocity concentrates. By 2030, PwC projects 45% of global manufacturing will be fully automated. The same compression hits financial services, legal, consulting, and administration. Entire middle‑office functions disappear. Not downsized. Dissolved.

Phase Three: Infrastructure Wars (2030–2033). The battle shifts from software to physics. Land, energy, and compute become the contested resources. Data center capacity determines competitive advantage the way factory capacity determined industrial dominance a century ago. Community opposition has already blocked $64 billion in Big Tech infrastructure projects. The companies that secured energy‑sovereign, modular deployment capacity will own the production layer of the dark economy.

Phase Four: The New Feudalism (2033–2036). Wealth stratification completes. Those who own AI infrastructure, compute access, and energy resources occupy a position analogous to landowners in agrarian economies. The rest participate in an economy where their labor competes with systems that cost a fraction of a salary and never sleep. Oxford Economics projected 12 million Chinese manufacturing jobs absorbed by robots by 2030. The knowledge‑worker equivalent is already underway.

THE WEALTH MANAGER’S EXISTENTIAL QUESTION

Every wealth manager reading this faces a binary choice. Understand the dark economy and translate it into investment strategy for your clients. Or continue allocating between equities and bonds while the largest wealth creation event in history happens in server rooms you’ve never visited.

AI excels at analyzing massive datasets, recognizing patterns, and processing information at speeds no human can match. But AI does not understand the context of your client’s divorce. It does not feel the emotional weight of a family home. It cannot read the sudden shift in priorities after a health diagnosis. MIT’s 2025 research showed that top performers doubled their productivity with AI while the bottom third gained almost nothing. The same principle applies to wealth management: AI amplifies whatever capabilities you already possess. If you lack deep judgment, AI will not supply it.

The surviving wealth manager will be an orchestrator. A human who directs a fleet of AI tools while focusing on what no machine can provide: trust, relationship, context. But orchestration without understanding is automation of ignorance.

Your clients need you to understand that the next decade’s greatest returns will not come from traditional asset classes. They will come from ownership of AI infrastructure, compute capacity, energy access, and the land beneath data centers. 80% of high‑net‑worth clients already prefer a hybrid advisory model. Contracts must explicitly address AI’s role in portfolio decisions, data provenance, model accountability, and human escalation mechanisms.

GOVERNANCE IN THE DARK: WHO ANSWERS WHEN THE ALGORITHM FAILS?

76% of financial firms report efficiency gains from AI. How many have a functioning governance framework for when AI recommends a strategy that harms a client? In my experience: a fraction.

The EU AI Act entered force in August 2024. Since February 2025, bans apply to unacceptably risky AI systems including manipulative technologies and social scoring. By August 2026, key provisions for high‑risk systems take effect. Financial services face mandatory algorithmic transparency, European AI database registration, and safety compliance assessments. Penalties reach 35 million euros or 7% of global turnover.

In the Czech Republic, a national AI law is being prepared that will designate supervisory authorities: the CNB for financial services, CTU for telecommunications, UOOU for data protection. For wealth managers operating in this jurisdiction, this is not an abstract regulatory development. It is an existential compliance question.

Every AI recommendation must be auditable. Every automated trade must have a clear escalation mechanism. Every client must know where the machine ends and the human begins. Firms that ignore these principles are constructing castles on sand while a regulatory tsunami approaches.

THE PARTNER QUESTION: ARCHITECT OR VENDOR?

95% of corporate AI projects fail. Not because the technology malfunctions. Because the organization purchased a product instead of a solution. 60% of AI implementation success depends on integration, not algorithms. I learned this after hundreds of deployments. Firms that buy the best AI tools and deploy them on broken processes get faster chaos.

The global AI wealth management market will reach an estimated $9.8 billion in 2025. In that ocean of opportunity, it is trivially easy to drown in solutions that promise everything and deliver a presentation. Or a functional demo. Your single qualifying question for any technology partner: do you understand our business, or are you selling your software?

THE QUESTION THAT MATTERS

Within the next ten years, the largest fortunes will not be built on stock exchanges. They will be built in facilities that operate without light, by companies that operate without employees, generating returns that flow to those who control three things: land, compute, and energy.

Dark factories already produce ten million smartphones annually with zero human workers. Dark startups already ship software with one founder and forty‑seven AI agents. Dark portfolios are coming. Autonomous systems that monitor, rebalance, and optimize wealth around the clock without human intervention.

The wealth manager who understands this transition will help clients position themselves on the right side of the greatest economic bifurcation since the Industrial Revolution. The one who does not will be running a boutique with a typewriter on a digital highway.

Are you preparing your clients for a world where the greatest wealth arises not on trading floors but in server rooms?

If not, you are preparing them for yesterday.

ABOUT THE AUTHOR

JF is a serial entrepreneur with Fortune 500 executive experience, co‑founder of more than 110 startups, and author of the AI Executive Transformation Program in Prague. He writes at AI Off the Coast about the uncomfortable truths of AI implementation.