The AI Profits Drought: What Trustees and Family Offices Must Learn from History

In 2025, the numbers are sobering. Corporations have invested between $30–$40 billion into generative AI, yet 95% of organizations report no measurable return. Only a small fraction—about 5%—have managed to move beyond pilot projects to capture real financial or productivity gains.

The paradox is striking: while nearly half of all workers (45.6%) already use AI tools in their daily workflows, enterprise-scale initiatives remain experimental, siloed, and under-integrated. What is driving this disconnect? And what can trustees and family offices learn before they repeat the same costly mistakes?

The Illusion of Immediate Payoff

Executives who once brimmed with confidence in their AI strategies are now more uncertain. In just one year, the number of “very confident” C-suite leaders dropped from 82% in 2024 to 49% in 2025. This collapse in conviction is not due to lack of belief in AI’s potential, but to the painful realization that general-purpose technologies rarely deliver instant returns.

Economist Robert Solow captured this dynamic decades ago: “You can see the computer age everywhere but in the productivity statistics.” AI may be today’s version of that paradox. Steam, electricity, and computing all required decades of complementary infrastructure, skills, and organizational change before their full productivity benefits were realized. AI is no different.

The AI Profits Drought_ issues.jpeg

fiduciacorp.com

The J-Curve of Transformation

AI adoption is not a straight line. It follows a J-curve: an initial phase of disruption and underperformance, followed by long-term productivity gains for those who endure.

Firm-level Census data from U.S. manufacturing reinforces this pattern: companies that invested in AI saw initial dips in efficiency, as processes were restructured, workers retrained, and governance frameworks established. Yet those same firms later pulled ahead, capturing compounding efficiency advantages once their foundations were stable.

This insight is vital for trustees and family offices. The early stages of AI adoption are not about maximizing ROI—they are about building resilience, capability, and alignment so that long-term benefits can be harvested.

The AI Profits Drought_ operations.jpeg

fiduciacorp.com

Where Real Value Is Emerging

Despite widespread disappointment, not all AI projects are faltering. The most successful deployments are targeted, customized applications that improve operational efficiency without requiring wholesale organizational redesign.

In corporate settings, this means back-office automation and customer service augmentation—areas where AI systems are bounded, low-risk, and directly measurable. For trustees and family offices, it translates into:

  • Automating compliance workflows: AI tools that process regulatory reporting with greater accuracy and speed.

  • Enhancing client stewardship: AI-driven personalization to anticipate beneficiary needs and deepen trust relationships.

  • Streamlining operations: Document review, transaction reconciliation, and onboarding—all accelerated by AI without changing the fiduciary core.

These are not speculative. They are low-friction entry points that build capability while proving AI’s worth.

Interestingly, the most tangible productivity gains are often outside formal enterprise initiatives. Workers using personal AI tools—like GPT, Claude, or other assistants—are reporting higher efficiency than corporate projects designed to scale.

This shadow AI economy is a lesson in humility: adoption is not always top-down. Trustees and family offices should recognize and support the quiet, ground-level uses of AI by professionals who integrate it organically into their work. Instead of suppressing these experiments, governance frameworks can bring them into alignment with institutional security and compliance.

A Historical Perspective for Trustees

Trustees and family offices have seen this movie before. When electricity arrived, it did not immediately boost factory output. Firms needed to redesign entire floor layouts to take advantage of distributed power. When computing emerged, it took decades before enterprise resource planning (ERP) systems unlocked efficiency.

AI is not a plug-and-play upgrade—it is a structural shift that demands new operating models, talent, and governance. For fiduciary institutions, the lesson is not to chase quick returns, but to anchor AI adoption in long-term stewardship.

The AI Profits Drought_ What Trustees and Family Offices Must Learn from History.jpegThe AI Profits Drought_ What Trustees and Family Offices Must Learn from History.jpeg

fiduciacorp.com

The Path Forward: Silent Compounding

For trustees and family offices, the opportunity is quiet but immense. The path is not chasing headlines or speculative moonshots. It is:

  • Starting where ROI is clear: projects where value can be measured quickly and directly—compliance monitoring, onboarding automation, or document intelligence. These prove AI’s worth and build confidence internally.

  • Recognizing that AI implementation is not IT implementation: AI is not another system to be installed; it is an operating model shift. Unlike IT upgrades, AI touches governance, workflows, and fiduciary judgment.

  • Choosing the right organizational model from the AI Trustee Playbook:

    • Center of Excellence (CoE) – a centralized hub where AI expertise is concentrated, ensuring governance and consistency.

    • Hub-and-Spoke – specialized AI support in a central team, but execution distributed across business units.

    • Decentralized – each business unit empowered to adopt AI directly, with governance frameworks in place.

Each model balances speed, oversight, and specialization differently. The choice depends on maturity, scale, and culture.

The firms who select the right model, start with projects where ROI is tangible, and treat AI as a structural capability—not a tool—will be those who convert today’s drought into tomorrow’s structural advantage.

The drought is temporary. The compounding, inevitable.

FiduciaCorp: “Mastering AI, Empowering Wealth”

Frédéric Sanz

With over 20 years of elite financial expertise in Switzerland, I specialize in managing UHNWIs assets, leading high-performing teams, and driving innovation in wealth management. As a TEP, MSc., MAS, and Executive MBA with AI diplomas from MIT and Kellogg, I combine deep technical knowledge with strategic leadership for business growth.

A blockchain specialist, I deliver exceptional revenue growth while elevating client satisfaction. Fluent in Spanish, French, Italian, and English, I offer a global perspective, blending advanced AI-driven strategies with traditional wealth management.

Previous
Previous

The Strategic Center of Excellence: How Banks Secure AI Sovereignty

Next
Next

ISO 42001 and the EU AI Act: A Dual Architecture