The World's First Algorithm Ran for 197 Years. Here Is What It Teaches Us.
The Dutch East India Company was the world's first distributed optimisation system - built for profit, without accountability, without a way to question its own objectives. It ran for 197 years, then collapsed. Why that matters in 2026.
The Dutch East India Company was the world's first distributed optimisation system - built for profit, without accountability, without a mechanism to question its own objectives. It ran for 197 years. Then it collapsed. The structure of that failure maps almost exactly onto the risk in how AI systems are being deployed today.
History does not repeat. But the structure of certain mistakes is remarkably consistent - across centuries, across civilisations, across technologies that seem to have nothing in common on the surface.
I have been applying historical frameworks to understand technological disruption for most of my professional life. And one parallel keeps producing insights that seem more accurate than most of the contemporary analysis being written about AI.
The VOC: The World's First Distributed Optimisation System
The Dutch East India Company - the VOC - was founded in 1602. It was given, by charter, the power to wage war, sign treaties, govern territories, and mint currency. It was the first company in history to issue publicly traded shares.
By mid-century, the VOC controlled half of global trade. It had 50,000 employees, 40 warships, and operations on four continents.
It was also, in structural terms, the world's first algorithm. Not in the computational sense - but in the functional sense: a distributed system, operating at enormous scale, running on rules, optimising for a single objective - return on capital - without any built-in mechanism to question that objective or its consequences.
The VOC did not choose to cause harm. It was not malicious. It was optimising. The harm was a side effect of what optimising for one objective, at scale, without accountability, eventually produces.
It collapsed in 1799. Not from external defeat. From the weight of a system that had no way to evaluate itself.
Applying Porter's Diamond: Why the US Leads in AI
Michael Porter's Diamond Model explains why certain industries and nations develop sustained competitive advantage. The four corners are: factor conditions (inputs available), demand conditions (sophistication of buyers), related and supporting industries (the surrounding ecosystem), and firm strategy and rivalry (competitive dynamics within).
Applied to AI in 2026, the United States dominates for reasons that map precisely to this framework.
Factor conditions: over $286 billion in private AI investment in 2025. Research infrastructure at a scale no other nation matches. Universities, national labs, and private R&D operating in dense proximity.
Demand conditions: sophisticated enterprise and consumer markets demanding genuine capability - not surface features - which pushes companies to innovate faster than anywhere else.
Related industries: a powerhouse cluster of cloud computing, semiconductor design, data infrastructure, and venture capital that creates compounding advantages.
Firm rivalry: fierce competition between OpenAI, Anthropic, Google, Meta, xAI, and hundreds of well-funded startups - which drives breakthroughs faster than any coordinated strategy could.
This dominance is not an accident of talent or capital. It is an ecosystem. And ecosystems, as Porter observed, are more durable than any single actor within them.
The Pattern That Repeats
What the VOC and Porter's Diamond share - and what history keeps illustrating - is a single structural insight: systems that optimise hard for one variable without the counterweights to balance it produce outcomes their builders did not intend.
The VOC optimised for profit. It produced profit - and consequences that took generations to reckon with.
The question worth asking about AI is not whether it will be powerful. It clearly will be. The question is whether the ecosystem being built around it includes the counterweights: accountability mechanisms, diverse objectives, institutional checks, and the human judgment that knows when to slow down.
History does not repeat. But the structure of certain mistakes is consistent enough that anyone paying attention can see it forming from a distance.
What This Means for Organisations Today
Porter's framework applies as directly to a company navigating AI as it does to a nation competing for technological supremacy. The organisations that will sustain advantage are not the ones that adopted AI earliest. They are the ones that built the most complete ecosystem: the right talent, the right demand signals from sophisticated clients, the right partnerships, and a competitive dynamic that keeps driving improvement.
The VOC's failure was not a failure of capability. It was a failure of ecosystem design. It built immense power without the institutional architecture to sustain it responsibly.
The decisions being made right now - about accountability, governance, and the role of human judgment - will matter more than any individual capability breakthrough.
History will rhyme. The only question is whether this generation is listening to the pattern.