Almost everyone reading this has touched artificial intelligence by now. Whether it’s an embedded assistant in Google Workspace or a native copilot on your desktop, the raw capability can leave you speechless.
In my last post, we broke down how the abstraction of AI is fueling a massive hype engine, driving mania and billions in capital into bolting "smart" features onto every software tool in existence. But underneath the investment frenzy lies a deeper, historical question about how we organize work—and what happens when disruptive technology matures.
To understand where AI is heading, we have to look back to 1776, when Adam Smith published The Wealth of Nations.
Smith’s foundational concept—the division of labor—argued that breaking production into discrete, specialized tasks exponentially increases output. Over a century later, Henry Ford operationalized this on an unprecedented scale with the moving assembly line, revolutionizing manufacturing through process-driven specialization.
Today, the modern enterprise is built entirely on this premise. We compartmentalize business functions into rigid silos: Marketing, Sales, Operations, Finance, and Engineering.
Yet as AI capabilities surge, a glaring paradox emerges in the labor market: corporate hiring requisitions are still demanding laundry lists of hyper-specific tool certifications, rigid discipline silos, and arbitrary years of platform-specific experience.
The fact that hiring criteria remain hyper-fragmented reveals an uncomfortable truth: organizations are applying a transformational generalist tool to an entrenched, assembly-line mindset.
The issue isn't what AI can do today—it's the predictable lifecycle of tech disruption. When you look at the last forty years of digital innovation, transformative technologies reliably follow a three-act playbook:
Phase 1: Open Access ──► Phase 2: Optimization & Ads ──► Phase 3: Re-Bundling & Paywalls
In the 1990s, print media rushed online, giving away high-value journalism for free to capture digital territory. The result? The near-total cannibalization of print revenue. To survive, publishers had to construct paywalls. What began as the democratization of human knowledge retreated behind subscription gates.
Early search felt like direct access to the world’s index. Over time, search engines mastered monetization. Organic discovery yielded to heavily optimized SEO arbitrage, sponsored placements, and aggressive ad real estate designed to extract revenue before answering your question.
Streaming promised the ultimate consumer disruption: cut the expensive cable cord, eliminate commercial interruptions, and watch whatever you want on demand.
Two decades later, streaming has become the exact beast it killed. Monthly costs are rising, platforms are introducing ad tiers, and services are bundling back together. Cable didn't die; it just changed its delivery protocol.
AI is accelerating through this identical lifecycle at breakneck speed.
We are already seeing early signals of what Act 2 and Act 3 look like for frontier models:
We are racing toward an environment where a tool capable of broad economic enablement risks being channeled into standard corporate rent-seeking.
The promise of AI was universal capability—a tide to lift all generalists and dissolve artificial boundaries between disciplines. But if history teaches us anything, the era of open, low-friction, uncompromised access is always fleeting.
The division of labor isn't going away, and neither is the commercial reality of technology platforms. What we have right now is a brief, unbundled window before the walls go up, the ad engines take over, and the ecosystem settles back into old habits. Those who recognize the cycle for what it is will understand that the real value of a breakthrough tool isn't waiting around for the industry to redefine work for them—it’s realizing that the window to redefine it yourself is already open.