Most AI strategies are decoration, not transformation

The commentary opens with a diagnosis: most companies believe they have adopted AI, but what they have actually done is add AI licenses, run pilots, hire a chief AI officer, and create governance structures while leaving the underlying work unchanged. Salespeople still enter data into CRM systems. Managers still spend half their weeks gathering information from one team and passing it to another. Customers still wait in approval chains. The author calls this the "AI Shuffle"—swapping one technology logo for another while preserving every assumption about how work gets done. This creates activity but produces no advantage.

The author proposes a different starting point: What work should no longer exist? Rather than asking how AI makes a process 10% faster or which chatbot to license, companies should identify what decisions can move closer to the customer, what information no longer needs to be collected and passed up a chain before anyone acts on it, and what steps can be eliminated entirely. This distinction between adding AI to a company and becoming an AI-first company is central to the piece. The first path produces announcements; the second produces remade institutions.

The article illustrates this with sales forecasting. For decades, companies have asked individual sellers to enter projections into CRM, managers to interpret them, and leadership to negotiate a number everyone knows is partly theater. The data arrives late, incomplete, and distorted by incentives. The AI-era alternative is not a more elegant meeting but a system that analyzes the buyer's actual behavior, market conditions, timing, relationships, and signals directly. The goal is not to make the old ritual more efficient but to make it unnecessary. The author contends that companies winning with AI measure constraint movement, not tool usage.

The article then broadens its argument beyond software to organizational structure. Much of the traditional enterprise software stack was built to organize human data entry—storing records, routing tasks, generating reports, and reconstructing what happened after the fact. AI agents will increasingly observe activity directly, maintain context, initiate work, and recommend or execute action. But the management structures built around information brokerage are also at risk. The people who will matter most are builders—those who understand a real business problem, can use technology to solve it, and are close enough to customers and operations to know whether the solution works. The people who lose relevance will be those whose role depends on preserving friction, controlling information access, or managing processes no one would design from scratch today. The author cautions that too many companies have placed their AI future in the hands of people selected to prevent mistakes rather than create capabilities, and argues that responsible deployment does not require paralyzing every low-risk experiment.

On the competitive landscape, the author argues the real debate is not at the model layer—whose benchmark is best, which training run is largest—but at the orchestration layer above it. Models will improve and proliferate; capabilities that once seemed exclusive will become available to more companies. The durable advantage will not come from having access to a model everyone else can rent. It will come from proprietary context, trusted relationships, operating data, and feedback from real decisions. This is where lock-in lives, not in a prompt or interface. The author frames this as a learning system—a company's ability to connect proprietary data, relationships, and market feedback in ways competitors cannot easily replicate.

The article concludes by broadening the institutional question beyond corporate workflow to systemic friction across housing, infrastructure, energy, capital formation, and communications—all designed in a pre-AI world where collecting and interpreting information was slow, expensive, and centralized. The author argues that permitting reform and AI infrastructure investment deserve serious attention beyond bubble-versus-substance debates. The final choice before leaders is whether to use AI to preserve yesterday's institution or to build the company that should have existed all along—one that sees more, learns faster, acts closer to the customer, and spends less time administering work. The author warns that the window to choose is open now but will not remain open indefinitely.