Beyond automation. Toward organizational intelligence.
Notes by Estefanía MolinaStrategy consultant, New YorkIn personal capacity
I write about how ideas actually become value - where strategy meets execution, and where systems either hold or quietly break.
Most of what I publish sits at the intersection of AI, operating models, and enterprise strategy. Not in the abstract, but in how these things behave in real environments: how decisions get made, how incentives shape outcomes, and why well-instrumented systems still miss what matters.
This is not commentary on AI tools, models, platforms, or vendor stacks. My focus is the organizational and strategic layer underneath - where the design work that determines whether AI generates enterprise value actually lives.
Read full AboutThe most important work in AI strategy has nothing to do with AI. It is the bidirectional design pass - top-down outcomes and constraint hierarchies meeting bottom-up decision architecture - that happens before any model is selected. Two strategic payoffs emerge from the same act.
Most AI investment is framed as differentiation. I keep returning to the possibility that it is also accelerating convergence - at the level of organizational reasoning itself. The long-term differentiator may not be who deploys AI fastest, but who develops the institutional capacity to think coherently when execution no longer differentiates.
We keep debating AI ROI as if we know what we're measuring. We don't. Enterprises are funding "AI initiatives" without a stable definition of the object they're funding - and every measurement debate downstream is incoherent until the unit is fixed.
Every enterprise has two org charts: one is a document, one is an agreement. The visible functional org chart has been refined for decades. The decision org chart - who actually controls each trade-off, who absorbs the consequence - has never been designed. AI is exposing the gap.
Enterprises are moving toward the top-right of the AI matrix whether they choose to or not. Two curves rise together - value potential and structural pressure. The danger zone is the gap between them. The matrix is a trajectory, not a snapshot.
AI fails to scale at the enterprise level not because of model quality, but because decisions are not designed to work together. The constraint is not capability - it is the absence of a system that coordinates decisions.
Get new essays in your inbox. No commercial layer, no marketing - just the work as it is published. Unsubscribe anytime.
By subscribing, you receive new essays only. Your email is not shared, sold, or used for outreach.