What Forty Years Inside Service Businesses Taught Me About AI.
Created on 2026-04-03 14:42
Published on 2026-04-03 16:30
I spent four decades inside service businesses, carrying full operational responsibility rather than advising from the outside. For twenty three of those years, I owned and operated a private club in New Canaan, Connecticut, building it from acquisition to $3.5 million in annual revenue while sustaining 25 percent net operating income across multiple market cycles. The work required decisions across operations, finance, staffing, systems, and customer experience where results reflected those decisions immediately. Over time, the specifics of the industry mattered less than the underlying pattern that determined whether things held or broke.
Across hundreds of operating situations, that pattern proved consistent. When intent and execution were aligned, performance held. When they were not, problems surfaced quickly and tended to repeat regardless of the surface issue. Efforts focused on symptoms created movement without resolution, while work done at the foundation corrected the problem in a durable way. That observation became the basis for the BASES framework, built on the premise that most business problems originate at the level of foundation before they appear in day to day execution.
I saw the same pattern again when AI tools began to move into real operational use. The systems were capable of executing tasks with speed and fluency, but they did not operate from a stable understanding of what anyone actually meant. Intent entered the system and degraded as it moved through it, leaving outputs that often appeared correct but failed under real conditions. Continuity, judgment, and interpretation had to be supplied outside the system, which in practice meant a person stepping in to hold together what the architecture could not.
The Personal Semantic Layer is my response to that condition. It is designed as a foundational layer beneath the execution stack, where human intent is captured, maintained, and carried forward in a way that remains consistent across interactions. Instead of reconstructing meaning each time, the system operates from a shared understanding that includes context, relationships, and history. When that structure is in place, the way work gets done changes. Coordination that previously depended on constant human intervention becomes reliable within the system, allowing the person to focus on judgment, decisions, and direction rather than reconstruction and correction.
Within that structure, human approval is not an added safeguard but a central function. It provides a clear point where intent is confirmed before action, making the system usable in environments where accuracy, accountability, and traceability matter. Compliance, safety, and oversight are not layered on after the fact. They are inherent to how the system operates because the structure requires them.
I published the paper “Meaning Before Applications” in early 2026 to formalize this architecture and submitted it without attribution to multiple AI systems for independent evaluation. The responses were consistent in identifying it as reflecting senior level strategic and architectural thinking. Since then, I have been developing a series of applied use cases across domains such as email, CRM, accounting, and project management, each one tracing the same structural issue through a different context and showing how it resolves when meaning is held at the foundation.
I am based in Wilton, Connecticut, and I lead Hutchinson & Co., a private advisory practice. I am also developing the Personal Semantic Layer platform and am pursuing a pre seed raise and a technical co founder to build it.
The through line across all of this work is consistent. The same pattern that determined whether a business held together in practice is the same pattern that determines whether these systems will hold. The difference now is that the technology has reached a point where the gap is visible and can be addressed directly.
