REAL Talk With Sam Holcman

The InfoWorld article “Small language models: Rethinking enterprise AI architecture” argues that as large language models (LLMs) hit limits of scale, cost, and risk, enterprises are shifting toward small language models (SLMs) that are faster, cheaper, and more private for well-defined, repetitive tasks. It highlights three primary advantages: division of labor between small and large models, radical economic efficiency for high‑volume workloads, and improved privacy at the edge when models run on devices or on‑premises. The underlying message is that competitive advantage will increasingly come from how well organizations architect and orchestrate multiple models around their own data, workflows, and governance.

Ford’s new electric truck program is being called its “Model T moment” – not because it’s just another vehicle, but because Ford is tearing up a century of manufacturing practice to build something fundamentally different. In doing so, they have replaced the traditional assembly line with what they call an “assembly tree”: a modular way of building that uses far fewer parts, far less complexity, and a completely new production logic.
Enterprise Architecture (EA) and Business Architecture (BA) are at a similar crossroads. Most organizations are still using an architectural “assembly line”: long sequences of phases, hand‑offs, documents, and exams that were designed for a slower, more predictable world. These legacy, exam‑centric certifications, methodologies, and frameworks add steps and artifacts, but struggle to produce architectures executives actually use to run and change the business.
The Enterprise Architecture Center Of Excellence (EACOE) and The Business Architecture Center Of Excellence (BACOE) represent the modern alternative. They are not “another framework” running down the same line. They are the architecture equivalent of Ford’s assembly tree: a new way of producing outcomes, built around a small number of powerful, reusable components that can be combined quickly to solve real business problems.

Enterprise Architecture and Business Architecture are at the same inflection point that software engineering has reached. When a field confuses vocabulary for competence, titles for capability, and exams for professional readiness, it creates a class of certified people who can describe the work without being able to do the work.
That model is no longer good enough. The next era of Enterprise Architect and Business Architect certification must move away from exam-centric credentialing and toward practitioner-based professionalization grounded in evidence, judgment, and demonstrated delivery.

Enterprise Architecture (EA) and Business Architecture (BA) have moved decisively toward the center of enterprise transformation in the last 30 days, especially as organizations try to operationalize AI at scale. Recent market and thought-leadership signals show a clear shift away from architecture as static documentation and toward architecture as a business decision capability focused on speed, risk, investment, and operating-model change.

In a recent CIO article, by Martin De Saulles, “How effective are semantic hubs in moving agentic AI forward?” the authors argue that semantics are now the backbone of enterprise AI, especially as organizations rush to deploy agentic AI systems at scale. They highlight a critical shift: the challenge is no longer just moving and storing data but ensuring that data means the same thing wherever and however it is used.
That is precisely the problem space Enterprise Augmented Intelligence (EAI™) was designed to address. And it is the space EACOE and BACOE have been working in for more than fifty years.