EACOE launches AI architecture initiative for CIOs

The Enterprise Architecture Center Of Excellence says its new program is designed to help CIOs turn AI spending into measurable business results. The effort pairs EACOE’s architecture frameworks with AI-ready data discipline to move organizations from pilots to production.

Why it matters: - EACOE is targeting a common enterprise problem: AI budgets are rising, but many pilot projects never become scalable business systems. - The initiative is designed to help CIOs choose AI use cases with clearer payoff, stronger governance and better data foundations. - The stated goal is to move organizations from experimentation to measurable outcomes faster.

What happened: - The Enterprise Architecture Center Of Excellence announced a new Enterprise Architecture initiative for CIOs and enterprise leaders. - The program combines EACOE’s R.E.A.L. Enterprise Architecture frameworks with AI-ready data discipline. - EACOE said the initiative is intended to help enterprises connect AI investments to business objectives and operating priorities. - The announcement was made Sept. 19, 2026, from Pinckney, Michigan.

The details: - EACOE says the approach applies its R.E.A.L. framework — Realistic, Enabling, Actionable and Logical — to AI planning and execution. - The initiative also draws on EACOE Managing Director Samuel B. “Sam” Holcman’s work on AI-ready data. - EACOE says the model links AI decisions to strategy, operating model, business processes, information needs and data realities. - The company says that gives CIOs a way to decide which AI projects to fund, defer or redesign. - Holcman said the main issue is data quality, not AI technology. - Holcman has said many organizations fail because they skip the foundational work needed to prepare data for AI use. - His Four Types of Data framework is part of the initiative and is also featured in his book, AI-Ready Data: What Leaders Must Get Right Before AI Works. - Holcman said unstructured or poorly governed data can undermine AI outcomes regardless of the model, platform or vendor. - EACOE says the initiative is meant to help leaders identify the business, governance, data and architecture conditions needed before committing major resources. - The program is built on EACOE’s Enterprise Framework and Quick Start Methodology. - EACOE says those tools are designed to help organizations develop an actionable enterprise architecture in months rather than years. - The company says expected outcomes include clearer links between AI and business goals, better prioritization, improved governance and a roadmap from proof of concept to implementation. - EACOE says its framework and methodology have been used by General Motors, Ford Motor Company, The Walt Disney Company, American Express, Walmart, government agencies and other organizations. - Tony Scott, former corporate vice president and CIO of Microsoft, called the EACOE approach “a book of action, not surveys or choices, moving from theory to implementation.”

Between the lines: - The announcement positions enterprise architecture as the gatekeeper for AI spending, not just a supporting function. - EACOE is also making a broader case that AI success depends on data discipline and governance before model selection or vendor choice. - The framing suggests CIOs should treat AI as an enterprise operating issue rather than a standalone technology purchase.

What’s next: - EACOE is offering the initiative through training and certification workshops, plus ongoing support and mentoring. - CIOs, enterprise architects, business leaders and data leaders can use the program to apply EACOE methods to their own priorities and architectural challenges. - Upcoming workshops are available for registration at EACOE workshops. - EACOE says executive briefings, media interviews and review copies of Holcman’s books are available on request.

The bottom line: - EACOE is betting that better architecture and cleaner data, not more AI experimentation, will determine which enterprise AI projects deliver real business value.