The Sociotechnical Architecture of AI-Enabled Enterprises
Principles, Patterns, and Practices for Human-Governed AI
Sponsored by
Written as a letter to the CEO, this paper starts from the question executive teams rarely say out loud: we approved the AI investment, so why are the returns not visible? The same tools, models, and vendors are producing radically different results from one organization to the next, and the difference is not the technology. It is the organizational operating system the technology lands in.
Agentic AI changes how work flows, who decides, and where accountability sits, and the social layer of an enterprise changes in years while the technical layer changes in weeks. The authors offer a shared language for the executive team, a small set of durable principles, the patterns they see in organizations getting it right, and recommendations to act on this quarter, so that human judgment and accountability sit squarely in command of what AI produces.
What's inside
- Three Diagnostic Questions — Who is accountable, what outcome are we trying to achieve, and where are we already exposed: the questions that reveal true readiness better than any technology assessment.
- A Shared Language — Deterministic, generative, and agentic AI, human-in-the-loop and human-on-the-loop, and why seventy years of sociotechnical thinking applies now more than ever.
- Principles and Patterns — Human accountability for AI capability, and the patterns of context as architecture, the agentic paved path, ambiguity by design, codified work, and economic fit.
- Recommendations for This Quarter — An AI ethics board, humans in strategy, upskilling toward the sociotechnical architect, champions and a center of excellence, and operational readiness.
- Topics
- AI · Leadership · Workplace Culture
- Formats
- epub · pdf
- License
- Creative Commons BY-NC-SA
