The Operating System for the AI Era

Governance. Context.
Control.

A body of work on how leaders build, govern, and orchestrate intelligent systems — from context engineering to agentic architectures to the transformation operating model.

Explore the Work About the Book

The AI Executive's Handbook: Harnessing the Ungoverned Machine by Misha Sulpovar

The Discipline

The Framework

Four interconnected domains that define how organizations navigate the AI era. Each is a body of research, not just a topic.

Wired Cognition OS

AI Strategy

Product, organization, and transformation in the AI era.

Agentic Surface

Multi-agent architectures, control planes, and orchestration.

Knowledge Architecture

Writing, selecting, compressing, and isolating context at scale.

AI Governance

Frameworks, compliance, ethics, and the EU AI Act.

Latest Research

Recent Work

View archive

In a concrete gallery, an orange wire sculpture stands on a plinth surrounded by glass panels holding glowing indigo copies of it, while a bundle of indigo threads reaches for one of the copies instead of the original

Context EngineeringAnalysis

What Is a Stale-Version Attack?

When a page changes, the old copy rarely goes away, and neither your retriever nor a bigger context window can tell which one is current. What that costs a business, and what it looked like against Amazon Bedrock: one answer in six.

Nov 2, 202611 min readApplied

Also Recent

What Coding Agents Ship When You Govern Their Context

I gave the same coding model the same vague request three ways, changing only which context each step was allowed to see. On a real app, the difference was the gap between a demo that renders and software that works.

Aug 19, 20267 min readApplied

Context Engineering

Strategic Ignorance

I have argued a lot for broad context. Here is the nuance: the value is not more context, it is the right context, at the right time, for the right purpose. Sometimes the highest-leverage move is making an agent forget.

Aug 12, 20263 min readApplied

A governance layer deciding which artifacts are approved, current, and verified before retrieval runs

Context EngineeringFramework

What 'Context Governance' Actually Means

The plain-language version of our arXiv paper: the layer beneath retrieval that decides whether an artifact was approved, current, and verifiable before the agent ever reads it, and two experiments showing why that layer is cheaper and more correct.

Jul 15, 20263 min readOperational

A knowledge graph forming a brain over a terminal — a governed context frame for a second brain

Context EngineeringField Note

Your Second Brain Needs a Better Context Frame

Notes pile up, the AI forgets, and the whole thing turns to slop. The missing piece isn't a better model or another app — it's a context frame. Here's the one I built, on video, plus a guide you can hand to your own AI.

Jul 7, 20263 min readApplied

A hash-chained, version-verified context layer sitting beneath retrieval and an AI agent

Context EngineeringFramework

Context Governance, Now With a Paper

ContextNest is now a peer-reviewable spec. Our new arXiv paper formalizes context governance — the layer beneath retrieval that decides what's approved, current, and verifiable — and backs it with two controlled experiments.

Jul 3, 20262 min readOperational

Clinical AI moving from retrieval-only tools toward governed, frontier-model reasoning

Context EngineeringAnalysis

Healthcare Is Quietly Moving Past RAG

Specialized medical AI tools like OpenEvidence were supposed to beat general models by retrieving over peer-reviewed literature. In 2026 benchmarks, general frontier models are catching and passing them. The lesson isn't that RAG is dead — it's that retrieval was never the moat.

Jun 24, 20263 min readApplied

View all articles

Stay Connected

Executive AI intelligence,
delivered.

Governance frameworks, agentic architecture patterns, and strategic implementation insights. For leaders building the AI era.

Governance. Context. Control.