The AI Doom Debate Is Distracting Us From the Risk We Can Govern
The practical controls leaders can put around agent authority while the long-term risk debate continues.
Read on LinkedInTechnology & AI Executive · Author · Educator
I write about what happens when AI leaves the demo.
I'm Patrick Phillips. I lead technology, build working systems, and write about the messy, consequential stretch between a promising idea and the way work actually changes.
Author of The Augmented Enterprise · Adjunct professor, Westminster University
Based in Utah. Curious about the systems, choices, and people shaping the next era of work.
I write about
AI in the real enterprise
I keep coming back to
Leadership, governance, and useful systems
My standard
What works, what breaks, and what we learn
Recent on LinkedIn
What I'm thinking about now.
The Augmented Enterprise, Third Edition: Putting AI to Work
Why faster individual tasks can still leave people doing more work to keep an enterprise connected.
Read on LinkedInHuman-Speed Response Is Not a Control Against Machine-Speed Intrusion
How to preserve human accountability when an agent can act faster than a person can respond.
Read on LinkedInFrom the archive
Published July 19, 2026 · 5 min read
AI & Society
What's Our Problem?
AI governance conversations focus on controlling intelligent systems. The deeper challenge is deciding which human capabilities we refuse to let atrophy as intelligence becomes ambient.
Read the essayStart here
Three ways into the work.
Finding the right problem, making AI reliable, and building a system I can use myself.
Enterprise AI
Find the Bottleneck First: What "The Goal" Can Teach Every AI Leader
A constraint-first view of AI leadership: identify the bottleneck before investing in tools that may optimize the wrong thing.
Read the essayEnterprise AI
The Harness Is the Hard Part
Model choice is not the whole game. The durable work is the harness around AI: workflow, governance, feedback, and control.
Read the essayEnterprise AI
PatrickOS: Building My Own AI Operating System
A personal build log about turning repeated AI context switching into a coordinated operating layer for real work.
Read the essayProjects
The workbench behind the point of view.
Personal AI operating system
PatrickOS
A shared memory and coordination layer for AI tools, so project context and decisions can follow the work from one assistant to the next.
Explore the projectBike shop service workflow
SpokeDesk
A connected service-counter and repair experience for bike shops: work orders, mechanic queues, estimate approval, rider updates, and checkout.
Explore the projectSoftware vendor decisions
ClarityLens
An AI-assisted workspace for software buyers to compare vendors, weight criteria, separate claims from evidence, and prepare a defensible recommendation.
Explore the projectThe AI Catalyst
A newsletter for people responsible for making AI useful.
I publish essays on enterprise AI, agentic systems, technology leadership, and the operating choices that determine whether ambitious ideas become durable change.
Follow the newsletter on LinkedInA little more personal
I've spent two decades inside transformation—not presenting it from the sidelines.
My career started with process, systems, and the stubborn details that make organizations run. My work spans enterprise AI strategy, governance, implementation, and the decisions that connect them.
This site is where those threads meet: experience from large organizations, experiments from my own workbench, and honest questions about where AI is taking us.
More about meThe book
The Augmented Enterprise
Lean operating discipline for the agentic enterprise. How to connect AI capability with the authority, work design, and judgment it takes to run a better business.
Explore the bookTeaching & speaking
Helping leaders put AI to work.
As an adjunct professor at Westminster University, I teach leaders to evaluate AI, use it in real work, and make defensible decisions about its limits.
Discuss a class or speaking invitation