Welcome to the Morneau Letter

Slow questions for fast technologies

For most of my career in technology education, one question guided our work: what do students need to know and do? We built curricula around the answer, and for a long time the question served us well. It no longer fits the situation our students, and the people who hire them, now face. When a capable tool can produce a plausible answer to almost any question in seconds, knowing the answer is no longer the scarce thing. The scarce thing is judgment: telling a good answer from a bad one, deciding which question was worth asking, and standing behind the result.

This letter exists to think about that gap, slowly and in the open. One letter arrives each month, on the first Sunday. Each takes up a single question about AI, education, and the formation of professionals, and tries to resolve it into something you can use. Now and then Ponder, the owl who keeps me company on the website, leaves a note in the margin. There is no funnel behind it and no scheme. It is written for people who can tell insight from content.

The gap that matters

I came to this through research, not opinion. In 2006, as part of my doctoral work at Pepperdine on experiential learning in information assurance education, I ran a four-round Delphi study with seventeen experienced practitioners across healthcare, government, consulting, the military, and education. The panel confirmed what employers had been saying for years: half of the real problems professionals face are about people and process, not technology, and credentials measure almost none of that half.

The finding that organized everything afterward is simple to state and hard to act on. Forming a solutions architect depends on situated judgment, and situated judgment develops in practice fields, not in lectures. Design is distinct from construction. Knowing how to build something is not the same as knowing what to build and why.

The ideas you will meet here

Some of the vocabulary will recur, so here it is in one place. Each has an essay of its own on keithmorneau.com.

• Value-delivery skills: the communication, negotiation, and organizational navigation that make technical skill useful to another person. They cannot be taught by lecture. They have to be experienced. (Read the essay)
• The curveball: a deliberate disruption placed inside a learning exercise to force adaptation. A group of educators and practitioners working on how to prepare technology graduates gave it the name in 2003, and the research found it has to run through a whole program, not wait for a capstone. (Read the essay)
• The fifty-percent problem: about half of what goes wrong at work is people and process, and our credentials barely touch it. (Read the essay)
• Practice fields and real fields: safe settings where work can be rehearsed, and real ones with real clients and real stakes. Formation needs both, in that order.
• The boundary practice: a living community between schools and the professions that keeps supplying authentic problems, so learners can cross from one world into the other. Most advisory boards do not work this way. (Read the essay)
• The Sputnik mistake: the habit of answering a technological moment by moving professional education away from apprenticeship and toward theory. The research suggests AI is creating the same pressure, in the same direction. (Read the essay)
• The solutions architect: the professional with technical and business depth, emotional intelligence, and the ability to deliver creative solutions for clients. It is the human role that remains when routine construction is automated.

Underneath these sits the instructional model I use to organize the work: Prepare, Practice, Perform, Assess. It will get a letter of its own.

What AI changes, and what it does not

I am not writing against the tools. The floor has risen because AI now handles much of what used to be entry-level technical work, and professionals who use it well are more capable than those who do not. The ceiling is higher too. The people who use AI well do more than run prompts. They frame the right problems, direct the tool, catch errors in the output, and take responsibility for the results.

The boundary sits where the research would put it. Drafting, summarizing, and constructing are increasingly automatable. Problem framing, critical evaluation of output, and judgment under uncertainty are not, and they are learned the way a craft is learned: slowly, in person, and from someone who has done the work.

That is the part that worries me. Entry-level work was where much of that learning used to happen. If the first rungs of the ladder are automated, who trains the next generation of professionals? That question is at the heart of my book, Wisdom is not a feature you can install, the first volume of The Apprentice Cycle, and it will keep returning here.

What to expect

One letter a month, on the first Sunday: a single question, the evidence behind it, and something you can do with it. The essays on the site go deeper on individual ideas, and you can put questions to Ponder there. If you teach, hire, train, or lead people who now work alongside AI, this is written for you.

I would like to close with a question, and I would like to hear your answer. Who taught you to judge your own work, and what did they have you do that no course ever did?