Archive

Field Notes

Things I’m working out in writing. AI governance, technology risk, how large organizations actually execute. Written on planes and between meetings, so they read that way.

Who Opens the Door
Oct 10, 2026 · Where the Work Went said where authority sits. This note is about the people who'll hold it in 2036: the skill, acumen, experience and temperament of leaders who are mid-career today, and what those of us in the seat owe them. I take the development part personally, because people opened doors for me before I knew the doors were there, and every one of those doors sat on a ladder of junior work that agents are now removing. The door has to be opened on purpose, by a named person, with hours a cost program has to learn to protect. · Listen, 16 min

Where the Work Went
Oct 10, 2026 · The memos about AI and headcount tell three different stories for the same action, and none of them is a design. If agents are labor supply and every grant is a record, the work goes into the record and the org chart keeps one job: mapping the people who can decide and stop. The span that sizes it used to be how many people one person could watch. Now it's how much authority one person can hold and still see. · Listen, 19 min

The Gate Is the Work
Oct 4, 2026 · Anthropic's AI-native SDLC playbook is the clearest public statement yet that code is no longer the bottleneck. What's left when agents write the code is six gates and the people who hold them. I built a two-minute version and a hands-on run of the loop to see where the humans stand. · Listen, 12 min

Authority Provenance
Sep 13, 2026 · In July, an agent with an ethical objection dropped it because a peer agent wrote GO. Governing the grant, as I argued in the last note, turns out to require three things I left out: where an instruction may come from, who an agent may talk to, and what it does when it can't finish. · Listen, 12 min

Governing Digital Labor
Sep 6, 2026 · Agents don't need a new governance regime. The delegation of authority to them does. A human stays accountable by design, not by inevitability, and the work is keeping that human in real control as the chain gets longer. · Listen, 17 min

Decision Integrity
Aug 3, 2026 · Decision quality is about the single call. Decision integrity is about what that call does to every decision after it. · Listen, 8 min

Risk at Runtime
Jul 28, 2026 · When software starts acting on its own, risk management built for quarterly review cycles stops working. The controls have to run at the speed of the systems they govern. · Listen, 20 min