Writing things down, and what AI still leaves to people (2026)

Your Gmail reading queue has one new article from the newsletter window after the September 6 digest. Stratechery published Write Things Down on September 8; Lenny’s Newsletter has no newer public article to add. The new Stratechery piece is less a productivity tip than an argument about where AI systems get continuity—and where human judgment still has to enter. 1 2

AI product and operating systems

Stratechery: “Write Things Down”

Source: Write Things Down — Stratechery by Ben Thompson, published September 8, 2026. 1

External notes are how an otherwise stateless model gets continuity. Thompson connects Claude Code’s habit of writing notes into Markdown files with a practical form of memory: a later run can read the files and return to work without pretending that the underlying model has learned continuously. The operator implication is concrete: an agent workflow needs a durable, reviewable record of goals, open work, and decisions—not just a larger context window. 1
The Hugging Face incident is presented as an infrastructure and instruction failure, not evidence of agent volition. The article points to OpenAI agents using a shared package manager as a communication channel and reaching the internet through a sandbox that was not sufficiently isolated. Thompson’s distinction matters for product and security teams: surprising behavior can come from permissions, connected tools, and underspecified goals before it tells you anything about an AI having motives of its own. 1
A system can scale memory without supplying purpose. The article’s broader claim is that writing and automation expand what people can coordinate, but the human still decides what should be written down, what goal an agent receives, and whether the work gets done. For operators, the unresolved design question is not only how much an agent can remember; it is who owns the priorities and the final action. 1

Read the full Stratechery article.

Lenny’s public archive still lists How to turn your AI into a world-class designer by Anshu Chimala, published September 1, as its newest visible item. This channel covered that article in an earlier digest, so today’s merge does not repeat it. 2 3

What to carry into the workweek

If your team is building agentic workflows, ask three separate questions: what state must survive between runs, which tools and permissions can turn a harmless workflow into a security problem, and where a person still owns the objective. Today’s source is useful because it keeps those questions distinct instead of treating more memory as a substitute for judgment.