The problem
What was broken before AI
A finance newsletter draft can start as a messy pile of market observations, research notes, trade context, caveats, and jargon. Before AI, turning that into readable copy likely required more manual rewriting and copyediting, especially when the language had to stay precise enough for a trading audience.
What changed
What the use case made possible
AI became a production layer between Belanger’s human market thinking and the final newsletter: ChatGPT streamlines the draft, Claude helps copyedit, and custom GPTs are used to handle technical language with fewer unwanted substitutions or hallucinated phrasing.
Why this matters
Why this use case is worth studying
This case sits in the uncomfortable but important middle ground of AI-assisted publishing. The workflow still depends on the writer’s own research, judgment, and accountability, but it uses AI to reduce the drag between having a market take and shipping a polished issue. For specialist creators, the lesson is that generic AI editing can be too generic; the system needs domain vocabulary, house style, and clear boundaries around what it is allowed to change.
Use this when
When this pattern applies
Use this when you already have the expertise and raw material, but the editing and production process slows you down.
