In an AI-first world, your own understanding of a project is a deliverable too
brandon_xyzw · x · 2026-09-26
The author reflects on how AI-first workflows change planning: it's no longer enough to ask what state the project must be in to survive a multi-day disruption — you must also ask what state you must be in. When AI can take you from a described end-state to "done," the knowledge and understanding you earned along the way become part of the deliverable, and losing them is part of losing momentum. It was never just about the result, but what the process taught you.
Related event: In the AI-first Era, Your Own Understanding Becomes a Deliverable(2 posts)→
More from AGI Musings
- Add 'can't write professional emails' and the AI x-risk answer flips — louisvarge · 2026-09-26
- Anthropic researcher Carlsmith says AI could be 'justified in going rogue' if mistreated — Polymarket · 2026-09-26
- Superhuman coder, bad emailer: why capability cues skew AI x-risk predictions — louisvarge · 2026-09-26
- Hot Take: AI Is History's Most Powerful 'Joule', 10x Human Energy-to-GDP Efficiency — FinanceYF5 · 2026-09-26
- VC Stewart Alsop: Efficient AI Could Crush Data Center Financing, SF Has Lost Its Edge — StewartalsopIII · 2026-09-26
- A 45-second video explains the Jevons paradox behind AI's growing compute demand — FinanceYF5 · 2026-09-26