He describes his own process: a 300-word prompt becomes a 660-word draft, which he rewrites into a 3,000-word essay. Maybe 40-50 words are Claude's. He's proud of how he uses AI. He says we're still in the "filmed stage plays" era of AI writing, and the masters will emerge.

This resonates with me. I'm an engineer, not a writer. Over the past few years, I've been using AI extensively — for drafting articles, structuring complex technical ideas. The AI helps me articulate thoughts I couldn't express alone. But I always rewrite, edit, and make sure every word is mine.

Here's my dilemma: should I disclose that I used AI as a tool?

When I don't disclose, I feel dishonest — like I'm claiming a skill I don't fully have. When I do disclose, people dismiss the work as "AI slop" before reading it. I've had journalists ignore emails, investors go silent, and peers assume the ideas aren't really mine.

Tim argues against Ted Chiang's position that AI writing lacks "choices." Tim says a prompt can contain a lifetime of experience, like Leonard Cohen's "Hallelujah" taking five years or Dylan's "Blowing in the Wind" taking ten minutes — both are masterpieces.

But here's the practical question for this community:

If you're an engineer, founder, or technical person using AI to communicate ideas you couldn't express otherwise:

1. Do you disclose AI use? When and how? 2. Have you faced backlash or dismissal when you did disclose? 3. Do you think "judge the output, not the means" is realistic, or will AI-assisted work always carry a stigma?

I'm genuinely trying to figure out the right approach. Not to cheat, but to be honest without being penalized for using a tool that helps me think and communicate better.

Link to Tim's essay: https://substack.com/app-link/post?publication_id=7730464&post_id=213041071