You know that thing where you ask AI to "write in your voice" and it produces something that technically uses words you might use but sounds like a corporate LinkedIn influencer possessed your keyboard? That's not an AI problem. That's a you problem. (I say this with love. And maybe three drinks. Mostly love.) AI can't follow patterns you haven't documented. When you say "write like me," AI has nothing to work with. So it guesses. Badly. Same structures, same rhythms, same soul-sucking monotony it uses for everyone else. The written equivalent of beige paint. The fix isn't better prompts. It's better inputs. Enter: The Tempo Extraction Exercise Tempo is one of four layers in VAST (Vocabulary, Architecture, Stance, Tempo). It's the most tangible. The layer where you can literally count patterns instead of gesturing vaguely at "vibes" like you're describing a coffee shop's aesthetic. I published the full exercise on Substack, but here's the actionable version for those of you who actually do the work. (All twelve of you. I see you.) THE EXERCISE (20 minutes, 3 parts) Part 1: Sentence Length Distribution (7 min) Grab 3 pieces of your writing. 500+ words each. Count sentences in each bucket: Short: 1-7 words Medium: 8-18 words Long: 19+ words Calculate your rough ratio across all three samples. Mine: 35% short, 50% medium, 15% long. This ratio tells AI what rhythm to target. Without it? Monotonous medium-length sentences that put readers to sleep. (AI defaults to medium the way middle managers default to "let's circle back.") Part 2: Sentence Starters (6 min) List the first word of every sentence in your samples. Look for repetition. Common patterns: 🧉 "I" = first-person dominant 🧉 "But/And/So" = conjunction starters (conversational energy) 🧉 "The" = noun-leading, content-focused 🧉 Verbs = command/imperative style Your top 3-4 starters are your defaults. Document them. Mine lean heavy on conjunction starters. "But" and "And" everywhere. English teachers hate me. Readers don't seem to mind.

I need to tell you something about this community. It's not an accident that it exists. It's not a "lead magnet" or a "top of funnel play" or whatever the growth hackers are calling it this week. It's a bet. I spent five years posting on X. Under 100 followers. Five years of playing the algorithm game exactly how you're supposed to play it. And I finally figured out why it wasn't working: the game changed and nobody sent me the memo. Here's the short version: The ROI of algorithm-chasing has collapsed. Not for consuming content (TikTok still works great for melting your brain). But for building anything sustainable as a creator? The slot machine is broken. Meanwhile, the ROI of trust networks is exploding. Discord. Paid Substacks. Communities like this one. We're in what I'm calling Phase 2 of the Content Reformation. The skepticism phase. People are actively retreating from public platforms into private spaces where trust can actually be verified. By Phase 3 (2026-2027), the drawbridge goes up. The people inside trust networks become the gatekeepers. Everyone else gets stuck in the algorithmic wasteland fighting for scraps. This community is my lifeboat. And yours. I'm building here because I believe the relationships we form now—before the retreat accelerates—are the ones that matter in three years. So: introduce yourself if you haven't. Start a conversation. Help someone. The people in this room are going to be your referral network when word-of-mouth becomes the only discovery that works. The drawbridge is still down. But it's going up. Let's make sure we're all inside when it does. What's your read on this? Am I being dramatic, or does this match what you're seeing too?

Alright, confession time. I've been using AI collaboration ass-backwards for months. Maybe you have too. Here's what I was doing: 🧉 Use AI for brainstorming 🧉 Use AI for drafting 🧉 Use AI for generating variations 🧉 Edit alone, exhausted, at midnight See the problem? All my AI collaboration was expansion—more ideas, more words, more options. Then I'd do contraction alone. The cutting. The tightening. The "what is this piece actually about" reckoning. At night. Tired. When my judgment was sharp as a spoon. The insight that changed everything: When AI expands your content, its voice creeps in. More words = more chances for generic phrasing. When AI contracts your content, your voice survives. Because you wrote the original—AI just identifies which of your words are strongest. More AI involvement in contraction = more authentic output. (I know. Sounds backwards. Felt backwards. Then I tried it and my editing time dropped 40%.) The rhythm that actually works: Expand → Contract → Expand → Contract 1. Generate ideas (with AI) 2. Pick one direction (with AI) ← most people skip this 3. Draft it out (with AI) 4. Cut to essential (with AI) ← and this Prompts that hit hard: "Cut 40% while keeping everything that matters. Tell me what you removed and why." "Read this as a skeptical reader one boring paragraph from closing the tab. What would you skip?" "Which sentences could be in anyone's article? Which sound like no one else would write them?" That "tell me why" is doing heavy lifting. Forces judgment, not just mechanical reduction. I put together a full toolkit on this: 25 contraction prompts organized by phase—finding the core, cutting with judgment, choosing between options, tightening. Plus checkpoint questions so you stop expanding forever and actually ship. Download: The Contraction Prompts Worksheet → https://sendfox.com/lp/m52jjn Discussion question: What's your ratio of expansion to contraction when you're working with AI? Be honest. Are you doing the hard part alone?

Last week I caught myself doing something embarrassing. I was sorting through article ideas and realized I'd created two mental piles: "Good enough to publish" "Too valuable to share yet" And I'd been calling that second pile "strategy." Save it for the course. Save it for when the audience is bigger. Save it for when it "matters." Here's the problem: that audience I'm saving content for? They don't exist yet. They're currently being trained on my copper while I hoard the gold. There's a 500-year-old economic principle called Gresham's Law: "Bad money drives out good." When two currencies trade at the same face value but have different intrinsic worth, people spend the cheap one and save the expensive one. Every time. AI did this to content. Production costs hit the floor. The gap between "adequate" and "actually good" became a canyon. And we all responded rationally—publish the copper, vault the gold. The result? The commons flood with slop while everyone hoards their best stuff for "later." Later = a course that might not get built. Later = an audience that's learning to expect less. Later = never. New rule I'm testing: The instinct to hoard is now my signal to share. When I catch myself thinking "this is too valuable for free content," that's exactly when I need to publish it. ******** Question for you: What's in YOUR vault? Not the polished version. Just the topic. What idea have you been "saving for later" that your audience might actually need right now? Drop it below. Sometimes naming what you're hoarding is the first step to cracking it open.

Most of your AI rewrites happen because AI assumed when it should have asked. You feed it your Voiceprint. You give it the topic. It generates 800 words. And somewhere around paragraph three, you realize it invented a personal story you never had, took a stance you'd never take, or structured the argument backwards from how you actually think. Now you're rewriting instead of editing. Again. Here's the fix. Add this exact line to every collaboration prompt: ``` Before drafting anything substantial, identify 2-3 places where you'd need my specific input—personal experiences, opinions, examples—and ask me first. Don't assume. Ask. ``` What changes: Before: AI generates → You read → You discover the problems → You rewrite After: AI asks → You answer → AI generates with your actual input → You edit details The difference isn't subtle. When AI asks "What's your personal experience with this?" before writing, it uses YOUR story instead of inventing generic filler. When it asks "How strongly do you want to make this claim?" it calibrates stance before drifting into guru-mode. The places AI needs to ask are what I call "you-shaped holes"—sections where only your experience, your opinion, your specific take belongs. AI can't fill these. It can only fake them. And faked content is exactly what makes output feel hollow even when the voice sounds right. Try it on your next piece. Count how many corrections you give versus last time. *** Quick challenge: Drop your "ask first" instruction in the comments. What specific questions do you want AI to ask before generating? Seeing each other's versions helps everyone refine theirs.