· personal  · 3 min read

Stop Asking "What Can AI Do?" Start Asking "What Do I Want to Be Exponentially Better At?"

DRAFT

Outline

Hook: Everyone’s asking “What can AI do?” The better question is “What do I want to be exponentially better at?” One leads to playing with toys. The other leads to transformation.

Core Argument: AI adoption fails when it’s capability-driven (“Look what this can do!”) instead of goal-driven (“Here’s what I want to achieve”). The most powerful use of AI isn’t learning its tricks—it’s identifying your leverage points and amplifying them 10x.

Key Sections:

  1. Why “AI Capabilities” Is the Wrong Starting Point

    • The technology-first trap: solutions looking for problems
    • FOMO-driven adoption: using AI because everyone else is
    • The “shiny object” cycle: ChatGPT → Midjourney → next thing
    • Why most AI experiments fizzle out after the honeymoon phase

  2. The Personal Amplification Framework

    • Step 1: Identify your high-leverage activities (what creates disproportionate value?)
    • Step 2: Find bottlenecks in those activities (where do you slow down?)
    • Step 3: Map AI capabilities to bottlenecks (not the other way around)
    • Step 4: Measure impact on the leverage activity, not AI usage
    • Example walkthrough: Content creation workflow

  3. My Personal Leverage Points (And The AI I Use For Each)

    • Writing: AI for outlining and research, not drafting → keeps my voice
    • Coding: Copilot for boilerplate, I handle architecture → faster builds
    • Music: AI for production variations, I write melodies → more output
    • Knowledge work: RAG for recall, I do synthesis → better decisions
    • Idea capture: 99 Minds for organization, I do evaluation → execution rises

  4. Questions to Ask Instead of “What Can AI Do?”

    • “What would 10x my creative output?”
    • “Where do I waste time on things that don’t require my unique skills?”
    • “What am I avoiding because it’s too tedious?”
    • “What would I do if I had a brilliant assistant with perfect memory?”
    • “Which of my strengths could be amplified vs. weaknesses compensated?”

  5. The Danger of Replacing vs. Amplifying

    • AI that replaces: You lose the skill, become dependent
    • AI that amplifies: You get better at the skill, maintain agency
    • How to tell the difference: Are you learning or abdicating?
    • The “deskilling” trap and how to avoid it

Examples/Stories:

  • Personal: Tried using AI for songwriting → felt hollow. Used AI for production → 3x output
  • Client story: Law firm wanted AI to write briefs → failure. AI to research case law → success
  • Counter-example: Someone who let AI do all their coding, couldn’t debug anymore
  • Success metric: Projects shipped per month doubled, not “AI queries per day”

Takeaways:

  • Start with desired outcomes, not AI features
  • AI should make you better at being you, not replace being you
  • The goal is leverage, not automation
  • Measure results in your domain, not AI metrics
  • Best AI usage is invisible—it’s just how you work now

Cross-Links:

  • ← “RAG, But Make It Real Life” (Series 1-4)
  • → “How I Design AI Systems” (Series 1-7)
  • → “How I Audit My Life Like a Product” (Series 3-25)
  • → “Personal Blueprint” (Series 3-22)
Share:
Back to Blog