· technical  · 2 min read

Why "AI-First" Doesn't Mean "No Soul" – Designing Tools That Still Feel Human

DRAFT

Outline

Hook: “AI-first” has become startup speak for “we replaced the human touch with algorithms and called it innovation.” But AI-first doesn’t have to mean soulless. It should mean: AI handles the tedious stuff so humans can focus on what makes us human.

Core Argument: The best AI tools feel human not despite their intelligence, but because they’re designed to augment human creativity, emotion, and judgment. Soul in AI products comes from intentional design choices that preserve human agency, provide transparency, and create emotional resonance.

Key Sections:

  1. What “Soul” Actually Means in Software

    • Not personality or chattiness—it’s resonance and trust
    • The feeling that the tool “gets you”
    • Examples: Spotify Wrapped (emotional connection), VS Code (respects your workflow)
    • Anti-examples: Clippy, overly-chatty AI assistants

  2. The Three Pillars of Soulful AI Design

    • Transparency: Show the AI’s reasoning, don’t hide it
    • Agency: User always in control, AI makes suggestions not decisions
    • Resonance: Design for emotional impact, not just efficiency
    • Case study: 99 Minds voice interface—natural but not pretending to be human

  3. Where AI Should (and Shouldn’t) Touch the Experience

    • AI for heavy lifting: categorization, search, pattern recognition
    • Human for meaning-making: final decisions, creative direction, emotional weight
    • The “last mile” problem: AI can get you 80% there, human finishes
    • Why fully automated anything feels hollow

  4. Building Warmth Into Cold Logic

    • Micro-interactions matter: loading states, transitions, confirmations
    • Language design: conversational but not fake-friendly
    • Error handling: turn failures into learning moments
    • Visual design: avoid sterile “tech aesthetic”

  5. Testing for Soul (Yes, Really)

    • User emotional response metrics beyond NPS
    • “Would you recommend to a friend?” vs. “Does this feel like it was made for you?”
    • Long-term engagement patterns
    • Qualitative feedback: what do users say about the product?

Examples/Stories:

  • 99 Minds: Why voice-first creates intimacy vs. typing into a form
  • Law firm tool: Lawyers loved AI summaries but hated when it made assumptions
  • Personal music project: Using AI for production but keeping songwriting human
  • Counter-example: A generic chatbot that felt like talking to a wall

Takeaways:

  • “AI-first” is a tech choice, “soulful” is a design choice—you can have both
  • Preserve human agency at every interaction point
  • Test for emotional resonance, not just task completion
  • The goal: AI that makes you feel more capable, not more replaced

Cross-Links:

  • ← “From Mixtapes to Machine Learning” (Series 1-1)
  • → “What I Learned Building 99 Minds” (Series 1-3)
  • → “How I Design AI Systems” (Series 1-7)
  • → “How to Design Products for Non-Tech People” (Series 2-19)
Share:
Back to Blog