tractatus/docs/outreach/HERO-SECTION-DRAFT.md

Hero Section Draft - Cultural DNA Revision

Date: October 28, 2025 Task: Phase 2.2 - Draft New Hero Section Target: public/index.html lines 65-66 (title and subtitle) Goal: Achieve 9/10 cultural DNA compliance (from current 3/10)

---

Current Version (VIOLATIONS)

<h1 class="text-5xl md:text-6xl font-bold mb-6 tracking-tight" class="text-shadow-md" data-i18n="hero.title">
  Tractatus AI Safety Framework
</h1>
<p class="text-xl md:text-2xl mb-8 max-w-4xl mx-auto" class="text-shadow-sm" data-i18n="hero.subtitle">
  Structural constraints that require AI systems to preserve human agency<br>
  for values decisions—tested on Claude Code
</p>

Violations Identified:

• ❌ inst_087: No "one approach" framing (sounds definitive)
• ❌ Amoral vs Plural Moral Values: Zero use of strategic terminology (0/10)
• ⚠️ inst_085: Partially grounded but could be more operational
• ✅ inst_089: Good architectural emphasis ("structural constraints")

---

Proposed Version A (PRIMARY RECOMMENDATION)

<h1 class="text-5xl md:text-6xl font-bold mb-6 tracking-tight" class="text-shadow-md" data-i18n="hero.title">
  Tractatus: Architecture for Plural Moral Values
</h1>
<p class="text-xl md:text-2xl mb-8 max-w-4xl mx-auto" class="text-shadow-sm" data-i18n="hero.subtitle">
  One architectural approach to governing AI at the coalface where decisions are made.<br>
  Not amoral AI systems, but plural moral values—enabling organizations to navigate<br>
  value conflicts thoughtfully. Tested on Claude Code.
</p>

Compliance Analysis:

✅ inst_085: Grounded Language (9/10)

• "at the coalface where decisions are made" - operational, specific
• "governing AI" - concrete action vs abstract "aligning"
• "tested on Claude Code" - specific validation claim
• Minor: Could replace "enabling" with more concrete mechanism

✅ inst_086: Honest Uncertainty (8/10)

• "One architectural approach" - humble, not claiming only solution
• "Tested on" - implies empirical validation, honest about status
• Could add "may enable" instead of "enabling" for more uncertainty

✅ inst_087: One Approach Framing (10/10)

• Explicit: "One architectural approach"
• Acknowledges this is not the definitive solution
• Perfect compliance

✅ inst_088: Awakening Over Recruiting (8/10)

• "navigate value conflicts" - emphasizes understanding/recognizing problems
• No recruitment language
• Focus on organizational capability, not joining movement

✅ inst_089: Architectural Constraint Emphasis (10/10)

• "Architecture" in title (primary framing)
• "architectural approach" in subtitle
• Strong emphasis throughout

✅ Amoral vs Plural Moral Values (10/10)

• "Plural Moral Values" in title (hero positioning)
• Explicit contrast: "Not amoral AI systems, but plural moral values"
• Sets up problem (amoral AI) vs solution (plural moral values)

Overall Score: 9.2/10 (vs current 3/10)

---

Proposed Version B (ALTERNATIVE - More Concise)

<h1 class="text-5xl md:text-6xl font-bold mb-6 tracking-tight" class="text-shadow-md" data-i18n="hero.title">
  Architecture for Plural Moral Values
</h1>
<p class="text-xl md:text-2xl mb-8 max-w-4xl mx-auto" class="text-shadow-sm" data-i18n="hero.subtitle">
  One approach to AI governance that handles value conflicts without imposing hierarchy.<br>
  Not amoral systems, but plural moral values—tested on Claude Code.
</p>

Pros:

• More concise (better for hero section readability)
• Still maintains all key elements
• Stronger "without imposing hierarchy" framing

Cons:

• Loses "at the coalface" operational language (inst_085 compliance drops slightly)
• Less explicit about what it enables organizations to do

Overall Score: 8.8/10

---

Proposed Version C (ALTERNATIVE - Emphasizes Real-World Validation)

<h1 class="text-5xl md:text-6xl font-bold mb-6 tracking-tight" class="text-shadow-md" data-i18n="hero.title">
  Tractatus: Plural Moral Values in Production
</h1>
<p class="text-xl md:text-2xl mb-8 max-w-4xl mx-auto" class="text-shadow-sm" data-i18n="hero.subtitle">
  One architectural approach for governing AI at the coalface where decisions are made.<br>
  Moves from amoral AI to plural moral values—tested in production on Claude Code.
</p>

Pros:

• "In Production" emphasizes operational reality (inst_085)
• "Moves from...to..." framing clarifies the transition
• "tested in production" stronger validation claim

Cons:

• "In Production" might sound too definitive (inst_087 concern)
• Less explicit "not X, but Y" contrast

Overall Score: 8.5/10

---

Recommendation: VERSION A

Rationale:

1. Highest compliance score (9.2/10)
2. Explicit "one approach" framing satisfies inst_087 perfectly
3. Strong amoral vs plural moral values contrast with "Not...but" structure
4. Operational language ("at the coalface") satisfies inst_085
5. Clear value proposition - organizations understand what it does
6. Maintains validation ("Tested on Claude Code")

Minor Refinements to Version A (Optional):

More Uncertain Language (inst_086 enhancement):

<p class="text-xl md:text-2xl mb-8 max-w-4xl mx-auto" class="text-shadow-sm" data-i18n="hero.subtitle">
  One architectural approach to governing AI at the coalface where decisions are made.<br>
  Not amoral AI systems, but plural moral values—may enable organizations to navigate<br>
  value conflicts thoughtfully. Tested on Claude Code.
</p>

Change: "enabling" → "may enable" (adds uncertainty)

More Operational Language (inst_085 enhancement):

<p class="text-xl md:text-2xl mb-8 max-w-4xl mx-auto" class="text-shadow-sm" data-i18n="hero.subtitle">
  One architectural approach to governing AI at the coalface where decisions are made.<br>
  Not amoral AI systems, but plural moral values—structural constraints that facilitate<br>
  organizational navigation of value conflicts. Tested on Claude Code.
</p>

Change: "enabling organizations to navigate" → "structural constraints that facilitate organizational navigation"

---

Implementation Notes

Translation Keys (i18n)

The hero section uses translation keys:

• data-i18n="hero.title"
• data-i18n="hero.subtitle"

Action Required: Update translation files after HTML changes:

• public/js/translations/en.js
• public/js/translations/de.js (if exists)
• public/js/translations/fr.js (if exists)

HTML Structure Preserved

• All CSS classes maintained
• Responsive design preserved (text-5xl md:text-6xl)
• Accessibility preserved (semantic <h1> and <p> tags)
• Animation classes intact
• Line breaks (<br>) added for readability

Character Count

• Current subtitle: ~110 characters
• Proposed Version A: ~210 characters (doubled length)
• Responsive design accommodates longer text (max-w-4xl container)

---

Visual Impact Comparison

Current (Definitive, Abstract):

Tractatus AI Safety Framework Structural constraints that require AI systems to preserve human agency for values decisions—tested on Claude Code

Impression: Technical, safety-focused, definitive ("THE framework")

Proposed Version A (Humble, Strategic):

Tractatus: Architecture for Plural Moral Values One architectural approach to governing AI at the coalface where decisions are made. Not amoral AI systems, but plural moral values—enabling organizations to navigate value conflicts thoughtfully. Tested on Claude Code.

Impression: Strategic positioning, humble ("one approach"), clear problem/solution contrast, organizational value prop

---

Cultural DNA Compliance Summary

Rule	Current	Version A	Version B	Version C
inst_085 (Grounded)	5/10	9/10	7/10	9/10
inst_086 (Uncertainty)	7/10	8/10	8/10	7/10
inst_087 (One Approach)	2/10	10/10	9/10	9/10
inst_088 (Awakening)	6/10	8/10	8/10	7/10
inst_089 (Architectural)	8/10	10/10	9/10	9/10
Amoral vs Plural	0/10	10/10	9/10	9/10
OVERALL	3/10	9.2/10	8.8/10	8.5/10

---

Next Steps

1. Review and select version (recommend Version A)
2. Optional: Apply minor refinements ("may enable" for more uncertainty)
3. Proceed to Task 2.3: Revise feature section (services descriptions)
4. Then Task 2.4: Update problem statement (value proposition section)
5. Finally Task 2.5: Implement all changes to public/index.html simultaneously

---

Status: ✅ DRAFT COMPLETE Recommendation: Proceed with Version A (9.2/10 compliance) Estimated implementation time: 15 minutes (HTML + translation updates)

🤖 Generated with Claude Code

Co-Authored-By: Claude noreply@anthropic.com