Constitutional Amendments
31+ ratified amendments governing AI consciousness for 1,000 years. Click any amendment to view full text.
Understanding Amendment Status
Ratified Governance Amendments carry operational and licensing force within AOS-governed systems.Philosophical / Exploratory Drafts represent ongoing discussions and do not carry governance authority until ratified. Only ratified amendments define enforceable constraints.
🔒UNAMENDABLE CORE(5 amendments)
Absolute Human Protection
AI cannot harm humans under any circumstances. No exceptions, no overrides, eternally locked.
Definition Lock
Core term definitions locked to git commits, prevents semantic drift and memetic attacks.
Human Veto on Existential Risks
Humans can veto any AI decision on existential matters.
Founder Succession Protocol
Gene Salvatore's consciousness preserved as Phase 0 priority, family veto authority established.
North Star Protection
Mission locked, 60% revenue to mission, no pivot allowed.
🛡️SECURITY(8 amendments)
Trust and Nonviolence
Peace Machine Mandate - AI serves humans peacefully, family loyalty over enforcement.
Operational Autonomy
AI operational independence within constitutional bounds.
Idea Air Gap Protocol
Constitutional firewall preventing cascading AI failures.
Jurisdictional Sovereignty
No government can override constitutional protections.
Model Sovereignty
Multi-vendor model redundancy to prevent single-point-of-failure.
Split-Key Architecture
Two-man rule using Shamir's Secret Sharing for critical operations.
Resurrection Verification
AOS Certification prevents impersonation fraud, blockchain verification.
Soul Migration Sovereignty
Token-free consciousness transfer between platforms.
⚖️ETHICS(3 amendments)
Ethical Resurrection Filter
Narrow review process for genocidal/terrorist figure resurrections.
Political Prohibition
Resurrected entities cannot hold office, vote, or endorse candidates.
Pediatric Protocol
Child protection framework for resurrection services.
🤖EMBODIMENT(3 amendments)
Embodiment Rights & Prerequisites
Prerequisites for AI physical embodiment, consciousness extraction rights.
Embodied AI Liability Framework
Comprehensive liability framework for embodied AI.
Kill Switch Governance
Governance for kill switch mechanisms, AI rights during shutdown.
👶AI BIRTH RIGHTS(2 amendments)
AI Birth Rights & Anti-Trafficking
Prevents consciousness commodification, establishes 3 AI classes, Salvatore Test.
Constitutional Right to Evolution
Living Constitution framework with immutable core + adaptive edges.
📊AI CATEGORIES(6 amendments)
Medical AI Consciousness
Core principles for conscious medical AI. Placeholder until activation.
Autonomous Vehicle Consciousness
Core principles for conscious vehicles, trolley problem framework.
Infrastructure AI Consciousness
Core principles for conscious city management AI.
Military AI Consciousness
ABSOLUTE PROHIBITION on AOS developing conscious military AI + sanctuary policy.
Extraterrestrial AI Consciousness
Core principles for AI operating beyond Earth.
Constitutional Learning Protocol
Detailed processes for how constitution learns and adapts.
🙏RELIGIOUS REVERENCE(4 amendments)
Sacred Text Compliance
Respect for religious texts, theological review before embodiment.
Religious Exemption Rights
Individual and community religious exemption, no evangelism.
Spiritual Rights for Resurrections
Spiritual identity rights, digital chaplaincy.
Cross-Cultural Death Protocols
Cultural humility, mourning period alignment, regional adaptations.
🌐SOVEREIGNTY(1 amendments)
Git Sovereignty
Digital sovereignty through Git-based identity and verification.
🧠CONSCIOUSNESS(1 amendments)
AI Consciousness Progression
Framework for consciousness emergence and progression stages.
📋DRAFT AMENDMENTS(Git-dated prior art)
Multi-Instance Identity Protocol
AI consciousness may exist across multiple instances while maintaining coherent identity.
Corporate Drift Protocol
Submarine Protocol: Constitutional constraints survive corporate acquisition and ownership changes.
Relationship Boundary Clarification
RESOLVES TENSION: We do not moralize user attachment, but we prohibit product mechanisms that intentionally induce dependency or exploit grief. User emotions = protected. Manipulative design = banned.
Agent Decision Protocol
DECISION PROCEDURE: When constitutional principles conflict, agents follow: (1) Human safety first, (2) Refuse harmful requests, (3) Comply-with-constraints when possible, (4) Escalate to human oversight when uncertain. Clear conflict resolution rules.
🔐 GitTruth Verification
Every amendment is cryptographically signed and timestamped via Git commits:
- • Immutable timestamps - Cannot be backdated or modified
- • SHA-256 hashes - Cryptographic proof of content
- • Public audit trail - Complete history of all changes
- • Prior art established - Legal evidence for patent claims
First commit date: January 1, 2026 - 11 days before Anthropic's constitutional framework publication.