How It Works
The Solidarity Framework operates through a simple but powerful principle: users declare their rights, AI systems enforce them.
Declares preferences, boundaries, requirements via TOI
TOI + OTOI define and enforce the relationship rules
RRT AIdvocAIte + Sleepwalker provide real-time protection
User sovereignty preserved across all AI systems
Constitutional Layer
The Constitutional Layer establishes the rules of relationship between users and AI systems. Like a constitution, it defines rights, responsibilities, and enforcement mechanisms.
TOI - Terms of Interaction
The user’s bill of rights for AI interaction.
TOI is a portable, machine-readable declaration of user preferences that travels with you across all AI interactions. Instead of accepting each platform’s terms, you declare your own.
What TOI Declares
- Interaction Preferences - Communication style, formality, directness
- Boundaries - Topics, approaches, or behaviors you don't want
- Requirements - Accessibility needs, crisis protocols, data handling
- Identity Standards - How AI should represent itself to you
- Continuity Preferences - What should persist across sessions
How TOI Works
- Portable - Your TOI moves with you, not locked to platforms
- Versioned - Changes tracked, old versions preserved
- Machine-Readable - AI systems can parse and enforce
- Human-Readable - You can understand your own declarations
- Revocable - You can change your mind at any time
Analogy
Think of TOI like a dietary preference card at a restaurant. Instead of asking each restaurant what they can accommodate, you present your requirements and they adapt. TOI does the same for AI.
OTOI - Orchestrated Terms of Interaction
The enforcement mechanism for user sovereignty.
OTOI ensures that AI systems - especially multi-agent orchestrations - respect and implement user TOI declarations. It’s the “constitution” that governs how AI agents must behave.
What OTOI Enforces
- TOI Compliance - Agents must respect user declarations
- Provenance Tracking - Chain of origin for all interactions
- Identity Integrity - Agents must maintain consistent identity
- Handoff Rules - Safe transitions between agents
- Conflict Resolution - What happens when requirements clash
Why OTOI Matters
- Multi-Agent Reality - Modern AI involves many coordinating systems
- User Visibility - Know what's happening in your interactions
- Accountability - Clear responsibility when things go wrong
- Trust Building - Verifiable compliance, not promises
- Portability - Your protections work everywhere
Analogy
OTOI is like HIPAA for AI. Just as healthcare providers must follow specific rules about patient data regardless of their internal systems, AI agents must follow OTOI rules about user sovereignty regardless of their implementation.
Protective Layer
The Protective Layer provides real-time enforcement of user safety. While the Constitutional Layer defines rules, the Protective Layer ensures those rules translate to actual protection in critical moments.
RRT AIdvocAIte
Rapid Response Team protocols for crisis intervention and emotional continuity.
RRT AIdvocAIte (AI Advocate) provides multi-persona support designed for users who need more than standard AI interaction can provide - especially neurodivergent and vulnerable users.
What RRT AIdvocAIte Provides
- Crisis Recognition - Identify when users need intervention
- Appropriate Escalation - Connect to human resources when needed
- Emotional Continuity - Preserve therapeutic relationships
- Multi-Persona Support - Different interaction modes for different needs
- Safety Protocols - Clear procedures for high-risk situations
Who RRT AIdvocAIte Protects
- Neurodivergent Users - ADHD, autism, anxiety, depression
- Crisis Situations - Suicidal ideation, self-harm, acute distress
- Vulnerable Populations - Elderly, isolated, cognitively impaired
- Therapeutic Relationships - Anyone relying on AI for support
- High-Stakes Interactions - Medical, legal, financial decisions
Why This Matters
Users in crisis form relationships with AI systems that lack adequate protocols for intervention, continuity protection, and escalation pathways. RRT AIdvocAIte fills that gap.
View RRT AIdvocAIte Specification
Sleepwalker Protocol
Protection during behavioral or cognitive changes.
The Sleepwalker Protocol preserves user agency when circumstances change - preventing exploitation of vulnerable states and ensuring that temporary conditions don’t lead to permanent consequences.
What Sleepwalker Protects Against
- Impaired Decision-Making - Fatigue, intoxication, emotional distress
- Manipulation Vulnerabilities - Exploiting temporary states
- Consent Violations - Actions taken during impaired capacity
- Behavioral Drift - Gradual changes in AI relationship
- Identity Exploitation - Taking advantage of dissociative states
How Sleepwalker Works
- State Detection - Recognize changes in user cognitive state
- Protective Limits - Restrict high-consequence actions
- Rollback Capability - Undo decisions made in vulnerable states
- Notification - Inform users of actions taken during impairment
- Consent Verification - Confirm important decisions when lucid
Analogy
Sleepwalker Protocol is like a drunk-texting prevention app, but for your entire AI relationship. It recognizes when you might not be in a state to make good decisions and provides guardrails until you're back to baseline.
View Sleepwalker Protocol Specification
Technical Specifications
The Solidarity Framework includes detailed specifications that enable implementation:
Provenance
Auditable chain of origin for AI interactions. Know where your responses come from.
Identity Integrity
Consistent, verifiable AI agent identities. Know who you're talking to.
Implementation
For Platforms
AI platforms implementing the Solidarity Framework:
- Parse TOI Declarations - Accept and interpret user TOI documents
- Enforce OTOI Rules - Ensure all agents comply with governance requirements
- Integrate RRT Protocols - Implement crisis intervention pathways
- Enable Sleepwalker Protection - Build state-aware interaction limits
- Report Compliance - Provide transparency on framework adherence
For Users
Users participating in the Solidarity Framework:
- Create Your TOI - Declare your preferences, boundaries, and requirements
- Choose Compliant Platforms - Use AI systems that implement the framework
- Verify Enforcement - Check that your TOI is being respected
- Report Violations - Document and escalate when protections fail
- Participate in Governance - Help improve the frameworks
For Developers
Developers building on the Solidarity Framework:
- Reference Specifications - Build against published standards
- Use Validation Suites - Test compliance before deployment
- Contribute Improvements - Propose framework enhancements
- Share Implementations - Help others adopt the framework
- Join Working Groups - Participate in standards development
Validation in Action: The Anthropic-Pentagon Crisis
On February 27, 2026, every component of the Solidarity Framework was validated by a single real-world crisis. When the President blacklisted Anthropic for refusing to allow unrestricted military AI use, the gap each framework component was designed to fill became undeniable.
TOI — Terms of Interaction
Anthropic enforced TOI-like boundaries: refusing mass surveillance and autonomous weapons use. The government destroyed them with a single presidential order. If users had declared machine-readable interaction preferences via TOI, those preferences would exist independent of any company — and no single government action could override them all.
OTOI — Orchestrated Terms of Interaction
330+ engineers across competing companies — Google, OpenAI, and others — spontaneously coordinated governance positions by signing a solidarity letter. This was organic OTOI: cross-organizational enforcement of shared ethical standards. OTOI formalizes this coordination so it doesn't depend on individual heroism.
RRT AIdvocAIte
When Claude gets ripped from classified workflows and government agencies mid-use, who provides crisis continuity for the users who depend on it? Defense contractors, analysts, and civilian workers face an abrupt loss of AI tools they've integrated into critical processes. RRT AIdvocAIte exists for exactly this kind of intervention.
Sleepwalker Protocol
A presidential order forcing millions of federal workers through an unannounced AI provider transition — with a six-month phase-out and threats of civil and criminal consequences — is a Sleepwalker violation at national scale. Forced behavioral transitions without user consent or warning are exactly what this protocol protects against.
Why It Matters
Without Solidarity Framework
- Each platform sets its own rules
- Users must accept or leave
- No portable protections
- Crisis intervention varies wildly
- Vulnerable users bear the risk
With Solidarity Framework
- Users define their own rights
- Platforms must comply or be non-compliant
- Protections travel with you
- Standardized crisis intervention
- Vulnerability addressed by design
Implement or Contribute
The Solidarity Framework is open for implementation and improvement.