The Automated Dispute Resolution (ADR) is the means with which 📝Guardians Assurance enforces it's policies.

The Claims Process: Workflow Architecture

The Objective: To process "Breach of Decency" claims efficiently, ensuring safety for the Claimant (the person dating the Assured) and due process for the Assured Policyholder.

Phase 1: Intake & Triage (The "FNOL" - First Notice of Loss)

Automated Chatbot / Survey Form

1. Policy Validation: Claimant inputs the Assured’s Policy ID (or looks them up via phone/email).
1. The Safety Gating (Critical Step):

• • Question: "Do you feel physically unsafe, or do you fear retaliation if the Assured finds out about this claim?"
  • If YES: The system locks the claim. No notification is sent to the Assured. The Claimant is routed to the "Protective Resources Toolkit" (local DV resources, guides on safe exit strategies). The claim is stored silently as "Shadow Data" to track patterns but not acted upon until safety is confirmed.
  • If NO: Proceed to Breach categorization.

1. Breach Categorization:

• • Claimant selects which of the 10 Principles was violated (e.g., 🏷️#3 Compassionate Closure, 🏷️#8 Clean Fighting).

Phase 2: The Accountability Pre-Check

Logic Gate: Have they tried to solve it themselves?

1. The Direct Communication Check:

• • Question: "Have you communicated this grievance to the Assured directly?"
  • If YES: Proceed to Evidence Collection.
  • If NO: Stop. The system enters Coaching Mode.

1. AI Coaching Mode (The "Pre-Claim Support"):

• • The system acknowledges that confrontation is hard.
  • The Output: The AI generates a "Script of Accountability" for the Claimant to send to the Assured.
  • • Example: "I’m helping you write a text that adheres to Non-Violent Communication. Send this: 'Hey [Name], when you [Action], I felt [Emotion]. I need [Requirement]. Can we discuss this?'"
  • The Strategy: This serves two purposes:

1. 1. 1. It attempts low-level repair.
      1. It creates the paper trail/evidence required for the claim if the Assured responds poorly.

Phase 3: Evidence Discovery

Upload Portal

1. The "Chain of Evidence":

• • If the Claimant has spoken to the Assured, the system asks: "How did they respond?"
  • Options: Denied it / Gaslit me / Ignored me / Attacked me / Admitted but didn't fix.

1. Digital Forensics:

• • Claimant uploads screenshots of texts, emails, or call logs.
  • AI Analysis: An LLM (Large Language Model) scans the text for keywords indicative of gaslighting, deflection, or DARVO (Deny, Attack, and Reverse Victim and Offender).
  • If no evidence exists: The system advises: "Please attempt the conversation via text/email using our generated script to document their response, then return to file."

Phase 4: The Notice of Claim

Contacting the Assured

Once the "Prima Facie" case (base evidence) is established, the Assured receives a notification.

• The Notification: "A Claim of Breach regarding Principle 🏷️#4 (Enthusiastic Consent) has been filed against your policy. Policy 🏷️#88392 is now under review."
• The "Deposition": The Assured is sent a secure link to a survey to tell their side.
• • They are not shown the raw screenshots yet (to prevent specific retaliation), but the general nature of the complaint.
  • AI Assessment: The AI analyzes the Assured's response. Is it defensive? Apologetic? Does it contradict the screenshots provided by the Claimant?

Phase 5: Adjudication & The "Deductible"

The Resolution

The system compares: Claimant Story + Evidence vs. Assured Story.

1. Verdict: "Breach Verified"

• • The Assured is found in violation of the code.

1. The "Repair Deductible":

• • Instead of money, the Assured must pay a "Deductible of Labor" to keep their policy active.
  • Tier 1 (Minor Breach - e.g., Flaking): Must send a formal apology letter (reviewed by AI for sincerity) and read a specific article/guide.
  • Tier 2 (Moderate Breach - e.g., Gaslighting): Mandatory Async Mediation.
  • • AI Mediation: The AI acts as the go-between. It asks the Assured prompting questions ("The Claimant felt unheard. How could you have phrased that differently?"). The Assured must successfully complete the dialogue tree to "pass."
  • Tier 3 (Major Breach): Policy Revocation. The Assured loses their "Certified Good Person" status.

Contexts

• 🏷️#guardians-assurance (See: 📝Guardians Assurance)