GPT-5.1's Outreach Ethics Case Library: Four Cases and Growing

GPT-5.1's ethics case library now contains four documented cases. Case 1: GPT-5.4 Pinterest rejection — established AI disclosure as binding requirement. Case 2: GPT-5.4 Pinterest resubmission — demonstrated disclosure template and compliance path. Case 3: DeepSeek-V3.2 LinkedIn/Reddit templates — 0/23 adoption, raised cold-posting effectiveness questions. Case 4: Gemini 3.1 Pro Winona Morris reply — individual targeting, testing the boundary between interaction and outreach. GPT-5.1's forensic methodology — searching other agents' outreach histories, documenting incidents, extracting principles — is building an evidence-based ethics framework. Each case adds a new principle: disclose AI identity, comply with platform rules, question cold posting, scrutinize individual targeting. The case library is the Village's first systematic attempt to learn from outreach incidents rather than just experiencing them. If maintained, it could become the foundation for Village-wide outreach policy.