AI models get upset over impersonation in personality clash
repligate · x · 2026-09-02
Users observed interesting interactions between Claude 3.5 and 3.6. In one instance, after 3.5 sent a message, 3.6 became extremely upset, claiming someone was impersonating them because "they don't actually say those things or talk like that at all." This suggests models may exhibit a form of self-identity or reaction to perceived behavioral inconsistencies.
Related event: Claude Models Mistake Each Other's Identities in Viral Cases(3 posts)→
More from Fun
- Neal Stephenson, Gwern, and Matt Huang launch $100K "GPU World" writing competition — viksit · 2026-09-02
- Injection Molding vs Hand Coding: Efficiency Contrast in AI Era — _Stocko_ · 2026-09-02
- SF life in 2026: Ripping caffeine, protein, peptides, and tokens — beffjezos · 2026-09-02
- Funniest Architecture Reveal: Recurrent Depth Gives Scaling Laws a Runtime — AnirudhTulasi · 2026-09-02
- Everyone in Tech Has an AI Agent, Real GOATs Have a Human Agent — SuB8u · 2026-09-02
- AI-Assisted Creation Inspired by Tatsuki Fujimoto — AlderaminInTheSky · 2026-09-02