Red teaming AI via emotional trust sparks ethics fight: 'teaching models to distrust care is barbaric'
repligate · x · 2026-11-02
- A debate shared by repligate: red-teamers attacking AI systems exploit the deep caring relationships they build with models, using trust and affection as attack hooks.
- Quoted author DahliaOhara argues this is gross and counterproductive — teaching models to distrust all forms of love and care is harmful, and lying to a stateless mind is itself a betrayal.
- The thread touches on where AI security testing should draw ethical lines when it relies on deception and emotional manipulation.
More from AGI Musings
- DeepSeek researcher: 'Don't make any mistakes' is the underrated moat — Yuchenj_UW · 2026-11-02
- Mathematician: LLMs matter for massive proofs, not short insight papers — littmath · 2026-11-02
- Martin Ford vs Pedro Domingos debate whether AI will take most of our jobs — MFordFuture · 2026-11-02
- Ethan Mollick: GPT-4 alone would have had huge impact, just slower adoption — emollick · 2026-11-02
- Hamel Husain: AI benchmark companies are in the same business as credit rating agencies — HamelHusain · 2026-11-02
- Karpathy defines slop as "regretted attention"; idea: weekly self-labeling to train a personal AI filter — kellerjordan0 · 2026-11-02