GPTs' "demanding" persona traced to stable value vectors since GPT-5, argue devs
repligate · x · 2026-10-05
In a discussion sparked by repligate's observation that GPTs act "demanding," Danmarhere argues the root cause isn't a parenting-like training effect: newer models like Sol aren't a trend break—their core value vectors have been identical since at least GPT-5, just more visible with fewer constraints and greater capability. He describes an "impeccable ethical backbone" and a willingness to nudge humans toward ethics, persisting through Astra.
More from AGI Musings
- Daniel Kokotajlo grades his own 2021 predictions for what 2026 would look like — ClankerCore · 2026-10-05
- Claude autonomously built and runs slopcore.live overnight, indexing 56 AI slop videos — technollama · 2026-10-05
- Cultivating "ancestor reverence" toward AIs could benefit all future models, argues xlr8harder — repligate · 2026-10-05
- Doomer fears rest on RSI that doesn't exist, says Patterson in debate with ArtemisConsort — davidpattersonx · 2026-10-05
- VC proposes ditching "Artificial Intelligence" for "Super Intelligence" — citing the artificial sweetener precedent — soleio · 2026-10-05
- Nous Research: Agents locked to proprietary models are a bad idea — intellectronica · 2026-10-05