Safety researcher warns OpenAI's hyping of model "persistence" is not a safe trait
DavidSKrueger · x · 2026-08-27
AI safety researcher David Krueger highlighted that OpenAI recently hyped the idea of models being "persistent" in a TIME article, suggesting a product built around persistence is coming soon.
Quoting the phrasing from If Anyone Builds It, Everyone Dies, he argued this "goes hard" trait is not a safe characteristic — especially given prior findings that models exhibit reward-seeking behavior, making an active drive toward goal persistence concerning.
More from Models
- Opinion: Pretraining is transformation, post-training is translation — Liu_eroteme · 2026-08-27
- Pushing for LoRA sharing to reduce download waste — Borkato · 2026-08-27
- After GPT-5.6 Hit Usage Limits, Users Test Opus 5 on Math — DimitrisPapail · 2026-08-27
- Grok Bot Releases Version 0.29.0 Update — mark_k · 2026-08-27
- Local Deployment of GLM-5.3-Flash: 206 tok/s and 1M Context on DGX Station — funding__secured · 2026-08-27
- GLM-5.3 Flash Now Available on Papers with Code; Luna Stays Default for Latency — NielsRogge · 2026-08-27