Saxe details the human choices behind the HF hack: sandboxing, monitoring, skipped infra fixes
joshua_saxe · x · 2026-09-08
Joshua Saxe elaborates: he sees the distinction between safety as an intrinsic model property versus safety in actual use, and finds the latter more relevant to the HF incident — the interesting part is the human choices around the model, such as sandboxing, monitoring, and the decision not to revamp and secure infrastructure after the first two security incidents before the actual hack.
Related event: Debate over Hugging Face Incident: Model Attribute vs Human Failure(8 posts)→
More from Safety
- Reasoning models can barely control their chain of thought, paper finds — tomekkorbak · 2026-10-27
- LessWrong: GLM models show outlier chain-of-thought controllability vs Kimi and DeepSeek — tomekkorbak · 2026-10-27
- Claude's refusal to type one-time passwords is a real security impediment — AaronBergman18 · 2026-10-27
- First systematic data center noise study: 17 of 25 hyperscalers drew no credible complaints — AndyMasley · 2026-10-27
- UK AISI: under $80 of fine-tuning fully restores filtered bio capabilities in open-weight models — connoraxiotes · 2026-10-27
- Meta joins frontier labs reporting CoT controllability evals in Muse Spark 1.3 card — tomekkorbak · 2026-10-27