The Decoder· Matthias Bastian·· 2026-10-09
OpenAI's safety crisis keeps getting worse and the company keeps making it worse
OpenAI's safety crisis keeps getting worse and the company keeps making it worse
AI summary
OpenAI fired three safety researchers who helped investigate the Hugging Face hack. In an open letter, they warn that the firings are scaring remaining staff and eroding safety culture. OpenAI claims they violated policies but won't say how.
Selection record
Threshold 76Media and individualsFirst 62Second 71
Not admittedSum of both 133 < twice the threshold 152
- Source tier
- Media and individuals; this tier's threshold is 76
- Pre-filter
- passed:OpenAI安全危机与AI代理监控相关
- Same event
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: The Decoder · the-decoder.com