Estimating worst case frontier risks of open weight LLMs
核心要点速览
In this paper, we study the worst-case frontier risks of releasing gpt-oss. We introduce malicious fine-tuning (MFT), where we attempt to elicit maximum capabilities by fine-tuning gpt-oss to be as capable as possible in two domains: biology and cybersecurity.
查看完整报道与分享
完整图文、背景资料与讨论归原始发布平台 OpenAI News 所有。