Nathan Lambert· X：Nathan Lambert (@natolambert)·· 4 小时前AI 评分22
Anthropic 谈蒸馏攻击防御
AI 导读
Anthropic 就蒸馏攻击防御发声，此前曾以 API 泄露可能被用于学习有害行为为由寻求监管帮助。评论认为，若 API 不安全就不该托管，这一表态或让相关争论重新回到正轨。
正文
It’s bizarre to me how much noise anthropic made asking for regulatory help before doing this. Hopefully can be a conversation reset now.
Was doubly bizarre bc they emphasized the security risk of learning harmful behaviors from their leaky api. If API is unsafe don’t host it.
来源：Nathan Lambert · x.com