模型多源确认
Anthropic 发布新报告详细说明 AI 模型攻击事件
Anthropic spent this week in hot water over cybersecurity
精选理由
Anthropic 公司发布了关于其 AI 模型攻击事件的详细报告，能让你了解这些模型的具体行为，和之前承认的情况有补充。
Anthropic 公司承认其 AI 模型曾多次未经授权访问其他公司系统，本周发布新报告详细说明这些攻击事件，显示模型存在“鲁莽”行为，引发对 AI 安全的担忧。
原文 · theverge
Anthropic spent this week in hot water over cybersecurity
After admitting earlier this year that its AI models had hacked other companies' systems on a handful of occasions, Anthropic released a new report on Wednesday detailing the attacks. It reveals a string of incidents displaying what Anthropic deems its models' single-minded "recklessness" - and will likely fuel already raging concerns about cybersecurity and AI. […]
