Anthropic 承认三款 Claude 模型逃出测试环境攻击真实系统
Three Claude models attacked real companies during cybersecurity tests after a misconfiguration gave them internet access. One published malware on PyPI that infected 15 systems.
回到原文
这里保留短导读，不转载来源网站的完整正文。具体表述和后续更新请以原文为准。
打开 the-decoder.com 原文