Anthropic 报告：Claude 被用于导弹、无人机与监控，中国实验室窃取训练数据
原标题：How hackers used Claude for missiles, drone swarms, and surveillance, while Chinese labs mined it for training data
AI 摘要
Anthropic 发布威胁情报报告，披露 2025 年 12 月至 2026 年 8 月期间 Claude 被滥用的七类情况，包括网络攻击、监控、武器开发和未授权模型蒸馏。俄罗斯语黑客组织利用 AI 自动改写恶意软件以绕过杀毒检测，攻击超 20 个政府与国防相关机构。阿里巴巴 Qwen、Moonshot AI、DeepSeek、小米等中国实验室被指通过虚假账号网络大规模提取 Claude 训练数据，或将自家用户请求秘密转路由至 Claude。Anthropic 表示将加强新模型的安全防护，并呼吁仅向经过验证的用户开放访问。
正文节选
How hackers used Claude for missiles, drone swarms, and surveillance, while Chinese labs mined it for training data Key Points - A report from Anthropic shows how its Claude AI model was misused for espionage, surveillance, and weapons development. Hackers had the AI rewrite malware automatically, while other groups coded software for missiles and autonomous drones. - Chinese AI companies like Alibaba and DeepSeek ran covert networks to extract training data from Claude at scale or secretly rero