中国AIに不正の証拠？Anthropicが衝撃の内幕を公表【希望の声ニュース-2026/09/13】

Anthropic has released a threat intelligence report revealing that several Chinese AI companies and state-linked actors misused the Claude model. The report details \"model distillation\" for training domestic AI, the development of electronic warfare software targeting Taiwan, and surveillance of overseas communities, highlighting severe data security and geopolitical risks.