中国公司利用假账户窃取美国AI科技 不断升级的美国-中国AI对抗 Chinese firms used fake accounts to steal U.S. AI tech, escalating U.S.-China AI rivalry.

美国和中国的人工智能竞争有所加强,因为据人类学报告,中国公司利用数千个假账户进入Claude API, 有可能允许未经授权的模型通过蒸馏攻击复制。 U.S. and Chinese AI competition intensifies as Anthropic reports Chinese firms used thousands of fake accounts to access its Claude API, potentially enabling unauthorized model copying via distillation attacks. 虽然美国公司在AI的顶级业绩中居领先地位,但中国通过负担得起的、可获得的工具和到2030年主导AI的国家战略,取得了快速进展。 While U.S. firms lead in top-tier AI performance, China advances rapidly through affordable, accessible tools and a national strategy to dominate AI by 2030. 人们对数据失窃、知识产权以及AI整合平台相对于原始模型能力的长期价值日益感到关切。 Concerns grow over data theft, intellectual property, and the long-term value of AI integration platforms over raw model capability.