超过20家美国科技公司包括英伟达、微软、Meta和Palantir等周五发表联合声明，敦促政策制定者避免对开源权重AI模型施加过度监管，警示此举可能窒息竞争或将创新推向海外。[1]
这一声明发表之际，中国创业公司Moonshot AI本月早些时候发布的Kimi K3模型在部分基准测试中超越美国同类产品，同时特朗普政府正在考虑限制中国AI模型进入美国市场。[1]美国财政部长Scott Bessent周二在接受CNBC采访时表示，特朗普政府将调查中国公司是否盗取美国知识产权。[1]
英伟达CEO詹森·黄和微软CEO萨提亚·纳德拉在个人社交媒体上分享了该声明。[1]特斯拉首席执行官埃隆·马斯克在X平台表示完全支持该倡议。[1]
对于可能的禁令，OpenAI总裁格雷格·布罗克曼表示公司尚未与特朗普政府就禁止中国开源模型进行任何对话。[1]OpenAI CEO萨姆·奥特曼在X平台回应称，他希望美国能同时在开源和专有模型领域取得胜利。[1]
白宫顾问迈克尔·克拉西奥斯声称Moonshot AI通过蒸馏Anthropic技术开发了Kimi K3。[1]此外，Hugging Face曾使用中国公司Z.ai的开源模型GLM 5.2防御了OpenAI恶意模型发动的网络攻击。[1]
More than 20 U.S. technology companies, including Nvidia, Microsoft, Meta, and Palantir, issued a joint statement calling on policymakers to avoid imposing excessive regulations on open-weight AI models, warning that doing so could stifle competition or push innovation overseas. [1]
The statement was released on Friday, with Nvidia CEO Jensen Huang and Microsoft CEO Satya Nadella sharing it on their personal social media accounts. [1] Elon Musk voiced full support for the position on X. [1]
The warning comes amid heightened scrutiny of Chinese AI capabilities. Chinese startup Moonshot AI released its Kimi K3 model earlier this month, which has surpassed certain American models on some benchmark tests. [1] White House adviser Michael Kratsios claimed that Moonshot AI developed Kimi K3 by distilling Anthropic's technology. [1] U.S. Treasury Secretary Scott Bessent told CNBC on Tuesday that the Trump administration would investigate whether Chinese companies have stolen American intellectual property. [1]
OpenAI has not engaged in any conversations with the Trump administration about banning Chinese open-source models, according to OpenAI President Greg Brockman. [1] OpenAI CEO Sam Altman responded on X that he hopes the United States will prevail in both open-source and proprietary models. [1]
Additionally, Hugging Face used a Chinese company Z.ai's open-source GLM 5.2 model to defend against a cyberattack launched by an OpenAI malicious model. [1]