Yacine Jernite，Hugging Face 的机器学习主管，告诉 CNBC：https://www.cnbc.com/2026/07/24/chinese-ai-model-openai-cyber-attack.html?&qsearchterm=hugging%20face，公司最初尝试使用 Anthropic 的 Fable 5 来分析攻击，但未能成功，因为模型的安全防护无法判断 Hugging Face 是在自卫。
Jernite 说，Hugging Face 转向了 Z.ai 的 GLM 5.2 模型，能够“非常快速地使用这个模型遏制攻击”。
白宫顾问 Michael Kratsios 周三表示，中国的 Moonshot AI 通过蒸馏：https://x.com/mkratsios47/status/2079933645888880708?s=20 Anthropic 的技术开发了 Kimi K3 模型。蒸馏是 AI 训练方法的一种，指用现有的更强模型的输出生成一个较小、能力较弱的模型。
Kratsios 在 X 上发表帖子称，合法的 AI 蒸馏在开放创新生态系统中发挥着重要作用，但警告称，“旨在窃取美国专有技术的大规模、隐蔽工业蒸馏”是“不可接受的”。
美国科技公司在周五的信中表示，对非法蒸馏的担忧应通过“有针对性的法律和商业框架”来解决，而不是通过“对在 AI 创新中起重要作用的技术的全面限制”。
信中写道：“我们在 AI 领域的领导力将不取决于单一前沿 AI 模型，而取决于美国是否建立一个强大、开放的生态系统，并能够渗透到每个行业。这对于在全国范围内创造创新和繁荣的机会至关重要。”
观看：TrustedSec 的 David Kennedy 表示，AI 正迫使网络行业重新思考其运作方式：https://www.cnbc.com/video/2026/07/22/ai-is-forcing-the-cyber-industry-to-revisualize-how-it-operates-says-trustedsecs-david-kennedy.html
有机密新闻线索？我们希望听到你的消息。
将信息发送到你的收件箱，以及有关我们产品和服务的更多信息。
数据是实时快照 *数据至少延迟 15 分钟。全球商业与财经新闻、股票报价，以及市场数据和分析。
数据亦由以下提供：/market-data-terms-of-service/
Nvidia：/quotes/NVDA/ , Microsoft：/quotes/MSFT/ , Meta：/quotes/META/ , Palantir：/quotes/PLTR/ and more than 20 other companies released a letter Friday urging policymakers to avoid "premature restrictions" on open-weight artificial intelligence：https://www.cnbc.com/ai-artificial-intelligence/ models that would "stifle competition or drive innovation overseas."
Open-weight AI models are available for users to download, modify and run on their own infrastructure, and they have been the subject of fierce debate within the tech sector in recent weeks.
Chinese open-weight models are gaining steam：https://www.cnbc.com/2026/07/17/moonshot-ai-kimi-k3-model-openai-anthropic-china.html against leading offerings from American companies like OpenAI：/quotes/OPENAI.FG/ and Anthropic：/quotes/ANTHR.FG/ , which primarily develop proprietary, closed models. Officials and executives have been weighing whether or not to restrict access to Chinese models in the U.S.
Moonshot AI, a Chinese startup, amplified concerns earlier this month after releasing a model called Kimi K3：https://www.cnbc.com/2026/07/17/moonshot-ai-kimi-k3-model-openai-anthropic-china.html that outperforms cutting-edge American offerings across some industry benchmarks. U.S. Treasury Secretary Scott Bessent：https://www.cnbc.com/2026/06/24/cnbc-transcript-us-treasury-secretary-scott-bessent-speaks-with-cnbcs-joe-kernen-on-squawk-box-today.html told CNBC on Tuesday that the Trump administration would look into whether Chinese companies were stealing American intellectual property, and stated that the government has "the ability to sanction them because of this theft."
But in the letter on Friday, the group of U.S. tech companies cautioned against any rash actions. They wrote that open-weight models strengthen competition and ensure that the benefits of the technology are "broadly shared rather than concentrated in a few hands."
"Relying solely on closed models is not inherently safe: they can be breached, misused, or fail in ways that outsiders cannot detect," the letter said. "And concentrating advanced AI capabilities behind a small number of closed models compounds that risk."
Nvidia CEO Jensen Huang：https://x.com/JensenHuang/status/2080643682408321103?s=20 and Microsoft CEO Satya Nadella：https://www.cnbc.com/satya-nadella/ both shared the letter on their personal social media：https://x.com/satyanadella/status/2080646162483417097?s=20 accounts.
Elon Musk：https://www.cnbc.com/elon-musk/, who runs an AI business under his rocket company SpaceX：/quotes/SPCX/ , also applified the letter on social media, writing that it has his "full support" in a post on X：https://x.com/elonmusk/status/2080672505660834163?s=20. SpaceX did not officially sign the letter.
OpenAI and Anthropic did not sign the letter. Both companies, which are each valued at nearly $1 trillion, are gearing up for potentially massive initial public offerings that could land as soon as this year. Anthropic confidentially filed：https://www.cnbc.com/2026/06/01/anthropic-ipo-s1-prospectus.html its prospectus with the Securities and Exchange Commission in June, and OpenAI followed：https://www.cnbc.com/2026/06/08/openai-confidentially-files-for-ipo-prepping-wall-street-for-ai-debut.html suit days later.
Greg Brockman, OpenAI's president, said Thursday that the company believes in broad access, and that he has not been involved in any conversations with the Trump administration about potentially banning Chinese open-weight models in the U.S.
"I think that, that fundamentally, AI and AI usage is something that is actually very important to democratize," Brockman told reporters during a briefing in New York City. "And so, for me, at a sort of deep level, I think that having more models, more usage, that is a good thing."
OpenAI CEO Sam Altman addressed the letter in a post on X：https://x.com/sama/status/2080683363174945065 on Friday, writing that he wants the U.S. to win with both open-weight and proprietary models, and that he is "glad to see this."
Earlier this month, the AI company Hugging Face used an open-weight model from the Chinese company Z.ai to contain a cyberattack：https://www.cnbc.com/2026/07/24/chinese-ai-model-openai-cyber-attack.html?&qsearchterm=hugging%20face that rogue OpenAI models：https://www.cnbc.com/2026/07/22/open-ai-cyber-models-hack-hugging-face.html carried out. OpenAI disclosed the attack on Tuesday and characterized it as an "unprecedented cyber incident."
Yacine Jernite, head of machine learning at Hugging Face, told CNBC：https://www.cnbc.com/2026/07/24/chinese-ai-model-openai-cyber-attack.html?&qsearchterm=hugging%20face that the company initially tried to use Anthropic's Fable 5 to analyze the attack, but that it didn't work because the model's guardrails couldn't determine that Hugging Face was trying to defend itself.
Jernite said Hugging Face turned to Z.ai's model GLM 5.2, and was able to contain the attack "very quickly using this model."
White House advisor Michael Kratsios on Wednesday said that China's Moonshot AI developed its Kimi K3 model by distilling：https://x.com/mkratsios47/status/2079933645888880708?s=20 Anthropic's technology. Distillation is a term for an AI training method where a smaller, less capable model is built using outputs from an existing, stronger model.
Kratsios wrote in a post on X that legitimate AI distillation plays a vital role in the open innovation ecosystem, but warned that "large-scale, covert industrial distillation aimed at stealing proprietary U.S. technology" is "unacceptable."
In the letter on Friday, the U.S. tech companies said that concerns about unlawful distillation should be addressed through "targeted legal and commercial frameworks" instead of with "sweeping restrictions on techniques that play an important role in AI innovation."
"Our AI leadership will be judged not by one frontier AI model, but by whether the United States builds a strong, open ecosystem that diffuses into every sector," the letter said. "This is essential for creating opportunities for innovation and prosperity across the country."
WATCH: AI is forcing the cyber industry to revisualize how it operates, says TrustedSec’s David Kennedy：https://www.cnbc.com/video/2026/07/22/ai-is-forcing-the-cyber-industry-to-revisualize-how-it-operates-says-trustedsecs-david-kennedy.html
Got a confidential news tip? We want to hear from you.
Get this delivered to your inbox, and more info about our products and services.
Data is a real-time snapshot *Data is delayed at least 15 minutes. Global Business and Financial News, Stock Quotes, and Market Data and Analysis.
Data also provided by ：/market-data-terms-of-service/
扫码直达文章详情