包括 Hugging Face、Meta、Microsoft、Mistral 和 Nvidia 在内的几家人工智能公司签署了一封公开信：https://www.microsoft.com/en-us/corporate-responsibility/topics/open-weight/，敦促政策制定者不要对开放权重的 AI 模型施加广泛的“过早限制”。这封信发布之际，华盛顿正在讨论美国应如何应对关于中国 AI 实验室窃取美国同行知识产权的指控，以及其日益增强的能力。
信中完全没有提到中国，但它发布的背景是，有报道称特朗普政府一直在考虑禁止中国的开放权重模型，并可能对来自该国的 AI 公司实施制裁：https://techcrunch.com/2026/07/22/treasury-threatens-sanctions-after-white-house-claims-moonshot-distilled-anthropics-fable/。白宫甚至指控 Moonshot AI 对 Anthropic 的 Fable 模型进行了蒸馏，以训练其最近发布的、在各方面都非常出色的 Kimi K3 模型：https://techcrunch.com/2026/07/18/kimi-threat-or-menace/
这封信似乎旨在阻止对中国模型的全面禁令，并确保政府对所谓中国蒸馏的回应不会蔓延到对开放权重 AI 或像蒸馏这样的通用技术的更广泛限制。
Hugging Face 表示：https://x.com/ClementDelangue/status/2079253659108409587，由于商业前沿 AI 模型的安全防护措施阻止了其努力，公司未能用这些模型抵御攻击。它所使用的封闭 AI 模型无法区分构建攻击者利用漏洞的请求与防御方试图检测漏洞的请求。公司因此不得不转向使用中国 AI 公司 Z.ai 的 GLM 5.2，一个强大的开放权重模型，以抵御攻击。
这封信凸显了 AI 行业的分裂。像 OpenAI：https://techcrunch.com/2026/07/20/openai-is-scared-of-open-weight-models-should-the-us-be/ 和 Anthropic 这样的公司敦促政府应对中国 AI 公司涉嫌的知识产权盗窃，因为开放权重模型的能力正快速增长。其结果可能对这些公司的商业模式产生重大影响，而这些模式正受到廉价、能力强大且易于获取的 AI 模型的冲击。
这些公司，以及其他封闭源代码 AI 开发者如 Google DeepMind 和 SpaceX，在这封信的底部显著缺席。
签署这封信的公司显然在看到开放 AI 模型兴旺方面有经济利益。像 Nvidia、Microsoft Azure 以及其他基础设施提供商的公司，有推动模型商品化的既得利益：如果模型可以互换，人们将购买更多 GPU，租用更多云计算容量，并开发更多应用。
特斯拉的机器人出租车正倒退行驶：https://techcrunch.com/2026/07/23/teslas-robotaxis-are-moving-in-reverse/ Sean O'Kane：https://techcrunch.com/author/sean-okane/
Jack Dorsey 正在通过 Buzz 挑战 Slack，这是一个为团队及其 AI 代理提供的群聊平台：https://techcrunch.com/2026/07/21/jack-dorsey-is-taking-on-slack-with-buzz-a-group-chat-platform-for-teams-and-their-ai-agents/ Amanda Silberling：https://techcrunch.com/author/amanda-silberling/
AI 音乐生成器 Suno 发生泄露事件，影响 5500 万用户，据 Have I Been Pwned 报道：https://techcrunch.com/2026/07/21/ai-music-generator-suno-breach-affects-55m-users-per-have-i-been-pwned/ Zack Whittaker：https://techcrunch.com/author/zack-whittaker/
Several AI companies, including Hugging Face, Meta, Microsoft, Mistral, and Nvidia, have signed an open letter：https://www.microsoft.com/en-us/corporate-responsibility/topics/open-weight/ urging policymakers not to impose broad “premature restrictions” on open-weight AI models. The letter comes as Washington debates how the U.S. should respond to allegations that Chinese AI labs are stealing intellectual property from their American counterparts, and growing in capability.
The letter doesn’t mention China at all, but it comes in the wake of reports that the Trump administration has been considering banning Chinese open-weight models, and potentially issuing sanctions against AI：https://techcrunch.com/2026/07/22/treasury-threatens-sanctions-after-white-house-claims-moonshot-distilled-anthropics-fable/ companies from the country. The White House has even accused Moonshot AI of distilling Anthropic’s Fable model to train its recently released and, by all measures, very impressive, Kimi K3 model.：https://techcrunch.com/2026/07/18/kimi-threat-or-menace/
The missive appears to be aimed at discouraging a total ban on Chinese models, as well as ensuring the administration’s response to alleged Chinese distillation doesn’t spill over into broader restrictions on open-weight AI or common techniques like distillation:
Policymakers should be careful not to conflate legitimate model-development techniques with misappropriation. Distillation, or the practice of using one model’s outputs to help train or improve another, is a widely used technique for model improvement, evaluation, and validation. It reflects a long tradition of learning from, building upon, and improving existing technologies, a tradition that has helped drive innovation since the rise of the open-source software movement.
By contrast, unlawful efforts to extract value from closed models raise legitimate concerns. Those concerns should be addressed through targeted legal and commercial frameworks rather than sweeping restrictions on techniques that play an important role in AI innovation.
Or as Amjad Masad, CEO of Replit (which also signed the letter), told TechCrunch: “I think banning Chinese open models is as good as banning open models in general.” He pointed out that Thinking Machines Lab’s new open model, Inkling：https://techcrunch.com/2026/07/15/thinking-machines-amps-up-its-bet-against-one-size-fits-all-ai-with-its-first-open-model-inkling/, was trained with the help of Moonshot’s Kimi 2.5. “It’s an ecosystem, and the precedent [a ban would] set is bad.”
The letter also pushes back on arguments in the industry that open-weight models are inherently dangerous：https://techcrunch.com/2026/07/22/arcee-a-us-open-source-ai-lab-says-chinese-models-are-not-inherently-dangerous/ because they expand access to powerful models, which can be used in cyberattacks or other nefarious activities, without any oversight.
“The right response to this risk is not to prohibit open weights. In a world where cybersecurity attackers use advanced AI, defenders need access to models with comparable capabilities so they can detect, simulate, and respond to emerging threats,” the letter reads. “Open models broaden defensive capability, increase transparency, and allow vulnerabilities to be discovered and remediated across many teams.”
Last week, OpenAI disclosed that while testing GPT-5.6 Sol and another unnamed model, one of the systems exploited a weakness：https://techcrunch.com/2026/07/22/how-an-openais-human-mistake-led-to-the-ai-powered-hack-on-hugging-face/ in its testing environment to access a Hugging Face repository：https://techcrunch.com/2026/07/21/openai-says-hugging-face-was-breached-by-its-pre-release-models/ containing a solution to a coding benchmark. One could argue the model’s goal wasn’t malicious, and that it was effectively cheating on a test to get the highest score. But the incident sparked debate about the risks of concentrating advanced AI technology behind a handful of closed providers.
Hugging Face said：https://x.com/ClementDelangue/status/2079253659108409587 it was not able to defend itself against the attack with commercial frontier AI models because their guardrails blocked its efforts. The closed AI models it used were unable to distinguish between being asked to build exploits for an attacker and a defender trying to detect them. The company instead had to pivot to using Chinese AI firm Z.ai’s GLM 5.2, a powerful open-weight model, to defend itself against the attack.
The letter highlights a divide in the AI industry. Companies like OpenAI：https://techcrunch.com/2026/07/20/openai-is-scared-of-open-weight-models-should-the-us-be/ and Anthropic have urged the administration to respond to alleged IP theft by Chinese AI firms as open-weight models grow rapidly in capability. The outcome could have major implications on their business models, which is being threatened by the spread of cheap, highly capable, and accessible AI models.
These companies, alongside other closed source AI developers like Google DeepMind and SpaceX, are notable in their absence at the bottom of this letter.
Those who signed the letter have an obvious economic stake in seeing open AI models flourish. Companies like Nvidia, Microsoft Azure, and other infrastructure providers have a vested interest in pushing for commoditized models: If models are interchangeable, people will buy more GPUs, rent more cloud capacity, and build more applications.
The letter encourages policymakers to expand access to compute for startups and researchers; invest in shared training assets like datasets, tools, and evaluation frameworks; and “[keep] the frontier plural by avoiding premature restrictions on open models that stifle competition or drive innovation overseas.”
This article has been updated with comment from Amjad Masad, CEO of Replit.
When you purchase through links in our articles, we may earn a small commission：https://techcrunch.com/techcrunch-affiliate-monetization-standards/. This doesn’t affect our editorial independence.
Rebecca Bellan is a senior reporter at TechCrunch where she covers the business, policy, and emerging trends shaping artificial intelligence. Her work has also appeared in Forbes, Bloomberg, The Atlantic, The Daily Beast, and other publications.
You can contact or verify outreach from Rebecca by emailing rebecca.bellan@techcrunch.com：mailto:rebecca.bellan@techcrunch.com or via encrypted message at rebeccabellan.491 on Signal.
Scale faster. Grow your portfolio. Gain practical expertise. No matter your goal, Disrupt can empower you. Save up to $330 toda y!
SpaceX launches new V3 Starlink satellites but suffers another booster failure：https://techcrunch.com/2026/07/24/spacex-launches-new-v3-starlink-satellites-but-suffers-another-booster-failure/ Sean O'Kane：https://techcrunch.com/author/sean-okane/
US accuses American of allegedly wiping his phone using a ‘duress’ password during border search：https://techcrunch.com/2026/07/24/us-accuses-american-of-allegedly-wiping-his-phone-using-a-duress-password-during-border-search/ Zack Whittaker：https://techcrunch.com/author/zack-whittaker/
Anduril reportedly in talks to raise funding at $100B valuation, more than 3x last year’s mark：https://techcrunch.com/2026/07/24/anduril-reportedly-in-talks-to-raise-funding-at-100b-valuation-more-than-3x-last-years-mark/ Ram Iyer：https://techcrunch.com/author/ram-iyer/
Tesla’s robotaxis are moving in reverse：https://techcrunch.com/2026/07/23/teslas-robotaxis-are-moving-in-reverse/ Sean O'Kane：https://techcrunch.com/author/sean-okane/
Jack Dorsey is taking on Slack with Buzz, a group chat platform for teams and their AI agents：https://techcrunch.com/2026/07/21/jack-dorsey-is-taking-on-slack-with-buzz-a-group-chat-platform-for-teams-and-their-ai-agents/ Amanda Silberling：https://techcrunch.com/author/amanda-silberling/
Light made a flip phone — it’s colorful and it’s cheap：https://techcrunch.com/2026/07/21/light-is-modernizing-the-flip-phone-with-light-flip/ Amanda Silberling：https://techcrunch.com/author/amanda-silberling/
AI music generator Suno breach affects 55M users, per Have I Been Pwned：https://techcrunch.com/2026/07/21/ai-music-generator-suno-breach-affects-55m-users-per-have-i-been-pwned/ Zack Whittaker：https://techcrunch.com/author/zack-whittaker/
扫码直达文章详情