Search Ctrl+K
Change language Switch ThemeSign In
[](https://www.bestblogs.dev/en/explore/brief "Daily Brief")[](https://www.bestblogs.dev/en/explore/newsletter "Weekly Picks")[](https://www.bestblogs.dev/en/explore/topics "Topics")
[](https://www.bestblogs.dev/en/settings "Settings")[](https://www.bestblogs.dev/en/docs "Help Center")
Narrow Mode
81
微软、英伟达、OpenAI 等，联合声明
Microsoft, NVIDIA, OpenAI, etc., issued a joint statement supporting open-weight AI models, saying they benefit innovation, competition, and risk dispersion, and warning that restrictions could stifle ecosystem development. Follow·
Yesterday·681 words (about 3 min)
AI Summary & Key Points
Summary
CCTV Finance reported that multiple U.S. technology companies and institutions issued a joint statement on July 24 advocating open-weight AI models. The statement said such models allow anyone to download, inspect, modify, and run them on their own infrastructure, helping to expand innovation opportunities, strengthen competition, reduce risks, drive AI into various fields, and prevent benefits from concentrating in a few entities. In response to concerns raised by some U.S. officials about restricting open-weight models, the statement emphasized that AI leadership should depend on building a strong, open, and widely adopted ecosystem rather than owning a single frontier model, and warned that restrictions could stifle competition. The article also noted that while open models carry risks of losing control, they should be addressed through comprehensive safeguards rather than outright bans, and pointed out that closed-source models likewise pose security risks, with concentrating advanced AI capabilities in a few closed-source models amplifying those risks.
Main Points
* 1. Definition of open-weight models and their advantages in promoting innovation and competition.
The statement describes open-weight models as AI models that anyone can download, inspect, modify, and run on their own infrastructure, believing this expands innovation opportunities, strengthens market competition, and reduces risk.
* 2. Response to U.S. government restrictions on open-weight models: leadership should depend on an open ecosystem rather than a single model.
The statement notes that judging U.S. leadership in AI should not be based on owning a frontier model but on the ability to build a strong, open, and widely adopted AI ecosystem across industries, and that restrictions could stifle competition.
* 3. Risk and response: open models carry loss-of-control risks, but should be addressed with comprehensive safeguards rather than outright bans; closed-source models are not necessarily safer.
The article mentions that releasing open-weight models may lead to loss of control by the original developers, but the proper approach is to implement comprehensive safeguards against potential harms; it also notes that closed-source models can be hacked, misused, or fail, and concentrating advanced AI capabilities in a few closed-source models amplifies risk.
Sign in to highlight text and take notes as you read. Sign in now
央视财经 2026-07-26 20:06 北京
!Image 7: 图片
美国多家科技企业和机构24日发表联合声明，支持开放权重人工智能模型。这被认为是对美国政府中一些人企图限制开放权重模型的回应。
目前已有 微软、英伟达、开放人工智能研究中心（OpenAI）等多家企业和机构签署了该联合声明。
联合声明说，开放权重模型指任何人都可以下载、检查、修改并在自有基础设施上运行的人工智能模型，这类模型有助于扩大创新机会、加强竞争、降低风险，有助于推动人工智能进入生产生活各个领域，并避免人工智能发展收益集中在少数主体手中。
针对美国政府中一些人讨论限制开放权重模型，声明说，对美国在人工智能领域是否有领导地位的判断，不应取决于是否拥有某一个前沿人工智能模型，而应取决于能否建立一个强大、开放并广泛应用于各行业的人工智能生态系统，相关限制措施可能会抑制竞争。
联合声明说，开放权重模型存在风险，如在发布后会脱离原始开发者控制，但正确的应对方式不是一禁了之，而是针对可能的危害采取综合性保护措施。
另一方面，仅依赖闭源模型并不必然更加安全。闭源模型同样可能有被入侵、被滥用或出现故障的风险，将先进人工智能能力集中在少数闭源模型中会让风险更严重。开放权重模型允许更多研究人员和开发者检查模型行为、发现漏洞、开发防护措施并持续改进模型。 猜你喜欢 ▌本文来源：央视财经（ID：cctvyscj） 你会 关注 胡塞武装，发动导弹袭击 2026AI竞速：向上突破天花板，向下扎根产业链 从“打卡中国”到“体验中国” 中国入境游何以“圈粉”世界 伊朗警告乌克兰 “剧”缺人！招聘需求暴涨179%，有岗位月薪2万元...... *
监制/柯成韵
主编/武洋
编辑/孙永慧
转发，关注！
Key Quotes
> Judging whether the U.S. has leadership in the AI field should not depend on owning a single frontier AI model, but on whether it can build a strong, open, and widely adopted AI ecosystem across industries.
> Open-weight models refer to AI models that anyone can download, inspect, modify, and run on their own infrastructure.
Tags
Open-weight models
Open AI
AI Governance
Open Source AI
Technology Policy
Related Articles
* US Military Launches 'Hellfire' Missiles * AI-Native Companies and Nations * Why Are AI Companies Bringing Philosophers into the Lab? * Severe Cases Can Be Fatal! Toxic 'Miracle Health Medicine' Sold Rampantly Online! Exposed by CCTV 'Financial Investigation' → * Revaluing Everything, Renaissance - 2026H1 AI Industry Observation * China's 'Light God' Who Loves Anime Solves a 125-Year-Old Math Problem * 2026 AI Race: Breaking Through Technical Ceilings and Rooting into Industrial Chains * Si Xiao: Building a New Form of Intelligent Economy, China's Comprehensive Advantages and Key Deployments * Domestic Heavy-Duty Trucks Go Overseas in Groups! Orders Set New Records * Have We Just Celebrated Humanity's Last Labor Day? Eight Changes in AI's New Careers
Make your daily reading actually fit you.A daily brief built from the sources you follow. Get started free HomeDiscoverSettings