🔶 反对开源 AI 的论点站不住脚
英文标题：🔶 The arguments against open source AI are bad
Kimi K3 的发布再次引发围绕开源 AI 的争论，一些媒体、企业领袖与政治人物将开放权重模型描述为安全和竞争威胁。作者认为，开源软件本就是商业软件与前沿模型的基础，而压制开源技术在历史上既难以奏效，也可能削弱本国企业的竞争力。文章指出，开源 AI 并非中国独有现象，芯片厂商、美国创业公司、企业用户和大型科技公司均有推动其发展的商业动机。对开发者和行业而言，开放模型将带来更低成本、更强定制能力与更广泛的安全审查，而试图阻止其扩散只会制造短期摩擦。
239.5 KB
80.4 KB
77.7 KB
中文
Kimi K3 的发布引发了新一轮焦虑和混乱讨论。有一大群记者、商业领袖和政治人物声称，开源 AI 是一种危险威胁。OpenAI 的 Dean Ball 表示：
在一个由开放权重模型主导的世界里，一种可能的结果是全面的 AI 共产主义……AI 不再是一种市场产品，而是一种“公共产品”。
人人都能自由获得 AI？太可怕了！
前沿实验室反对开源 AI 的核心论点基本是：开源模型很危险（而且不美国！）。我们应该打开 AI 的潘多拉魔盒，但只能由负责任的守门人来打开（最好是收费站经营者，也就是我们！）。只有可信用户（我们最赚钱的客户）才应该能够使用它。
我想回应一些反对开源 AI 的糟糕论点，但首先需要纠正这些论点的构建方式：
开源软件是商业软件的基础
Ball 的表述忽略了一个事实：_开源软件是所有专有软件的基础。_这也包括前沿模型，它们归根结底都是软件产品。
对于软件行业之外的人来说，开源软件可能有些反直觉：为什么要辛苦开发一个产品，然后免费送出去？
软件程序是由多个程序组成的技术栈，每一层都构建在另一层之上。要打造 Uber，你需要编程语言框架、收发网络流量的软件、数据分析工具以及无数其他组件。其中大多数并不是商业企业的差异化因素，因此，商业参与者在技术栈的底层组件上合作、在真正能区分产品的高层部分竞争，符合它们的利益。
前沿实验室会_非常希望_ AI 模型不要落入“太过普及、以至于不值得在其上竞争”的类别。是否会如此，还有待观察。
开源软件极难压制
实际上，关于压制开源模型的争论大多偏离了重点。历史告诉我们，压制开源软件_极其_困难，而这样做只会削弱企业面对国际竞争者时的实力。一段简短的加密技术历史很能说明问题：
如今，PGP 是人人都可使用的常见工具，大多数开发者至少也有所了解。但在 Phil Zimmermann 于 1991 年发明它时，美国政府将加密技术视为军事技术。政府还对 Zimmermann 展开了刑事调查。
当 Netscape 创建 SSL 时，美国政府只允许其在国际市场发布弱化版本。这些管制适得其反：弱化的“国际版”更容易获得，因此甚至许多美国人也在使用它。
出口管制未能如政府所愿限制加密技术。SSL、PGP 和类似工具在全球范围内都很容易获得，而这些管制反而让美国人处于不利地位。最终，法院裁定发布加密源代码属于受保护的言论，美国政府放宽了加密技术出口管制。
即使将压制范围缩小到“中国”模型，也不会让事情更容易。究竟什么使一个 AI 模型成为中国模型？如果像前沿模型所声称的那样，它是从美国模型蒸馏而来的，它算中国模型吗？如果美国人对中国模型进行微调呢？至多，这类 AI 监管会在一段时间内让美国人背负繁文缛节，并使其获得 AI 的能力弱于世界其他地区。
开源 AI 不只是中国现象
开源 AI 争论中隐含着一种假设：只有中国政府才有动力开发开源模型。实际上，许多商业参与者都有充足动力开发开源 AI：
- 芯片制造商：Nvidia CEO Jensen Huang 将 Nvidia 正在建设的体系描述为“token 工厂”。Nvidia 并不在意其芯片是用于运行前沿模型还是廉价开源模型——它只希望生产并创造尽可能多的 token 需求。实际上，Nvidia 自身也发布了一套开源模型。
- 美国创业公司：Thinking Machines Lab 最近发布了一款强大的开源模型。它们及其他公司押注于模型将成为商品化产品，而可防御的护城河可以建立在补充模型或为模型提供定制服务的辅助服务之上。
- 企业 AI 用户：前沿模型客户目前在开源 AI 开发中还不算活跃，但未来会参与其中。他们会希望用更低成本的模型处理低复杂度任务，并希望对面向客户的功能拥有更细粒度的控制。
- 大型科技公司：你可以肯定，Google 和 Meta 正在密切关注 OpenAI 的新广告产品。如果前沿模型的广告产品获得市场认可，这些巨头很可能会通过将无广告开源模型商品化，来压制广告竞争。
所谓“AI 竞赛”究竟是什么？
围绕中国模型的很多焦虑，核心是“输掉 AI 竞赛”。但这场竞赛的目标到底是什么？开发出最好的模型？卖出最多的 token？还是抢先毁灭人类？
谈论“AI 竞赛”并不比谈论“互联网竞赛”更有意义。我们并不是在竞争谁能第一个把火箭送上月球，而是在应对一项具有变革性的新技术。若国家之间确实存在竞赛，那也是一场吸收这场转型并推动经济增长的竞赛。按照这一框架，自由可用的 AI 模型是助益，而非威胁。
害怕中国 AI 模型的糟糕论点
中国在“倾销 AI！”
Scott Galloway 曾主张，免费的中国 AI 是一种长期消灭竞争者的尝试：
这就是中国对太阳能电池板、钢铁、电动车和电池所做的事。先是达到西方的质量水平，或者甚至还没达到。89%。很接近。实际上，汽车已经达到同等水平了，但继续说。然后他们把价格降三分之二，随后他们就拥有了整个市场。
但除了芯片之外，AI 并不是实体商品。太阳能电池板和钢铁需要实体供应链，其中每一环都难以独立存在。如果你的国家没有人在制造太阳能电池板，那么建立一家销售太阳能级硅晶圆的企业就很困难。
软件不是这样。来自中国的开源模型不会阻止一家美国微调企业取得成功——恰恰相反！
它们会传播宣传！
假设中国模型会带有亲中国的立场，并非毫无道理。但这不是压制它们的理由。_这些模型是开源的！_如果任何美国人不满中国 AI 模型的政治倾向，他们完全可以修改并发布一个“美国化”的版本。至少在美国境内，很难预见一个被视为带有扭曲亲中偏见的模型，能够胜过一个实质上相似、但带有扭曲亲美偏见的模型。
它们会加入后门！
AI 并未改变漏洞的基本市场规律：负责任的参与者修补漏洞，攻击者利用漏洞。限制负责任参与者可用的工具，只会帮助攻击者。
从理论上说，不良行为者可能在模型中嵌入隐藏的对抗性行为。但如果发生这种情况，负责任参与者的利益在于尽快发现这些漏洞；而做到这一点的最佳方式，就是让所有想检查它们的人都可以进行检查。
开源 AI 正在到来
政策制定者或商业领袖想要什么，其实并不太重要：开源 AI 过于强大，也过于难以控制。它正在到来，试图压制它的行为，最终不过是这一进程中的些许噪音。
English / 原文
The release of Kimi K3 has opened a fresh round of angst and confused discourse. There's a loud cohort of journalists, business leaders, and politicians arguing that open source AI is a dangerous threat. OpenAI's Dean Ball:
One probable outcome of an open-weight-model-dominant world is full AI communism... rather than a market product, AI is a "public good"
Freely available AI for anyone? The horror!
Frontier labs' case against open source AI is essentially: Open source models are dangerous (and un-American!). We should open the AI Pandora's Box, but only with responsible gatekeepers (toll collectors, preferably us!). Only trusted users (our most profitable customers) should be able to use it.
I want to address some bad arguments against open source AI, but some corrections on how the argument is being framed are in order:
Open source software is the foundation for commercial software
Ball's framing strolls past the fact that open source software is the foundation of all proprietary software. This includes frontier models, which at the end of the day are software products.
Open source software is counterintuitive to people outside of the software industry. Why work hard on a product, and give it away for free?
A software program is a stack of programs, with each layer built on top of another. To build Uber, you need programming language frameworks, software to send and receive web traffic, data analysis tools, and countless other components. Most of these are not differentiators for a commercial enterprise, so it serves commercial actors to cooperate on lower components in the stack and compete on the higher level pieces that actually differentiate their products.
Frontier labs would very much like AI models to not fall into the category of "so commonplace that it doesn't make sense to compete on". Whether that happens remains to be seen.
Open source software is very difficult to suppress
In reality, the argument about suppressing open source models is mostly beside the point. History tells us that suppression of open source software is extremely difficult, and attempting to do so only serves to weaken companies against international competitors. A brief history of encryption is illustrative:
Today, PGP is a commonplace tool anyone can use, and most devs are at least familiar with. But when Phil Zimmermann invented it in 1991, the U.S. government considered encryption to be military technology. A criminal investigation was opened against Zimmermann.
When Netscape created SSL , the U.S. government allowed it to only release a weakened version of it internationally. These controls backfired: it was much easier to acquire the weakened, "international" version, so even many Americans used it.
Export controls did not succeed in limiting encryption as the government wished. SSL, PGP, and similar tools were readily available throughout the world, and the controls disadvantaged Americans. Eventually, courts ruled that releasing encryption source code is protected speech, and the U.S. government relaxed encryption export controls.
Narrowing suppression to "Chinese" models won't make things easier. What, exactly, makes an AI model Chinese? Is it Chinese if, as frontier models allege, it was distilled from American models? What about if an American fine-tunes a Chinese model? At best, regulating AI in this way will (temporarily) encumber Americans with red tape and diminished AI access relative to the rest of the world.
Open source AI is not just a Chinese phenomenon
There's an assumption baked into the open source AI debate that open source models are something that only the Chinese government has an incentive to develop. In reality there are many commercial actors with ample incentive to develop open source AI:
- Chip makers : Nvidia CEO Jensen Huang has described what Nvidia is building as “token factories”. Nvidia doesn't care if its chips are used to run frontier models or cheap open source models - it just wants to produce and generate demand for as many tokens as possible. And indeed Nvidia has itself released a suite of open source models.
- American Startups : Thinking Machines Lab recently released a powerful open source model. They and others are betting that models will be commoditized, and a defensible moat can be built around auxiliary services that complement or customize models.
- Enterprise AI users : Frontier model customers aren't currently all that active in open source AI development, but they will be. They will want lower-cost models for low-complexity tasks, and more fine-grained control over customer-facing features.
- BigCos : You can be sure that Google and Meta are watching OpenAI's new ad product closely. Should frontier model ad products gain traction, it would be well worth it for these behemoths to commoditize ad-free, open source models to squash ad competition.
The "AI race" is... what, exactly?
Much of the angst around China's models centers on "losing the AI race". But what's the goal of this race? Is it to develop the best model? To sell the most tokens? To destroy humanity first?
Talking about an "AI Race" doesn't make more sense than talking about an "Internet Race". We're not competing to be the first to send a rocket to the moon, we're reacting to a new, transformational technology. To the extent there's a race between nations, it's to absorb this transition and grow economies. In this framing, free AI models are a boon, not a threat.
Bad arguments to fear Chinese AI models
China is "AI dumping!"
Scott Galloway has argued that free Chinese AI is an attempt to eliminate competitors in the long run:
This is what China did to solar panels, steel, EVs, and batteries. First, they match Western quality, or they don't even match it. 89%. Close. Actually, match it with cars, they've matched it, but go ahead. Then they cut the price by two thirds, then they own the market.
But apart from chips, AI isn't a physical good. Solar panels and steel require physical supply chains, each link of which cannot easily exist on its own. If no one is manufacturing solar panels in your country, it's difficult to build a business selling solar-grade silicon wafers.
Software isn't like that. An open source model coming from China doesn't prevent a fine-tuning business from succeeding in the US - quite the opposite!
They will spread propaganda!
It's not unreasonable to assume that Chinese models will be shipped with a pro-China point of view. But this is not a reason to suppress them. The models are open source! If any American has an issue with the political slant of Chinese AI models, they are free to change and release an "Americanized" one. At least within the U.S., it's difficult to foresee a model seen as having a distorted pro-China bias outcompeting a substantially similar model with a distorted pro-U.S. bias.
They will add backdoors!
AI does not change the basic market for vulnerabilities: responsible actors patch them, attackers exploit them. Limiting tools for responsible actors only serves attackers.
It's theoretically possible for a bad actor to embed hidden adversarial behavior in a model. But if this happens, it serves the interests of responsible actors to find these exploits as soon as possible, and the best way to do this is to let anyone who wants to inspect them.
Open source AI is coming
It doesn't matter much what policy makers or business leaders want: open source AI is too powerful, and too difficult to control. It's coming, and attempts to squash it will not amount to anything more than noise along the way.