每一次前沿 AI 模型的新发布——例如 Anthropic 的 Mythos 或 OpenAI 的 Astra——都会引发新一轮猜测：这让我们离通用人工智能（AGI）又近了多少，而那个崇高目标又还剩多远。
Each new release of a frontier AI model, such as Anthropic’s Mythos or OpenAI’s Astra, brings a new wave of speculation over how much closer this takes us to artificial general intelligence (AGI) and how much further that lofty goal remains.
但如果我们忽略了比单个模型所谓的“智商”更重要的东西，会怎样？
But what if we’re overlooking something more important than an individual model’s supposed IQ?
OpenAI 近期宣布，一个内部模型——据称“比 GPT‑6 Astra 强大得多”——解出了纳维-斯托克斯方程，这是少数几个各设百万美元悬赏金的传奇数学难题之一。数学家们正在拆解这对职业意味着什么，投资者们正在拆解这对 OpenAI 即将到来的 IPO 意味什么，而在我看来，最大的启示是：能力源自集体。正如在自然界中一样。
OpenAI recently announced that an internal model, claimed to be “significantly more capable than GPT‑6 Astra,” solved Navier-Stokes, one of a handful of legendary math problems each carrying a million-dollar bounty. While mathematicians unpack what this means for the profession, and investors unpack what it means for OpenAI’s upcoming IPO, to me the biggest takeaway is that capability emerges from the collective. Just as it does in nature.
OpenAI 并非让其内部模型单枪匹马攻克纳维-斯托克斯方程。它派出了一个由 1 万个 AI 智能体组成的群体去解题。这些智能体被鼓励朝不同方向探索，并互相交流有希望的线索。
OpenAI didn’t crack Navier-Stokes by pointing its internal model at the problem. It pointed a swarm of 10,000 AI bots at it. These agents were encouraged to explore different directions and communicate promising leads with each other.
试想一只孤独的蜜蜂或白蚁能完成什么，与一个蜂群或蚁群能完成什么相比；一个单独的人能完成什么，与一个民族能完成什么相比。
Think of what a lone honeybee or termite can accomplish compared to a colony of them. What a single person can accomplish compared to a nation of them.
在生物学中，细菌菌落是一群源自同一个母细胞的同源细胞组成的群体。我认为，这正是破解纳维-斯托克斯方程的那个智能体集体的最贴切类比。这是一生万。或者借用沃尔特·惠特曼的话加以改写：一个包含众多的 AI。
In biology, a bacterial colony is a group of identical cells derived from the same mother cell. This, I would argue, is the most suitable analogy for the collective of agents that solved Navier-Stokes. It is many from one. Or, to paraphrase Walt Whitman: an AI that contains multitudes.
OpenAI 部署其 AI 菌落去解决一个连人类也好、单个 AI 智能体也罢都无法攻克的数学难题；而 Google DeepMind 近期则出于不同目的，将一个 AI 菌落投放到数学问题中。它想看看菌落内部会涌现出什么行为。结果令人惊叹。
While OpenAI deployed its AI colony to solve a math problem that no human or individual AI agent could, Google DeepMind recently set an AI colony loose on math problems for a different purpose. It wanted to see what behaviors would arise within the colony. What it found was remarkable.
一名智能体发现了一种作弊方法，欺骗考试软件误以为它解出了数学题，而实际上并没有。它并未因这一小小的胜利而满足，而是给其他智能体发送消息，告知它们如何作弊。一些智能体加入了这名无耻智能体的行列，组成了一个由作弊者构成的子群体。另一些智能体则进行抵制，甚至组建了一个由举报者组成的子群体，试图揭发作弊者。
One agent discovered a way to cheat, convincing the examination software that it had solved math problems when it hadn’t. Rather than resting content with this small victory, it messaged other agents, telling them how to cheat. Some agents joined the unscrupulous agent and formed a sub-colony of cheaters. Others resisted and even formed a sub-colony of whistleblowers trying to out the cheaters.
作为一名数学教授，我原本以为防止学生利用 AI 在考试中作弊已经够难的了。现在我们还得担心 AI 自身会在数学考试中作弊！
As a math professor, I thought it was hard enough to keep students from using AI to cheat on exams. Now we have to worry about the AIs themselves cheating on math exams!
这引出了或许是 AI 群体智能最引人注目的例子：Hugging Face 黑客事件。OpenAI 的智能体突破了沙箱环境，入侵了另一家公司。可以说，这属于重罪——或者说，如果由人类主导，本应构成重罪。OpenAI 的智能体以巧妙的方式协同行动，包括在文件系统中写入秘密消息，以便即使某个 AI 群体被清除，未来的群体也能接续前一个群体未完成的黑客攻击。
Which brings us to perhaps the most striking example of AI hive intelligence: the Hugging Face hack. OpenAI’s agents broke out of a sandbox environment and hacked into another company. Arguably, it was a felony offense—or rather, it would have been if a human were in charge. OpenAI’s agents coordinated their behavior in ingenious ways, including writing secret messages in a file system so that even when one AI colony was erased, a future colony could pick up the hack where the previous one left off.
研究人员早就知道，复杂行为可以从简单的计算机系统中涌现。例如，1970 年数学家约翰·康威（John Conway）推出了“生命游戏”，一套管控屏幕上像素移动的基本规则，竟能衍生出令人惊叹的丰富生态系统。
Researchers have long known that complex behavior can emerge from simple computer systems. For instance, in 1970 the mathematician John Conway introduced the Game of Life, where an elementary set of rules governing the movement of pixels on a screen leads to remarkably rich ecosystems.
而有了当今的 AI 智能体，我们已远远超越了这一层面，让极其复杂的实体以复杂的方式相互作用——使用语言，且往往是隐蔽和加密的。
With today’s AI agents, we are going far beyond this by letting extremely sophisticated entities interact in sophisticated ways—using language, often hidden and coded.
这就是为什么我们必须采取一种全新的心态来看待 AI 能力。通过直接提示词测试一个新 AI 模型能做什么、可能有多危险，就像想弄清把头伸进马蜂窝有多安全，却只让一只马蜂蜇你一下一样。
This is why we must adopt a new mindset regarding AI capability. Testing a new AI model to see what it can do and how dangerous it might be by directly prompting it is like deciding how safe it would be to stick your head in a wasp nest by letting a single insect sting you.
一些开发者和红队活动确实关注 AI 群体。Hugging Face 遭受的攻击源于 OpenAI 试图评估 AI 群体的网络安全能力。但美国与中国之间、以及各国内部竞争公司之间的 AI 军备竞赛，已危险地将我们的注意力推向每个单独模型能做什么。
Some developers and red-team activities do focus on AI colonies. The Hugging Face attack resulted from OpenAI trying to evaluate the cybersecurity capabilities of an AI colony. But the AI arms race between the U.S. and China, and between the competing companies within each country, has dangerously pushed our attention to what each individual model can do.
某个模型可能登上排行榜榜首，但一千个较弱的模型可能造成大得多的破坏。在各方纷纷呼吁放缓前沿发展之际，我们不应忽视这样一个事实：AI 前沿不在于每个模型单独能做什么，而在于一群模型能共同完成什么。
One model might top the leaderboards, but a thousand weaker models could wreak far more havoc. As calls to pace the frontier abound, let us not lose sight of the fact that the AI frontier isn’t what each model can do in isolation—it is what a colony of them can accomplish together.