OpenAI’s Jev clone could help the frontier lab stop its swarming agents
OpenAI’s Jev clone could help the frontier lab stop its swarming agents
OpenAI 的 Jev 克隆版或将助力该前沿实验室管控其失控的智能体
One of the more intriguing announcements at OpenAI’s Dev Day event on Tuesday came in an aside from CEO Sam Altman, who revealed the company’s new “Decisions API.” The API apparently provides similar functionality to Jev, a model released by TypeSafe AI earlier this month that’s explicitly designed for software automation. 在周二的 OpenAI 开发者日活动中，最引人注目的公告之一来自 CEO 山姆·奥特曼（Sam Altman）的一段插话，他透露了公司全新的“决策 API”（Decisions API）。该 API 显然提供了与 Jev 类似的功能，而 Jev 是 TypeSafe AI 本月早些时候发布的一款专门用于软件自动化的模型。
A kind of super-powered classifier built on an LLM, developers can give Jev a set of choices that it outputs as probabilities cheaply and at high speeds. OpenAI’s Decisions API seems to be the same sort of product. At the event, Altman described the API as a way to give the lab’s Luna model a predefined set of options to choose between, such as categories in which to classify an image or different agent behaviors. 作为一种基于大语言模型（LLM）构建的超强分类器，开发者可以为 Jev 提供一组选项，它能以低成本、高速度输出这些选项的概率。OpenAI 的决策 API 似乎属于同类产品。在活动中，奥特曼将该 API 描述为一种为实验室的 Luna 模型提供预定义选项的方法，例如对图像进行分类的类别或不同的智能体行为。
“By focusing the model on that choice, we can make it extremely fast while keeping capabilities like image understanding, broad language support, and safety protections,” Altman said. TypeSafe didn’t respond to TechCrunch’s questions about the new product, but CEO Diogo Almeida, a former OpenAI engineer who co-invented reinforcement learning, joked on X about the beginning of the clone wars. “通过让模型专注于这些选择，我们可以在保持图像理解、广泛语言支持和安全保护等能力的同时，使其运行速度极快，”奥特曼说道。TypeSafe 没有回应 TechCrunch 关于该新产品的置评请求，但其 CEO、曾共同发明强化学习的前 OpenAI 工程师 Diogo Almeida 在 X 上开玩笑称“克隆战争开始了”。
He added that OpenAI’s interest could be “a sign…that building in a System One compatible way is the future.” (“System One” is TypeSafe’s term of art for fast, intuitive thinking, versus “System 2,” which it applies to deliberate reasoning.) The subtext here is that LLMs as we know them aren’t the right solution for a lot of software because they are comparatively slow and expensive. 他补充说，OpenAI 的兴趣可能是一个信号，“表明以‘系统一’（System One）兼容的方式进行构建是未来趋势。”（“系统一”是 TypeSafe 的专业术语，指快速、直觉式的思维，与之相对的是指代深思熟虑的“系统二”。）这里的潜台词是，我们所熟知的大语言模型并不是许多软件的最佳解决方案，因为它们相对缓慢且昂贵。
Developers have been using Jev to augment LLMs and, in doing so, have found that they’re faster and cheaper. It’s not clear how similar Decisions API will be to Jev, since OpenAI released it as a limited preview and, thus far, TechCrunch hasn’t spotted developers running it through its paces. However, there is clearly interest, according to the conversations on X. 开发者一直在使用 Jev 来增强大语言模型，并发现这种方式更快、更便宜。目前尚不清楚决策 API 与 Jev 的相似程度，因为 OpenAI 仅将其作为有限预览版发布，且到目前为止，TechCrunch 尚未发现有开发者对其进行全面测试。不过，从 X 上的讨论来看，市场对此显然很感兴趣。
Decisions API isn’t the only Jev-like API on the internet — other startups are rolling out similar models; OpenAI won’t be the last tech giant to produce one. A key question is how well calibrated each of these decision models’ outputs will be to real life. Almeida says his company’s moat is the synthetic data it creates to generate statistically useful outputs. 决策 API 并不是互联网上唯一的类 Jev API——其他初创公司也在推出类似模型；OpenAI 也不会是最后一家开发此类产品的科技巨头。一个关键问题是，这些决策模型的输出在现实生活中能达到多高的校准度。Almeida 表示，他公司的护城河在于其创造的合成数据，这些数据能生成具有统计学意义的有用输出。
“Fast and cheap is very easy, you know,” Almeida told TechCrunch last week. “If you want it really fast and cheap, use dice, right? Intelligence is the hard part, and my North Star is always pushing the intelligence-per-dollar Pareto curve.” “快速且便宜非常容易，你知道的，”Almeida 上周告诉 TechCrunch。“如果你想要极快且便宜，掷骰子就行了，对吧？智能才是难点，我的北极星指标始终是推动‘每美元智能价值’的帕累托曲线。”
After just weeks, it seems clear that these models have a future ahead of them, and one likely application is monitoring and securing AI agents. One of OpenAI’s new security measures following a series of incidents where its agents misbehaved on the open internet is using a separate model to watch for bad actions at “significant compute cost.” 仅仅几周后，这些模型的前景似乎已十分明朗，其中一个可能的应用场景是监控和保护 AI 智能体。在发生了一系列智能体在开放互联网上表现失控的事件后，OpenAI 的新安全措施之一是使用一个独立的模型来监测不良行为，但这需要“巨大的计算成本”。
Shapor Naghibzadeh, a long-time cybersecurity professional who leads the startup QueryStory, thinks that a model like Jev could make that possible far more cheaply. He built a demo for a hackathon held last weekend that uses Jev to check each agentic action against the task it was given, blocking actions it had high confidence were bad, flagging others for review, and permitting the rest. 长期从事网络安全工作的 QueryStory 初创公司负责人 Shapor Naghibzadeh 认为，像 Jev 这样的模型可以以低得多的成本实现这一目标。他在上周末举行的黑客马拉松中构建了一个演示程序，利用 Jev 对照智能体被分配的任务来检查其每一个动作，拦截那些高置信度判定为不良的动作，将其他动作标记为待审查，并允许其余动作执行。
In theory, such monitoring could have stopped the Hugging Face incident — and monitoring of that kind costs $2.94 with Jev, versus $372 with a frontier LLM. A key observation is that Jev is arguably cheap enough to run on every agentic action, which offers a layer of review that could improve the reliability of agents writ large. It’s the kind of thing TypeSafe was hoping to achieve — and now OpenAI has seen the value as well. 理论上，这种监控本可以阻止 Hugging Face 事件——使用 Jev 进行此类监控的成本仅为 2.94 美元，而使用前沿大语言模型则需要 372 美元。一个关键的观察是，Jev 的成本低到足以在每一个智能体动作上运行，这提供了一层审查机制，可以全面提升智能体的可靠性。这正是 TypeSafe 希望实现的目标，而现在 OpenAI 也看到了其中的价值。