以下是摘录的核心采访内容，Greg Brockman 这期播客最打动我的点是：AI 不是思考的替代品，而是判断力的放大器。越是 AI 变强，人越不能把方向交出去。

OpenAI’s Greg Brockman: Why Human Attention Is the New Bottleneck

来源：Training Data 

参与嘉宾：Alfred Lin, Greg Brockman 

1. 核心要点（Key Takeaways）

1. 计算业务比看起来更简单

The Compute Business Is Simpler Than It Looks

1. 缩放定律是深层物理真理，而非临时趋势

Scaling Laws Are a Deep Physical Truth, Not a Temporary Trend

1. 人类注意力是新的稀缺资源

Human Attention Is the New Scarce Resource

2. 摘要与对话（Summary & Transcript）

1. 核心主题（Key Themes）

计算业务比看起来更简单——且可无限扩展

OpenAI的核心计算商业模式非常直接：购买/搭建算力，以一定利润率出租/出售。需求端几乎没有上限，因为智能本身就是一种无限的资源。Greg Brockman直白地表示：

原文：”We buy, rent, build, compute, and we resell it at a margin. That is it. As long as the margin is positive, then you want to scale it. Because the demand for solving problems, the demand for intelligence, that’s unlimited.”

这是一种深刻的重构视角——OpenAI并非传统意义上的软件公司或AI实验室，而是一家智能公用事业公司。其限制因素从未是需求，而只有供给。

缩放定律是深层物理真理，而非临时趋势

Brockman并未将缩放视为一种可能会停滞的工程技巧，而是将其框架化为一种更接近自然法则的事物——经验性的、持久的，且尚未被完全理论化。

原文：”They feel deeply fundamental. It’s like the scientific truth that just like you think about physics and Newton’s laws and things like that, there’s somehow this truth of the universe. As you pour more compute into the models, they get correspondingly more capable. And it just keeps going, there’s no wall.”

这对投资具有重大启示：计算基础设施的建设并非由炒作驱动——而是在响应一条真实、持久的自然法则。

人类注意力是新的稀缺资源

随着AI处理执行层工作，关键瓶颈完全转移到了人类的判断与注意力上。这或许是对话中最具战略意义的洞见。

原文：”Human attention is going to be this incredibly scarce resource, right? The doing of things now is easy. Is this a good thing? Is this what I wanted? Is this aligned with my values, with my desires? That is going to become the most important bottleneck. And so I think building systems that take that into account and really think about the human factor, like, that’s the most important thing to do now.”

这重构了AI堆栈中价值的归属：价值不属于自动化工具，而属于那些能够智能地路由、分流并将决策升级给人类的系统。

2. 反主流观点（Contrarian Perspectives）

我们距离AGI大约80%——而非数十年之后

与主流观点认为AGI是遥远、抽象的里程碑不同，Brockman提出了一个惊人具体的主张：

原文：”According to my view of where we are, I think we’re about 80% of the way there, in that we have models that are smart. They’re very capable. I think that they are just so capable. It’s really remarkable. Like, does anyone here feel better at writing software than GPT-4?”

剩余的20%大概涉及自主代理、持续目标追求和现实世界的复杂性——但这种框架表明我们正处于最后阶段，而非中途。

公司的组织架构将从根本上瓦解——而非仅仅进化

Brockman并非主张渐进式的团队重组；他认为整个管理层级结构可能会被淘汰。

原文：”A lot of how we run organizations right now, there’s almost only one way to organize large groups of people where you have teams, you have management structures, and you have scopes, and you have these hierarchies and all these things. Maybe that can change. Maybe you can be much more flat, small teams that can really do incredible things.”

其隐含的反主流观点是：投资或构建基于传统层级组织结构的企业软件，可能是在建立一个正在萎缩的基础上。

个体创业者将能够打造过去需要整个公司才能实现的业务

这与规模需要大量员工的传统智慧背道而驰：

原文：”I think that we’re going to have this ability for solopreneurs to build very incredible businesses. And so anyone who has a vision, I think will be able to realize it.”

这对风险投资组合构建具有重大影响——下一个十亿美元公司可能只有5人团队，而非500人。

上下文缺失——而非模型能力——才是当前真正的瓶颈

尽管大多数辩论聚焦于模型智能，但Brockman指出了一个更简单、更直接的问题：AI被要求提供帮助，却未被纳入定义工作的会议、决策和背景中。

原文：”You have all these meetings, you don’t include the AI. You know, that’s not very nice to the AI. Like, you’re asking it to help you with things and it has no information. So I think really leaning into how you make sure the AI even has enough information in theory to solve the problem.”

这表明短期竞争优势不在于模型质量——而在于上下文架构。

3. 提及的公司（Companies Identified）

OpenAI

AI研究与部署公司。作为核心主体被重点介绍，处于模型能力的前沿。在内部跨职能（工程、财务、销售、IT）部署Codex，并构建Chronicle（一个持久记忆工具）。Brockman指出他们正“活在未来”，并根据自身使用场景共同设计工具：

原文：”One of the amazing things about being at OpenAI is you do get to live in the future, right? You do get to really see the shape of what’s emerging and we can design, right? We can really change the models, the harness, everything together in order to better serve the needs that we see.”

Stripe

全球支付基础设施公司。被提及为基础设施价值复利的证明点——Brockman是第4号员工和首任CTO：

原文：”Stripe as employee number four and then the first CTO. I just recently heard that they process 1.0% of the global GDP.”

4. 提及的人物（People Identified）

Greg Brockman

OpenAI联合创始人兼总裁，曾任Stripe首任CTO。被认定为同时在基础设施、模型和应用层思考的精英建设者与运营者。他将人类注意力定义为稀缺资源的框架，以及在内部亲手使用Codex的做法，展现了真正的技术深度与战略清晰度：

原文：”I’m not sure if there’s ever an official title, but I’ve been called many things. Let’s just say that.”

Matt Garman

AWS CEO。在对话中被顺带提及，但关于算力稀缺性的观点令人印象深刻：

原文：”I was just with Matt Garman. He says the GPU compute availability in 2026 rounds to zero.”

5. 运营洞见（Operating Insights）

将人类问责作为智能体工作流中不可妥协的检查点

OpenAI对Codex的内部政策——要求人工审核并签署每一次代码合并——是任何公司都可以立即采用的实用治理模板：

原文：”We still want a human to be accountable for all code that gets merged, right? So at the end of the day, is it a good thing to merge this piece of code? Is it well structured? Is it going to make our code base more maintainable? We want to make sure there’s a human who is signing off to say ‘yes.'”

与领域专家垂直部署AI，而非与通才水平部署

OpenAI的内部推广策略——将小型专业团队深度嵌入每个职能部门（财务、销售、IT），先理解领域需求再进行构建——是一种避免半生不熟、无人使用的部署的企业AI采用模型：

原文：”We are also going vertical within OpenAI to adopt these tools within finance, within sales, within IT. And there we have a small dedicated team who’s really deep understanding the domain, working with the people who are the experts in it, in order to build skills, in order to modify the tools AI, whatever it is in needed in order to get it to be good.”

从第一天起就将数据溯源构建到AI架构中

随着内部知识库可被AI查询，权限系统会失效——派生的工件可能暴露源文档后来被限制的数据。这是一个必须主动解决的架构问题，而非事后补救：

原文：”You need to make sure you have some way of tracking through the system to say, well, this output document came from the source one. The source one is no longer accessible to this audience. Let’s go in and validate that as well.”

6. 被忽视的洞见（Overlooked Insights）

OpenAI正悄悄为AI优先企业建立客户共同设计计划

在Codex讨论的末尾，Brockman透露OpenAI正主动挑选企业客户共同构建和定义这些智能体工作流——并且这是一个公开邀请。这并非公开宣传的项目，但它代表了早期访问OpenAI最强大内部工具的机会：

原文：”We are starting to work with certain customers as well. So for people who want to be very AI forward and want to be part of defining this revolution, that there’s a place for that. And I’d love to talk afterwards.”

对于在场的运营商和投资者而言，这是一个不明显的竞争护城河：与OpenAI共同设计，让你既获得优先访问权，又能塑造竞争对手未来将使用的工具。

AI驱动的科学发现比市场预期的更近——量子引力相关结果已经存在

Brockman简要提到了一个物理结果，其重要性使得严肃的研究人员曾认为该问题无法解决——而OpenAI的AI得出了物理学家认为是迈向量子引力的一步的公式。这一点只是顺带提及，几乎没有后续讨论：

原文：”We had this physics result where our AI came up with this very beautiful formula that physicists who have been working on this for quite some time thought was totally impossible. That thought it was like maybe an unsolvable problem… It’s real serious physicists who really view this as a step toward really being able to get to some sort of answer for quantum gravity.”

如果这一轨迹持续下去，那些在未来12-18个月内将AI嵌入科学发现工作流的公司和研究机构——特别是在物理学、材料科学和药物发现领域——可能会在更广泛的市场意识到正在发生的事情之前，捕获不成比例的价值。