# OpenAI Update and Q&A with Sam Altman
- Transcript ID: `1131297184`
- Source tier: `vimeo_auto`
- Original language: `en-x-autogen`
- Translation status: `agent_translated_reviewed` (151/151 chunks)
- Original source: https://vimeo.com/1131297184
- Raw evidence: `data/transcripts/vimeo/1131297184.json`
> 中文是派生翻译，英文是留存的源文本。中文未复核时，以英文和原始音视频为准。
## 00:00:06
**中文**
大家好，我是 Sam。这位是我们的首席科学家 Jakub。今天，我们有许多关于 OpenAI 的最新情况要分享。显然，今天的新闻是我们的新组织架构，我们会在临近结束时谈到它，但还有许多其他重要背景需要说明。首先，鉴于其中许多事项的重要性，我们将以不同寻常的透明程度，介绍一些具体的研究目标、基础设施规划和产品计划。
**English / Original**
Hello, I'm Sam. This is our chief scientist, Yoko, and we have a bunch of updates to share today about OpenAI. Um, obviously the news of today is our new structure. We're gonna get to that near the end, but there's a lot of other important context we would like to share. First, given the importance of a lot of this, we're going to go into, uh, an unusual level of transparency about some of our specific research goals and infrastructure plans and product.
## 00:00:26
**中文**
不过，我们认为，在这个时候全面说明这些事情非常符合公众利益。OpenAI 的使命——无论是非营利实体还是新的公益公司（PBC）——都是确保通用人工智能惠及全人类。随着我们越来越接近实现它，我们对这意味着什么也有了新的认识。
**English / Original**
But we think it's, uh, you know, sort of very much in the public interest at this point to cover all of this. So, our mission at OpenAI in both the nonprofit and our new PBC is to ensure that artificial general intelligence benefits all of humanity. As we get closer to building this, we have new insights into what that is going to mean.
## 00:00:45
**中文**
OpenAI 早期曾经认为，AI 或 AGI 会像高悬空中的神谕，为我们创造各种美好的东西。如今，我们的看法更加清晰：我们希望创造工具，再由人们使用这些工具创造未来。我们希望尽可能用 AI 赋能每个人，并相信那种贯穿人类历史的进程会延续下去——人们利用更新、更好的工具，不断创造更好的事物。
**English / Original**
There was a time earlier on in open AI where we thought that AI would, or a GI would be sort of this oracular thing in the sky, and it would make all these wonderful things for us. And we now have a sharper view of that, which is we wanna create tools and then we want people to use them to create the future. We wanna empower people with AI as much as possible, and then trust that the process that has been working for human history of people building, um, better and better things with newer and better tools will continue to go on.
## 00:01:12
**中文**
现在，我们已经能看到这样一种愿景：帮助打造个人 AGI，使人们可以在任何地方使用它，调用各种工具、服务和系统，协助工作与个人生活。随着 AI 不断进步，它甚至可以发现或帮助发现新的科学知识。
**English / Original**
We, we can now see a vision where we help build a personal a GI that people can use anywhere with all of these different tools, access to all these different services and systems to help with work and personal life and their personal life. And as AI gets better and better is AI can even do things like discover or help discover new science.
## 00:01:30
**中文**
人们借此能够创造出的东西，将令整个社会变得更好，也使个人生活更加充实；我们认为，这将非常惊人。OpenAI 有三大核心支柱：研究、产品和基础设施。我们必须在构建 AGI 所需的研究上取得成功；必须打造一个易用而强大的平台；还必须建设足够多的基础设施，让人们能够以较低成本使用他们想要的这些卓越 AI。
**English / Original**
What people will be able to create with that, um, to make all of society better in their own lives, more fulfilled, we think should be quite incredible. There are three core pillars we think about for OpenAI research, product and infrastructure. We have to succeed at the research required to build a GI. We have to build a platform that makes it easy and powerful to use, and we have to build enough infrastructure such that people can use at a low cost, all of this amazing AI that they'd like.
## 00:01:57
**中文**
这张简图展示了我们如何看待这个世界。最底层是芯片、机架及其配套系统，以及容纳它们的数据中心和能源。今天会多谈前三项，能源留待以后。我们在这些基础设施之上训练模型，再在模型之上构建 OpenAI 账户。
**English / Original**
Here's a little like cartoon of how we think about our world. So at the bottom layer here, we have chips racks and the systems around them, the data centers that these go into, and the energy. Uh, we'll talk more about the first three today and energy another time. Then we train models on top of these. Then we have an open air account on top of that.
## 00:02:14
**中文**
我们现在有一款名为 Atlas 的浏览器；未来几年还会推出设备，让你随时随地把 AI 带在身边。我们有 ChatGPT、Sora 等少数第一方应用，今后还会增加。但最让我们兴奋的，是右上角那块大的拼图。我们终于开始进入这样一个世界：人们可以利用 AI 打造卓越的服务——从 API、ChatGPT 中的应用，到未来的新企业平台、OpenAI 账户，以及更多更多东西。
**English / Original**
We have a browser now, uh, called Atlas. And we have devices coming, um, in the next few years that you'll be able to take AI with you everywhere. And we have a few first party apps like CHATT and Soar, and we'll have more over time. But mostly what we're excited about is this big puzzle piece in the upper right. We're finally getting to a world where we can see that people are going to be able to build incredible services, uh, with ai, starting with our API with apps and chat, bt new enterprise platform that we'll have over time, uh, an open AI account in way, way more.
## 00:02:44
**中文**
人们将能够把当今世界已有的各类事物，以及更多新事物，接入这个新的 AI 世界。我们希望促成这一点，并相信全世界会由此为所有人创造巨大的价值。这大致就是我们看到的经济图景。不过，还有一件我们思考已久、如今确实开始显现端倪的事情——无论你称之为曙光还是新芽——那就是 AI 对科学的影响。
**English / Original**
And people will be able to fit all of the kind of like current things in the world and many more into this new AI world. And we wanna enable that. And we, we believe the world will build just a huge amount of value for, um, for all of us. So that's kind of what we see the economic picture looking like. But one of the things that we've thought about for a long time, and we really see happening now or starting to happen now, glimmers of it, green shoots, whatever you wanna call it, is the, uh, impact that AI will have on science.
## 00:03:10
**中文**
上一页所说的经济影响，会对长期生活质量、社会改善和社会变迁产生巨大作用；但能够自主发现新科学，或帮助人们更快取得科学发现的 AI，我认为将是最重要的事情之一，也是我们正在努力理解其意义的事情。
**English / Original**
And, um, although the economic impact from that previous slide will be huge for the long term quality of life, uh, and improvement that, and change in society, AI that can autonomously discover new science or help people discover new science, uh, faster will be, I think one of the most important things and something that we're really trying to wrap our heads around.
## 00:03:31
**中文**
下面我把话筒交给 Jakub，请他介绍研究。正如刚才提到的，我们会大量分享内部目标，以及我们对当前进展的判断。谢谢 Sam。我们的核心是一家研究实验室，专注于理解一种叫作深度学习的技术。其中一个重点，是理解深度学习系统的训练规模扩大后会发生什么。
**English / Original**
So I'm gonna hand this over to Yako to talk about research. And as I mentioned, uh, we're gonna share a lot about our internal goals and our picture of where thing, where things are. Thanks, Sam. Um, at the core, we are a research laboratory focused on understanding a technology called deep learning. And so a particular focus of ours is, um, understanding, uh, what happens as you scale up training deep learning systems.
## 00:03:58
**中文**
我们经常讨论由此带来的一个结果——AGI，也就是通用人工智能。不过我们发现，从某种意义上说，AGI 这个说法甚至可能低估了潜在进步与变革的幅度。具体而言，我们认为，深度学习系统有可能在不到十年内迈向超级智能。
**English / Original**
Um, and one consequence we discuss a lot there, uh, is a GI artificial general intelligence, but we find that, um, in some way even this maybe understates a bit, um, the magnitude of, um, the possible progress and change here. Um, and so in particular, we believe, uh, that it is possible that deep learning systems are less than a decade away from super intelligence.
## 00:04:28
**中文**
也就是说，这类系统会在许多关键领域比我们所有人都聪明。当然，这是一件严肃的事，我们必须应对其众多影响。我们整个研究计划所围绕的一项重点，是这项技术以及通往它的技术有潜力加速科学发现、加快新技术开发。
**English / Original**
So systems that are smarter, uh, than all of us on, uh, a large number of, of, of, of critical AEs. And, um, this is of course, uh, a serious thing, right? There is, uh, a lot of implications of this to grapple with. And one, uh, particular focusing impact of this technology. Um, and the technology is leading up to it. And something that we organize our entire research program around is the potential, uh, to accelerate scientific discovery, um, to accelerate the development of new technology.
## 00:05:12
**中文**
我们认为，这或许会成为 AI 发展最重大的长期影响，并从根本上改变新技术进步的速度。若要判断我们距离这些目标还有多远，一个很好的方法，是观察模型能够完成的任务，通常需要人类花多长时间完成。
**English / Original**
Um, we believe that this will be, uh, perhaps the most significant, uh, long-term impact of AI development. Uh, and it'll fundamentally change the pace of progress on, on, on developing new technologies. Um, and so thinking about how far along we are, uh, uh, towards these goals, uh, one good way to think about progress is to look at the time horizon that it'll take people to accomplish the task that the models can perform.
## 00:05:46
**中文**
过去几年，这一时间跨度一直在迅速延长。当前一代模型大约达到了五小时。你可以从模型在国际数学或信息学竞赛等比赛中比肩最优秀选手的表现看出这一点。
**English / Original**
And so this is something that has been extending rapidly over the past few years. Um, so where the current generation of models is at right now is about five hours. So, um, if you, you can see this by looking at the models matching the best people in competition such as the international impact and mathematics or informatics.
## 00:06:07
**中文**
我们相信，这个时间跨度会继续快速延长。一部分原因是算法创新和深度学习的进一步扩展，尤其是沿着一个新维度扩展——上下文计算，也称测试时计算。在这个维度上，我们仍有多个数量级的提升空间。
**English / Original**
And we believe that this horizon will continue to extend rapidly. And this is in part as a result of algorithmic innovation and the, uh, just scaling deep learning further, and in particular, scaling, um, along this, um, new axis, um, in context compute, also called test time compute, uh, where we really see orders and orders of magnitude to go.
## 00:06:29
**中文**
这大致就是模型花多少时间“思考”。看看模型现在为问题投入的思考时间，再想想对于科学突破等真正重要的问题，我们愿意投入多少算力、多少时间——即便动用整座数据中心，也应当是可以接受的。
**English / Original**
Um, so this is roughly how much time the model spends thinking, right? And if you look at how much time the model's currently spent thinking about problems, and if you think about how much compute, how much time you would like to spend on problems that really matter, such as scientific breakthroughs, you should be okay using entire data centers.
## 00:06:47
**中文**
所以这里确实还有很长的路可走。预见到这种进步，我们当然会在内部据此制定计划，也希望提高外界对我们思路的了解。因此，我们想采取一个或许不太寻常的做法：公开我们迈向这些强大系统的内部目标和时间表。
**English / Original**
Uh, and so there is, there is really quite a way to go there. Um, and anticipating this progress, uh, we of course make plans around it internally and we want to provide some transparency around our thinking there. And so we want to take this maybe somewhat unusual step of sharing our internal goals and goal timelines, uh, towards these very powerful systems.
## 00:07:11
**中文**
当然，这些具体日期完全可能错得离谱；但它们代表了我们目前的判断，也是我们当前进行规划和组织工作的依据。作为一家致力于自动化研究的研究机构，我们自然会思考：这会怎样影响自身工作？能够加速下一代 AI 系统研发的 AI 系统，会是什么样子？
**English / Original**
And, you know, these particular dates, we absolutely may be quite drunk about them, uh, but this is how we currently think. This is currently how, how, how we plan and organize. And so as a research organization that is working on automating research, naturally, we are thinking about how does this impact our own work? And, uh, how will AI systems that accelerate development of future AI systems look like?
## 00:07:35
**中文**
它们如何赋能对齐等研究？我们的计划是，到明年 9 月实现能力相当不错的 AI 研究实习生：通过投入大量算力，它们能切实加速研究人员的工作。我们认为这一天其实已经很近。之后，我们的目标是到 2028 年 3 月，获得能够自主完成较大型研究项目、真正称得上全自动 AI 研究员的系统。
**English / Original**
How can they empower, um, research like alignment? And so, uh, we are making plans around getting to quite capable AI research interns that can meaningfully accelerate our researchers by expanding, uh, uh, a significant amount of compute, um, by September of next year. So we believe that's actually quite close. Um, and then we, uh, look towards getting a system capable of autonomously delivering on larger research projects and a, a meaningful, uh, fully automated AI researcher, uh, by March of 2028.
## 00:08:14
**中文**
当然，在迈向这些能力极强的系统时，我们会深入思考安全与对齐。事实上，我们在部署与安全方面的许多工作，以及对学习过程和能力发展的理解，都可以看作是在为这些强大模型做准备。
**English / Original**
Um, and so of course as we look towards these very capable systems, uh, we think a lot about safety and alignment, right? And in fact, a lot of our work, uh, both on deployments and safety, uh, but also just on, on, on on understanding the learning and the development capabilities side, we can think of as preparation for these very capable models.
## 00:08:36
**中文**
安全是一个多层面的问题。我们通常用五个层次来组织思考，从最内在于模型的因素，一直到最外部的因素。处于核心、也是我们认为超级智能长期安全中最重要的问题，是价值对齐。
**English / Original**
Um, safety is a multifaceted problem. And so the way we generally structure our thinking, uh, are these five layers ranging from, uh, factors that are most internal to the model, to ones that are most external. And so at the core, right? And what we believe is the most important, uh, long-term safety question for super intelligence is value alignment.
## 00:09:00
**中文**
所谓价值对齐，可以理解为：AI 从根本上真正关心的是什么？它能否遵循一些高层原则？面对不清晰且相互冲突的目标时，它会怎么做？它是否缺乏人性关怀？我们之所以认为驱动 AI 的这类高层目标或原则如此重要，是因为系统会开始进行非常长时间的思考。
**English / Original**
So to put this, um, this value alignment, uh, you can think of as what is, what's really the thing that the AI fundamentally cares about? Uh, can it adhere to some high level principles? Uh, what will it do if it's given unclear and conflicting objectives? Uh, does it lack humanity? Um, and the reason we believe that kind of this, this high level, uh, um, um, objectives or, or, or principles driving the AI are so important is that as we get to the systems that are thinking for very long,
## 00:09:44
**中文**
当系统变得非常聪明，开始处理处在人类能力边缘、甚至超出人类能力的问题时，要给出完整无缺的规范会变得相当困难。因此，我们必须依赖这种更深层的对齐。接下来是目标对齐：智能体是否与人互动？
**English / Original**
uh, as they become very smart, as they tackle problems that are, uh, um, at the edge or perhaps beyond human ability, uh, really getting to like complete specifications becomes quite difficult. Uh, and so we have to rely on this, on this deeper alignment. Um, then there's goal alignment. Um, does the agent, uh, interact with people?
## 00:10:09
**中文**
它如何与人互动，又如何遵照指令行事？再下一层是可靠性：AI 能否正确校准自己的预测？能否在简单任务上保持可靠、在困难任务上表达不确定性？能否应对略显陌生的环境？然后是对抗鲁棒性，它与可靠性密切相关，但关注的是对抗性场景。
**English / Original**
How does interact with people, how does it, um, do it following instructions? Um, then reliability. Um, can the AI correctly calibrate its predictions? Uh, can it be reliable on easy tasks, express uncertainty on hard ones? Can it deal with environments that are a little bit unfamiliar familiar? Um, then we have adversarial robustness, which is very related to reliability, but it's about adversarial settings.
## 00:10:34
**中文**
也就是说，AI 能否抵御人类或 AI 对手的定向攻击？最外层是系统性安全，即不依赖 AI 本身的智能或对齐程度，对整个系统行为作出的保障。例如系统安全机制、AI 能访问哪些数据，以及它能够使用哪些设备。
**English / Original**
So can, can the AI withstand targeted attacks from human or AI adversaries? And then the outer layer, systemic safety, which are guarantees about the behavior of the overall system that do not rely on the AI's intelligence or alignment. Um, so for example, this can be security or what data does the AI have access to, um, or, um, um, uh, um, what devices it it can use.
## 00:11:04
**中文**
我们在这些领域投资了多条研究路线；与此同时，仅仅是深度学习整体的发展和我们对它理解的提升，也带来了相当大的进展。下面我想稍微深入一些技术细节，介绍其中一个具体方向。
**English / Original**
And so we invest in multiple research directions across these domains, and we have seen, uh, quite a lot of progress also come from just the general, uh, development and improving understanding of deep learning, uh, as a whole. And, uh, I want to, I I want to take a slightly deeper technical dive here, um, and talk about the particular direction.
## 00:11:27
**中文**
价值对齐是个难题，显然还没有解决。不过，现在出现了一种很有前景的新工具，可以帮助我们研究它，那就是“思维链忠实度”。从首个推理模型开始，我们就对此投入很多。我们一直在探索这一新的可解释性方向，其核心思想是让模型内部推理的一部分不受监督。
**English / Original**
Uh, value alignment is a hard problem, right? It's definitely not solved yet. Um, however, there is a new promising tool that aids our study of it. Uh, and that is chain of thought faithfulness. Uh, it's something we invest in very heavily, um, starting from our first reasoning model. So we have been pursuing this new direction and interpretability, and the idea is to keep parts of the model's internal reasoning free from supervision.
## 00:11:55
**中文**
也就是说，训练时不要查看这部分内容，让它继续真实反映模型的内部过程。我们避免引导模型去想所谓“正确的念头”，从而让这些推理更忠实地呈现它实际在想什么。
**English / Original**
So don't look at it, uh, during training, and that's let it remain representative of the model's internal process. Um, so we refrain from, from, from, from kind of guiding the model to think good thoughts and, and, and, and, and so let it, let, let it remain a bit more faithful to, to, to what it actually thinks, right?
## 00:12:19
**中文**
当然，这并不保证一定有效。我们无法为深度学习给出数学证明，所以需要持续研究。不过，有两个理由让人保持乐观。第一，我们已经看到了很有希望的实证结果；这项技术在内部得到了大量应用。
**English / Original**
And this is not guaranteed to work, of course, right? Like we cannot make, uh, mathematical proofs about deep learning. And so this is something we study. Uh, but there are two reasons to be optimistic. One reason is that we have seen very promising empirical results. Uh, this is a technology we employed a lot internally.
## 00:12:38
**中文**
我们用它来理解模型如何训练，以及模型的行为倾向如何随训练演变。我们也与外部成功合作，例如研究模型的预谋行为。第二，它具有可扩展性：我们明确地让规模化目标不要与监控模型的能力相对抗。
**English / Original**
Uh, we use this to understand, um, how our models, um, um, train HHH how their propensities evolve over training. Uh, also we have had successful, uh, external collaborations on investigating the models prevent the scheme, for example. Um, and secondly, uh, it is scalable and in the sense that explicitly we make the scalable objective not adversarial to our ability to monitor the model.
## 00:13:08
**中文**
当然，让目标不与我们的模型监控能力相对抗，只是成功了一半。理想情况下，还应让它主动帮助我们监控模型。因此，我们正在大力研究这一点。不过，需要特别强调的是，思维链忠实度有一定脆弱性。
**English / Original**
Um, and of course, an objective not being adversarial to the ability to monitor our model is only half the battle. Um, and, you know, ideally you want it to, to to, to get it to help with monitoring the model. And so this is something we're, we're researching quite heavily. Um, but one important thing to underscore about chain of thought faithfulness is it's somewhat fragile.
## 00:13:31
**中文**
它确实要求划出清晰边界、建立明确抽象，并克制访问思维链的方式。从算法设计到产品设计，这一原则贯穿 OpenAI。以 ChatGPT 的思维链摘要为例：如果没有思维链摘要器，而是让完整思维链始终可见，会怎样？
**English / Original**
Um, it really requires drawing this clean boundary, uh, and having this clear abstraction, uh, and having restraint in in what ways you can access the chain of thought. And this is something that is present, uh, at OpenAI from algorithm design to the way we design our products, right? So, so if you look at the chain of thought summaries in charge GPT, uh, if we didn't have the chain of thought summarizer, if we just make the chain of thought fully visible at all times, right?
## 00:13:59
**中文**
久而久之，它会成为整体使用体验的一部分，也就很难不受到某种监督。因此，从长远看，我们相信，为模型保留这种适度、受控的隐私，可以维持我们理解其内部过程的能力。随着我们迈向能力极强、长时运行的系统，这可能成为一项影响重大的技术。
**English / Original**
That would make it kind of part of the overall experience over time, it would be very difficult to not subject it to any supervision. Um, and so long term, we believe that by preserving the same amount of this controlled privacy for the models, uh, we can retain the ability to understand their inner process. And we believe this can be a very impactful technique, uh, as we move towards this very capable long grounding systems.
## 00:14:22
**中文**
下面交还给 Sam。好吧，后面的内容很难接在这段之后讲，因为这显然是我们今天最重要的信息。不过再强调一次：我们完全可能判断错误；过去我们也定过目标，结果错得很惨。但根据眼前的图景，我们认为，到明年 9 月拥有大致达到实习生水平的 AI 研究助理，是有可能的。
**English / Original**
Um, and I'll hand this time, Okay, that's very hard to follow, uh, with the rest of this, and obviously that's the most important part of what we have to say. But, um, you know, just to, to reiterate, uh, we may be totally wrong. We have set goals and missed them miserably before, but with the picture we see, we think it is plausible that by September of next year, we have sort of a intern level AI research assistant.
## 00:14:45
**中文**
而到 2028 年 3 月——我想那恰好接近 GPT-4 发布五周年——我们可能拥有一名名副其实的 AI 研究员。这是我们研究计划的核心主线。还有两个领域需要讨论：产品和基础设施。产品方面，随着深度学习取得这些惊人进展，我们希望让它真正为人所用，帮助大家创造未来。
**English / Original**
And that by March of 2028, which I believe is almost five years to the month after the launch of GPT-4, um, we have like a, a legitimate AI researcher. And this is the core thrust of our research program. There are two other areas we wanna talk about, uh, product and then infrastructure on the product side. As we make this incredible progress with deep learning, we wanna make it useful to people to sort of invent the future, as we mentioned.
## 00:15:09
**中文**
传统上，这意味着 ChatGPT 中的 AI 超级助理；但现在我们将真正演变为一个供其他人搭建产品的平台，世界中需要接入的各类组件将由别人构建。在继续介绍之前，我想先播放一段短片，展示今天人们在 ChatGPT 中使用 GPT-5 的一些方式。
**English / Original**
And, um, what that's looked like traditionally for us is a AI super assistant inside of chat bt, but we're now really gonna evolve to a platform, uh, that other people will build on top of, and all of the pieces that need to fit in, uh, of the world, um, you know, will be built by others. Before we go talk about that, I wanted to just show a quick video of how people are using G PT five, some of the ways people are using GPT five in chat GPT today.
## 00:15:34
**中文**
我是量子物理学家。我是美甲师。人类免疫学家。钢铁工人。设计师兼开发者。经济学教授。我基本靠捕鱼为生。GPT-5 能预测我们甚至还没做过的实验结果。你能设计一套零件编号系统，让车间里的伙计容易使用吗？如果想在旧金山湾区捕珍宝蟹呢？我会问某个量子算符应用方面的问题。
**English / Original**
I'm a quantum physicist, I'm a nail technician, Human immunologist. I'm A steelworker, A designer and developer, a Professor of economics. I basically go fishing for a living. GPT five is able to predict the outcomes of experiments that we haven't even done. Can You create a part numbering system that's easy for my guys in the shop and want To catch Dungeness crab in the Bay Area Here? Would ask a question about the application of a certain quantum operator.
## 00:16:05
**中文**
它会给出非常详细的数学推导。我让它做一个相机应用，让我能在空中实时绘画。我们有了主题，也有了方向。把上百万乃至无穷无尽的创意收敛成 10 个或 20 个，然后开始测试。你会讲笑话，也会问它：经济学家应该做什么？
**English / Original**
This one gives me a very detailed mathematics. I Ask for basically a camera app where I can draw real time in the air. We have a theme, we have a direction to go. Let's go from a million or infinite amount of ideas to like, gimme 10 or 20. You just start testing it. You know, you tell the jokes, you ask it questions like, what should economists do?
## 00:16:27
**中文**
可以尝试不同饵料、不同水深。这些信息如果自己摸索，可能要花好几年和大量试错。头脑风暴的工作量很大，需要来回讨论。我可以很轻松地跟随它的推理，不必盲信结果，只要看看它做了什么。GPT-5 一次就生成了 1,001 个零件编号。
**English / Original**
Different baits. I can use different water depths. All this information that it would take years to figure out on your own. A lot of trial and error. It's tremendous work brainstorming, it's back and forth. I Can kind of easily follow the reasoning. I don't need to trust the result. I can just look, what did they do? GBD five, just like did this, uh, in one shot, 1,001 part numbers.
## 00:16:53
**中文**
如果让我自己编号，得花好几个星期，恐怕会把我逼疯。Jakub 稍后会回来和我一起问答，不过结束前还有一位特别嘉宾。我们很喜欢这些案例，希望有更多、希望到处都能这样。因此，我们希望 OpenAI 成为个人和企业都能在其上构建产品的平台。
**English / Original**
It would've taken weeks for me to number this, and it would've made me go Crossey. Uh, yaul joined me back for q and a in a little bit, but we're gonna have one special guest on before, before the end. Um, we love that. We want much more of that. We want that everywhere. So we want OpenAI to be a platform that people and companies can build on top of.
## 00:17:15
**中文**
我们现在大致看到了通往“AI 云”的路径：它不只存在于 ChatGPT，也不只限于我们创建的服务；我们希望尽可能开放技术，让更多人构建人们会依赖、使用并借以创作的东西。我想下面这句话，至少这个理念，最初来自比尔·盖茨。
**English / Original**
We can sort of see our way now to an AI cloud where this is not just in chat G pd. This is not just services that we create, but we want to expose our technology for as many people to build the things that people will depend on and use and create with as possible. I think this, this quote, or at least this like idea is originally from Bill Gates.
## 00:17:33
**中文**
至少我是先从他那里听到的：当平台上的建设者所创造的价值，超过平台创建者自身创造的价值时，你才真正建成了一个平台。这正是我们的目标。我们相信，从明年开始，可以把这项技术、这批用户以及我们建立的框架结合起来，让全世界在其上打造卓越的新公司、服务和应用。
**English / Original**
Um, at least that's where I first heard it, that, you know, you've built a platform when there's more value created by people building on the platform than by the platform builder builder. And that's our goal. Um, and that's our goal. Like next year, we, we really think we can now take this technology and this user base and this sort of framework we've built and get the whole world to build amazing new companies and services and applications on top of it.
## 00:17:56
**中文**
要做到这一点，我们还有许多方面需要演进。但在迈向平台的过程中，有两项基础原则值得说明。第一是用户自由。如果这是一个由各种人使用、建设和创作的平台，那么世界各地的人自然有非常不同的需求和愿望。
**English / Original**
To do that, uh, there will be many things that we have to evolve towards, but there's two foundational principles as we, as we move towards being this platform that I wanted to touch on. Um, one is about user freedom. If this is going to be a platform that all sorts of people are building, uh, on using, um, creating with people around the world have very different needs and desires.
## 00:18:16
**中文**
当然会有一些非常宽泛的边界，但我们希望用户能对使用方式拥有很大的控制权和定制空间。最近谈这件事时，我又犯了一个愚蠢错误；真希望当时举的不是情色内容这个例子。我原以为情色作品与色情机器人之间存在容易理解的区别。
**English / Original**
And there will, of course be some very broad bounds, but we want users to have a lot of, uh, control and customization of how they, of how they use it. Now, I made, you know, one of my many stupid mistakes when I tried to talk about this recently. Uh, I wish I had used an example other than erotica. I thought there was an understandable difference between erotica and porn bots.
## 00:18:35
**中文**
无论如何，我们想表达的是：人们需要很大的灵活性，也想以不同方式使用这些工具。在我们自己的第一方服务里，我们希望把成年用户当作成年人对待。第一方服务或许会有更严格的准则，但 AI 将成为人们生活中极其重要的一部分。
**English / Original**
But in any case, we were trying to show, the point we're trying to get across is that people need a lot of flexibility and people want to use these things in different ways. And we wanna treat our adult users, like adults in our own first party services. We may have, you know, tighter guidelines, but AI is gonna become such an important part of people's lives.
## 00:18:53
**中文**
因此，必须保留人的表达自由。与此同时，我们认为，对于隐私，世界需要采用不同于以往技术的思考方式。隐私对所有技术都重要，但对 AI 尤其重要，因为人们使用它的方式不同于过去任何技术。
**English / Original**
The freedom of human expression is going to need to be there. Along with that, we think that world will need to think about privacy in a different way than they have for previous kinds of technology. Privacy is important for all sorts of technology, of course, but privacy for AI will be especially important. People are using this technology in a different way than they've used the technologies of the past.
## 00:19:12
**中文**
人们像和医生、律师或配偶交谈一样同 AI 交流，分享生活中最私密的细节。我们当然需要强有力的技术手段保护这种隐私，也需要强有力的政策保障。我们谈过“AI 特权保密”等概念；如果 AI 将成为人们生活中的基础平台，那么强有力的隐私保护对我们而言至关重要。
**English / Original**
They're talking to it like they would to their doctor or their lawyer, their spouse. Um, they're sharing the most intimate details of their lives. And of course, we need strong technical protections on that, privacy of that privacy. But we also think we need strong, um, policy protections of that privacy. We've talked about concepts like AI privilege, um, but really strong protections if AI is going to be this fundamental platform in people's lives, seem super important to us.
## 00:19:38
**中文**
接下来谈基础设施。我知道外界对我们的基础设施建设现状存在许多困惑，所以我们决定完全透明地说明。到目前为止，我们作出的全部基础设施建设承诺合计略高于 30 吉瓦。
**English / Original**
Okay, and then I wanna go onto infrastructure. So I know there's been a lot of confusion about sort of where we are in our infrastructure build out, and we figured we would just be super transparent about that. So where we are today, um, all of our commitments total a little bit over 30 gigawatts of, uh, of infrastructure build out.
## 00:19:55
**中文**
这意味着未来多年里，我们承担的总财务义务约为 1.4 万亿美元。这是目前已经承诺的规模。我们当然希望以后做得更多，但根据今天看到的图景、预期的收入增长和融资能力，这是当前令我们感到稳妥的水平。
**English / Original**
Um, and that's about a $1.4 trillion total, uh, financial obligation for us over the next many years. Um, this is what we've committed to so far. We of course hope to do much more, but given the picture we see today, given what we think we can see for revenue growth, our ability to raise capital, this is what we're currently comfortable with.
## 00:20:14
**中文**
这需要大量合作伙伴。我们已经介绍过许多优秀的芯片合作伙伴；还有人为我们建设数据中心、提供土地和能源，也会涉及芯片制造设施。它已经需要相当多的供应链创新。我们很高兴能与 AMD、博通、Google、微软、英伟达、甲骨文、软银以及许多其他伙伴合作，真正把它实现。
**English / Original**
This requires a ton of partnerships. We've talked about, uh, many of our great chip partners. Uh, there are people building the data centers for us, land energy. Uh, there'll be chip fab facilities. This is already getting to require quite a lot of supply chain innovation, and we're thrilled to get to work with, uh, a MD, Broadcom, Google, Microsoft, and video, Oracle, SoftBank, many others to really make this happen.
## 00:20:35
**中文**
但这仍处于早期。如果 Jakub 所说的研究成为现实——我们相信会如此——如果这些技术产生经济价值，人们也希望使用所有这些服务，那么我们需要的规模会远不止这些。我想明确一点：我们还没有对此作出承诺，但已经开始讨论。
**English / Original**
But this is still early. If the, if the work that Yaku talks about comes to fruition, which we think it will, and if the economic value of these things, um, happen and people want to use all these services, we're gonna need much more than this. So I wanna be clear, we're not committing to this yet, but we are having conversations about it.
## 00:20:54
**中文**
我们的愿景是建成一座“基础设施工厂”，每周可以新增 1 吉瓦算力；还希望把成本显著降至每吉瓦约 200 亿美元——按设备五年生命周期计算。要做到这一点，需要大量创新、大量合作，当然也需要收入大幅增长。
**English / Original**
Our aspiration is that we can build an infrastructure factory where we can create one gigawatt a week of, uh, compute. And we aspirationally would like to get that cost down significantly, um, to like $20 billion a gigawatt over the five year lifecycle of, uh, you know, that equipment to do this will require a ton of innovation, a ton of partnerships, obviously a lot of revenue growth.
## 00:21:17
**中文**
我们可能得重新考虑机器人工作的方向，让机器人帮助建设数据中心，而不是去做其他所有事情。但这就是我们想去的方向。未来几个月，我们会做大量工作，看看能否抵达那里。距离我们在财务上真正有条件扣动扳机、启动这一计划，还需要一段时间。
**English / Original**
Um, we'll have to repurpose our thoughts about robotics to help us build data centers instead of doing all the other things. Um, but this is where we'd like to go. And over the coming months, we are gonna do a lot of work to see if we can get here. Um, it will be some time before we're in a financial position where we could actually pull the trigger and get going.
## 00:21:33
**中文**
1 吉瓦是个很大的数字，所以我们准备了一段短片帮助大家理解其规模。这是我们正在得州阿比林建设的数据中心，也是第一个“星际之门”站点。目前我们在全美建设多个这样的站点，这一处进展最快。每天仅现场施工人员就有数千人。
**English / Original**
On this one. Gigawatt is like a big number, but I figured we would show a little video to put this into perspective. Um, this is, uh, a data center that we're building in Abilene, Texas. This is the first Stargate site. We're doing several of these now around the country, but this one is the furthest along. There's like many thousands of people that work here every day just doing the construction at the site.
## 00:21:53
**中文**
在整个供应链中，为实现这一切而工作的人可能有数十万乃至数百万：设计芯片、制造芯片、组装设备；能源方面也包含大量工作。每新增 1 吉瓦，都需要完成数量惊人的事项。
**English / Original**
There's probably hundreds of thousands or millions of people that work in the supply chain to make all this happen, to design these chips, to fab these chips, to put them together. Um, there, you know, there's all of the work that goes into this for energy. There's an enormous amount of stuff that has to happen, um, for each one gigawatt.
## 00:22:10
**中文**
我们希望找到办法，让这一过程效率高得多、成本低得多，也更容易扩展，从而交付研究路线图以及人们各种使用方式所需的基础设施。接下来是我们的新组织架构。大家可能见过此前那张极其复杂、展示 OpenAI 各实体的示意图。
**English / Original**
And we wanna figure out how we can make this way more efficient and way cheaper and way more scalable so that we can deliver on the infrastructure that the research roadmap requires. And that all of the ways that people will want to use this need, um, to enable that. Uh, we have a new structure. So, um, maybe you saw before this like crazy convoluted, uh, diagram of all of the open entities.
## 00:22:34
**中文**
现在简单多了。我们有一家名为 OpenAI Foundation 的非营利组织，它掌握控制权，董事会设在这里；该基金会还持有名为 OpenAI Group 的公益公司（PBC）的一部分股权。也就是说，非营利组织居于控制地位，公益公司位于其下。
**English / Original**
Now it's much simpler. We have a non-profit called the Open Air Foundation that is in control of, uh, where the board sits, uh, or where, uh, let's come back to the board, uh, where the board also sits and, uh, owns a slice of our PBC Public Benefit Corporation called OpenAI Group. So, uh, nonprofit and control, public Benefit Corporation sits under it.
## 00:22:56
**中文**
我们希望 OpenAI Foundation 成为史上规模最大的非营利组织。正如刚才几次提到的，我们认为，科学以及让科学成果得到广泛分配的制度，是世界取得最大改善的途径之一。科学不会是基金会资助的唯一领域，但会是我们首批工作的一个重要重点。
**English / Original**
Um, we hope for the Open AI Foundation to be the biggest nonprofit ever. As I mentioned now, uh, a few times, science is one of the ways that we think the world most improves, along with the institutions that broadly distribute the benefits of that. So then science will not be the only thing that the nonprofit funds, but it will be an important first major area of the things that we do.
## 00:23:20
**中文**
非营利组织将治理 PBC。它最初持有 PBC 约 26% 的股权；如果我们表现非常好，通过认股权证，这一比例可以随时间提高。它会利用这些资源，根据技术所处阶段和社会需要，追求我们认为 AI 能带来的最大公共利益。PBC 则会更像一家普通公司那样运营。
**English / Original**
The nonprofit will govern the PBC. It will initially own about 26% of the PBC equity, but that can increase over time with warrants if, if we perform really well. And it will use these resources to, uh, pursue what we think are the best benefits of ai. Uh, given where the technology is and, and, and what society needs. The PBC, uh, will operate more like a normal company.
## 00:23:41
**中文**
它将承担同样的使命，并受该使命约束；在安全事务上尤其如此。同时，它也能吸引我们建设巨型基础设施所需的资源，以服务研究和产品目标。基金会未来会做更多事情，但在初期，我们希望先把一件事做到极致。
**English / Original**
It'll have the same mission. Um, it will be bound to that mission. Um, and, you know, and matters of safety will only be bound to that mission, but it will be able to attract the resources that we need for that gigantic infrastructure build out to serve the research and product goals that we have. So the initial focus of the, the foundation, we'll do more things over time, but we want to knock something outta the, uh, outta the park.
## 00:24:03
**中文**
希望如此。第一项是承诺投入 250 亿美元，用 AI 帮助攻克疾病。实现方式有很多：生成数据、提供大量算力、向科学家拨款，以及支持 AI 韧性。AI 韧性是一个新领域，我认为非常重要。下面请我们的联合创始人 Wojciech 上台，介绍它会是什么样。
**English / Original**
Uh, hopefully. First is a $25 billion commitment to use AI to help cure disease. There are a lot of ways this can happen. Um, generating data, um, you know, using a lot of compute grants to grants to scientists, and also for AI resilience. Um, AI resilience is a new and I think very important area, and I'd like to invite, uh, our co-founder EK up to talk about what this will, what this will look like.
## 00:24:26
**中文**
大家好，很高兴来到这里。谢谢你来。所谓“AI 韧性”，比我们以往讨论的 AI 安全范围更广。谈韧性时，我们认为高级 AI 会带来风险和扰动，希望建立一个由不同组织组成的生态系统，帮助解决其中的一系列问题。
**English / Original**
Hello, uh, glad to be here. Thanks for being Here. So, um, term AI resilience is, um, little bit broader than what we historically taught about AI safety. So in case of the, uh, resilience, we think that advanced AI comes with risks and disruptions, and we would like to have an ecosystem of organizations that, uh, can help to, um, solve a number of these problems.
## 00:24:55
**中文**
举个例子会更容易理解。我们都相信 AI 会推动生物学进步；但随着能力提高，也存在恶意行为者用 AI 制造人为大流行病的风险。在安全层面，一种缓解措施是确保模型拦截与病毒学相关的危险请求。
**English / Original**
Um, so let me give you an example to better illustrate it. Um, we all believe that the ai, uh, will advance in biology. And as it advances in biology, there is a, there is a risk that some bet actor could use AI to create manmade pandemics. So the, on the safety level, the mitigation would be to make sure that the models, uh, block the queries that have to do with, uh, uh, virology.
## 00:25:26
**中文**
然而，考虑整个 AI 行业，即使 OpenAI 拦截这些请求，仍很可能有人使用其他模型制造病原体。韧性的思路并不只是阻止它，还要在问题真的发生时具备快速响应能力。
**English / Original**
Um, however, um, if you consider the entire, uh, entire, um, ai, uh, industry, it's very likely that even if open AI blocks it, someone could use, uh, different models out there and still, uh, uh, produce, uh, pathogens and, um, the in case of resilience, we don't want just to block it, but also have a, a rapid response if the problem would occur.
## 00:25:52
**中文**
我想到的风险和扰动有很多：心理健康是一项，生物风险是另一项，工作岗位流失可能又是一项。我们认为需要一个生态系统。我喜欢用网络安全作类比：互联网早期，人们甚至不放心在网上填写信用卡号，因为太容易遭到攻击。
**English / Original**
So when I think about the, uh, risks, uh, and disruptions, there are just many. The, the mental health is one of them. Bio is, uh, another one, another one, uh, job displacement might be another one. And we think that we need the ecosystem. And maybe a good analogy that I like is cybersecurity. So at the beginning of internet, uh, it was actually, it was a place that people didn't feel comfortable putting their credit card numbers because it was so easy to get hacked.
## 00:26:24
**中文**
那时一旦出现病毒，人们会互相打电话，提醒把电脑断开网络。此后我们走了很远。如今已有完整的网络安全产业与基础设施，保护关键基础设施、政府、企业和个人用户，保护程度高到人们愿意把最私密的数据、乃至毕生积蓄放到网上。
**English / Original**
And, uh, when there was a virus, people are giving each other a call to disconnect the computer from internet. And we got a long way. At the moment. There's entire infrastructure of cybersecurity companies. They are protecting, uh, um, the critical infrastructure governments, corporations, and individual users to such extent that people are willing to put the most personal data, uh, online to have life savings, um, be online.
## 00:26:55
**中文**
网络安全已经发展得很成熟，我们认为 AI 领域也会出现类似的“AI 韧性层”。我非常兴奋，非营利组织将帮助推动并建立这样的生态系统。我也一样。现在正是开展这项工作的关键时刻，我很期待你找到让它真正落地的办法。
**English / Original**
Um, yeah, so the, the cybersecurity got really far and we think that something analogs will be present for ai, that there will be AI resilience layer. And, uh, I'm really excited that the, uh, nonprofit, uh, will help out, uh, to stimulate it to create, uh, such a ecosystem. So am I, I think this is a important time to be doing this, and I'm very excited to that you're gonna like, figure out how, how we, how we go off and make it happen.
## 00:27:23
**中文**
再次说明，这些不会是非营利组织资助的全部事项，但我们很高兴把它们作为最初两项重点：利用 AI 开发疾病疗法，以及在我们探索如何将 AGI 部署到社会时，推动新的 AI 韧性计划。刚才说了三大支柱，但如果这一切真的成功了呢？
**English / Original**
So a again, these are not the only things that the nonprofit will fund, but we're excited about these as the first two, using AI to develop cures and, uh, treatments for diseases and this new AI resilience effort as we figure out what the deployment of a GI into society is going to look like. So we mentioned that those are our three pillars, but you know what, what if this all works?
## 00:27:46
**中文**
我们认为，2026 年当年的模型有可能开始取得小型发现；到 2028 年，可能取得中等、甚至更大的发现。至于 2030 年和 2032 年会怎样，谁知道呢？如果 AI 能像以往那样持续推动科学进步，我们认为未来可以非常光明。
**English / Original**
We, we think it is plausible that in 2026 we start to see the models of that year begin to make small discoveries by 2028, medium or maybe even larger discoveries. And, you know, who knows what 2030 and 2032 are going to look like. If AI can keep advancing science as has happened in the past, um, we think the future can be very bright.
## 00:28:09
**中文**
当然，我们认为，让人类自主决定如何走过这个未来非常重要；但新科学进展为我们打开的可能空间令人惊叹。于是，我们请 Sora 回望过去，帮助想象一个彻底变好的未来。我们尤其感兴趣的是，科学史如何在一项项成果上层层累积。
**English / Original**
Of course, we think it's very important that humans can self determine our way through this future, but the open space that new scientific advances give us is quite impressive. So we asked SOA to help us imagine a radically better future by looking at the past. And we are particularly interested in how the history of science builds on itself.
## 00:28:27
**中文**
发现接着发现——这正是我们希望 AI 带来的。这里展示的是 200 年的科学发展；如果这些彼此支撑、复利累积的 200 年发现，不是在 200 年，而是在 20 年甚至两年内完成，看看这会加速多少，再想想此前不敢想象的可能性，一个截然更美好的未来就会变得现实：这里一座数据中心在发现癌症疗法，那里一座数据中心在制作史上最好的娱乐内容。
**English / Original**
Discovery after discovery, this is what we hope will happen with ai. You know, this is gonna be 200 years of science, but if you can do these 200 years of compounding discoveries of the scaffolding building up on each other, not in 200 years, but in 20 years or in two years, and if you look at how much this, this is accelerated, and think about what could be possible before you can imagine a world, uh, where a radically better future becomes quite possible, you have a data center here that is discovering a cure for cancer, a data center there that's making the best entertainment ever.
## 00:28:59
**中文**
这里的数据中心帮你找到未来的丈夫或妻子；另一座在建造火箭，帮助人类殖民太空；还有一座帮助解决气候危机。过去我们用传统方式完成所有这些事情，如今借助 AI，我们将能以大得多的力量塑造未来。
**English / Original**
A data center here that's helping you find your future husband or wife. This one is building rockets and helping you colonize space. This one is helping to solve the climate crisis. So we did all this stuff the old fashioned way. Um, and now with the help of ai, we'll be able to shape what comes next, uh, with maybe much more power.
## 00:29:15
**中文**
我们谈了一些 AI 医疗，也非常看好机器人；我们确实认为能源对世界极其重要。我们希望探索个性化教育的含义，设计新型材料，以及许多目前甚至还想不到的事情。在 OpenAI 进入下一阶段、而更重要的是深度学习持续进步之际，感谢大家今天参加。下面尝试一种新形式：直接回答问题。
**English / Original**
So we talked a little about AI medicine. We're very excited about robots. Um, we really think energy is very, very important to the world. We want to figure out what personalized education can mean, design novel materials, and probably a ton of other things that we can't even think of yet. So as we head into this next phase of OpenAI, and more importantly than that, this continual progress in deep learning, um, we thank you for joining us today and we're gonna try something new now, which is, we're gonna just answer questions.
## 00:29:43
**中文**
如果效果不错，以后我们会多做。Jakub 会回来参加问答。非常感谢。谢谢 Sam。这对我们是个新形式，第一次尝试，请大家包涵。如果确实有用，我们今后会经常进行。我们将按照问题获赞数从高到低回答。
**English / Original**
If this works, it's something we are, we'll try more in the future. Yakob is gonna rejoin for this q and a. Thank you very much. Thank you Sam. Um, but, uh, this is a new format for us, so bear with us as we try it this first time again. If, if, um, if this is useful, it's something we'll do, uh, again a lot more. And we're gonna try to just answer questions in the order they are most upvoted.
## 00:30:03
**中文**
准备好了吗？好，我们看看效果。大家可以在 Vimeo 链接中提交问题，我们会直接开始回答。Caleb 问：你们曾警告科技产品正变得令人上瘾并侵蚀信任，可 Sora 模仿 TikTok，ChatGPT 又可能加入广告。为什么重复自己批评过的模式？你们将如何通过行动而非言辞重建信任？
**English / Original**
Um, are we good to go? Alright, let's see how this works. So, uh, you can put questions in the Vimeo link and we will just start answering them. So from Caleb, we've warned that the tech is becoming addictive and eroding that tech is becoming addictive and eroding trust, yet Soar mimics TikTok and Chacha. BT may add ads. Why repeat the same patterns you criticize and how will you build, rebuild, trust through actions and not just words?
## 00:30:26
**中文**
这确实令我们担忧。我担心的不只是 Sora、TikTok、ChatGPT 广告这类已知问题——这些或许可以通过谨慎设计加以处理。我们还看到，人们会和聊天机器人建立我们未曾预料的关系。鉴于这种互动机制和现实世界的竞争，其中显然可能出现成瘾行为。
**English / Original**
We're definitely worried about this. Uh, I worry about it not just for things like SOA and TikTok and ads and Chacha, bt, which are maybe known problems that we can design carefully, but you know, we have certainly seen people develop relationships with chatbots that we didn't expect. And there can clearly be addictive behavior there, given the dynamics and competition in the, in the world.
## 00:30:47
**中文**
我猜一些公司会推出极易使人上瘾的新型产品。你们只能根据我们的行动来判断我们。我们会犯错，也会尝试撤回有问题的模型。如果推出 Sora 后，它变得极度令人上瘾而不再关乎创作，我们就会取消这个产品；大家要根据这样的行动评价我们。
**English / Original**
I suspect some companies will offer very addictive new kinds of products. Um, and I think you'll have to just judge us on our actions. We'll have to, you know, we'll make some mistakes, we'll try to roll back models that are problematic. If we ship Soar and it becomes super addictive and not about creation, we'll, you know, cancel the product and you'll, you'll have to just judge us on that.
## 00:31:06
**中文**
我希望、也相信，我们不会重蹈前人的覆辙。我想他们当初也并非有意犯那些错误；大家都在共同探索。不过，我们很可能会犯新的错误，只能快速调整，并保持紧密的反馈闭环。
**English / Original**
My hope and belief is that we will not make the same mistakes that companies before us have made. Uh, I don't think they meant to make them either. It's, uh, you know, we're all kind of discovering this together. We probably will make new ones though, and we'll just have to evolve quickly and have a tight feedback loop.
## 00:31:21
**中文**
我们可以想象这项技术为世界带来巨大福祉的各种方式，也能看到明显的负面用途。使命会指引我们持续改进产品。对于 ChatGPT、Sora 等产品应优化什么，有一点让我们很有希望：着眼于极长期进行优化。这与我们总体上不断延长模型可有效工作时间跨度的方向天然一致。
**English / Original**
We, we can imagine all sorts of ways. This technology does incredible good in the world, also obvious bad ones. And, um, you know, we're guided by a mission where we'll just continuously evolve, evolve the product. Um, one thing that we are quite helpful about for in terms of, um, what we opt, optimize for and, and, uh, products like Chad, GPT or, or Sora is thinking about, um, optimizing for the very long term, which is naturally very aligned with, uh, how we think in general about extending, uh, the horizon on which the models can work productively.
## 00:31:56
**中文**
我们相信，这方面还有很大的发展空间，最终可以让模型真正优化长期满意度和福祉，而不只是短期信号。下一个问题：成人模式上线后，是否可以永久保留 GPT-4o？
**English / Original**
Um, and so we believe that quite a lot of development is possible there, and we can eventually get to models that really optimize for, um, long term, uh, satisfaction and, and, and wellbeing instead of just short term signals. Um, okay, next question. Will we have an option to keep the four oh model, uh, permanently after adult load is installed?
## 00:32:20
**中文**
“负责任的成年人不需要更安全的模型。”我们目前没有让 4o 退役的计划。当然，也不会承诺一直保留到宇宙热寂；但我们理解，一些用户确实非常喜欢这个产品。我们也希望大家理解，为什么我们认为它不适合未成年人使用。
**English / Original**
We don't need a safer models responsible adults. Um, we have no plans to sunset four. Oh. Uh, we are not gonna promise to keep it around to the heat death of the universe either, but we understand that it's a product that some of our users really love. We also hope other people understand why it was not a model that we thought was healthy for miners to be using.
## 00:32:41
**中文**
我们希望不断打造更好的模型，让人们更喜欢。你生活中与之建立关系的人，也会成长、变聪明，并随时间略有变化；我们希望模型也能如此。不过，是的，目前没有让 4o 退役的计划。哇，关于 4o 的问题真多。
**English / Original**
Um, we hope that we build better models over time, that people like more, you know, the people you have a relationship with in your life, they evolve and get smarter and change a little bit over time. And we think that, we hope that the same thing will happen. But yeah, no, no plans to, uh, no plans to sunset four. Oh, currently, uh, wow, we have a lot of four oh questions.
## 00:33:02
**中文**
好吧，时间有限，不可能逐一回答。但我们希望人们能使用自己想用的模型，不希望用户觉得我们在不同模型之间暗中改道。我们希望成年人能够像成年人一样自己选择——前提是我们没有做类似贩卖海洛因之类的事；顺便说一句，那当然也不该做。
**English / Original**
All right. We're not gonna, in the interest of time, we will not go through, uh, all of these. But, but yeah, we don't, you know, we want people to have models that they want to use. We don't want people to feel like we're routing them around models. Um, and you know, we, we want adults to make choices as adults, As long as we think we're not, you know, selling heroin or whatever, which also you, you shouldn't, you shouldn't do.
## 00:33:24
**中文**
对于希望进行情感交流的人，正如我们说过的，我们希望允许更多这样的互动，也计划这样做。这里有个很好的匿名问题。Jakub，AGI 什么时候会出现？我想若干年后回望今天，我们会说，这几年就是 AGI 发生的过渡期。
**English / Original**
Um, so people that wanna have emotional speech, uh, as we've said, we wanna allow more of that and we plan to, okay, here's a good anonymous question for here, Jako, when will a GI happen? Um, so I think, I think in, in some number of years, we'll look back at these years and we'll say, you know, this was kind of the transition period when a GI happened.
## 00:33:51
**中文**
正如 Sam 所说，OpenAI 早期对 AGI 的想象带有很强的情感色彩：它仿佛是解决一切问题的终极事物，是一个划分“之前”和“之后”的单一时刻。
**English / Original**
Um, I think, you know, one way we thought about, um, I think as, as some said like early on at opening, we thought about a GI kind of emotionally as this like thing that is like that kind of ultimate solution of all the problems and, and, and, um, it's, it's this like single point, um, for which there is, uh, before and after.
## 00:34:15
**中文**
后来我们发现，现实比这更连续。过去看起来像迈向 AGI 的明显里程碑的各种基准，现在更像是在提示：我们大概还相距多少年。
**English / Original**
And, um, I think, um, we found that it's, uh, a bit more continuous than that. Um, and, and so in particular for like various kind of benchmarks that, you know, seemed at, uh, uh, uh, seemed like kind of the obvious like milestones towards agi, I, I think, I think we now think of them as kind of like indicating like, you know, roughly how far away we are in years.
## 00:34:42
**中文**
回顾一连串里程碑：计算机先在国际象棋中战胜人类，接着是围棋，再到能够使用自然语言交流、解决数学问题。这些成就显然正逐块拼合在一起。
**English / Original**
And so, uh, you know, if you look at the succession of, of, of, of, of milestones such as computers being humans at chess and then at go and then, uh, you know, computers being able to speak in natural language and computers being able to solve math problems, right? I think, well, they clearly kind of get, uh, puzzled together.
## 00:35:02
**中文**
我会说，“AGI”这个词已经承载了太多含义。正如 Jakub 所说，这是一个持续多年的过程，而我们正处于其中。我们今天作这番介绍的原因之一，就是与其试图用一个 AGI 定义满足所有人，不如明确说：我们的目标是到 2028 年 3 月实现真正的自动化 AI 研究员，并具体界定它意味着什么。这样更有用。
**English / Original**
Um, Yeah, I, I would say, I think it's the a GI term has become hugely overloaded and as Jakob said, it'll be this process over a number of years that we're in the middle of. Uh, but one of the reasons we wanted to present what we did today is I think it's much more useful to say our intention, our goal is by March of 2028 to have a true automated AI researcher, um, and define what that means, uh, than it is to sort of try to, you know, satisfy everyone with the definition of a GI.
## 00:35:32
**中文**
还有一点值得一提，而且可能有些反直觉：我们显然在研究一项非常复杂的技术，试图理解所有这些算法。起初，我们可能想象 AGI 是这样一个时刻：所有答案都已经找到，它就是某种最终形态。
**English / Original**
And, and maybe one other thing to mention, right? Like, I think like one kind of counterintuitive thing here is that obviously we're working with like a pretty complicated technology. We're trying to understand all these algorithms and maybe initially we kind of imagined that like a GI is the moment once you, where you kind of have figured out all the answers and, and, and, and, and it's kind of the, the final thing.
## 00:35:52
**中文**
如今我们越来越意识到，智能更像一条曲线，或许还是多维的，而人类处在其中某个位置。随着深度学习规模扩大、新算法不断开发，我们会逐渐接近那个点，最终超过它；事实上，在多个维度上已经超过了人类。
**English / Original**
And I think now we increasingly realize that, you know, there is kind of a, some, some curve of intelligence, maybe a multidimensional one, and, you know, humans are somewhere on it. And as you scale deep learning, as you develop these new algorithms, like eventually, well, you kind of inch closer to that point and eventually, and eventually will surpass it, and you already have surpass the multiplexis.
## 00:36:12
**中文**
但这并不意味着围绕它的所有问题都已经解决，这一点需要我们认真思考。能否说明内部模型比已部署模型领先多少？我们对下一代模型抱有很高期待。
**English / Original**
Um, and, and so, and that doesn't actually mean you have solved all the problems around it, which is something we, we need to seriously think about. Can you give us an idea of how far ahead internal models are compared to deployed ones? Um, I think, um, we, we have quite strong expectations for our, for, for, for our next models.
## 00:36:40
**中文**
我们预计未来几个月乃至一年内会取得相当快速的进展。不过，我们并不是藏着某个极其疯狂的成品。实践中通常是这样：我们开发出许多组件，每个都是来之不易的成果；然后我们知道，把它们组合起来会得到相当惊人的东西，而且对此能够作出较可靠的预测。
**English / Original**
Um, so I think, I think we, we expect quite rapid progress over the next, uh, uh, couple months and a year. Um, yeah, I think, um, but we, we we haven't been like withholding something, something, uh, extremely crazy about it. Yeah. Yeah. One, one of the ways this, uh, kind of often works in practice is there's like a lot of pieces that we develop and that, you know, they're all kind of hard won victories, and then we know that when we put them together, um, we will have something quite impressive and we're able to predict that fairly well.
## 00:37:16
**中文**
我们今天的部分目的，就是告诉大家，我们已经拥有许多这样的组件。并不是说我们正坐拥某个庞然大物却不向世界展示；而是我们预计，到一年以后——具体说，围绕 2026 年 9 月的目标——我们有切实机会在能力上迈出极其重要的一大步。
**English / Original**
Uh, part of our goal today is to say that, um, we have a lot of those pieces. It's not like we're kind of currently sitting on this giant, you know, thing that we're not showing to the world, but that we expect by a year from now, certainly with this September of 2026 goal that we have a a light mean, not likely we have a realistic shot at a, like, tremendously important step forward in capability.
## 00:37:44
**中文**
Ronan 问：OpenAI 对与 Anthropic、Gemini 或 xAI 等实验室开展联合研究、共享算力和安全合作持什么态度？我们认为，这在安全方面会越来越重要。各实验室需要共享安全技术和安全标准。可以想象，在进入递归式自我改进阶段前，全世界会说：我们真的需要一起谨慎研究。
**English / Original**
Um, what does OpenAI Ronan asks, what does OpenAI stance on partnering with labs like philanthropic, Gemini or x AI for joint research, compute sharing and safety efforts? Uh, we think this is gonna be increasingly important on the safety front. Uh, labs will need to share safety techniques, safety standards. Um, you can imagine a time when the whole world would say, uh, okay, before we hit a recursive self-improvement phase, we really need to all carefully study this together.
## 00:38:08
**中文**
我们欢迎这样的合作，我认为它会非常重要。以之前谈到的思维链忠实度为例，我们已经开始讨论建立行业规范，并与 Google、Anthropic 及其他实验室的研究人员开展了一些联合调查。
**English / Original**
Um, we welcome that collaboration. I think it'll be quite important. Uh, One thing to mention on the, on the on chain of thought faithfulness that I talked about earlier, um, we actually have, um, started talking about establishing industry in arms and, and we started some, some joint investigations with researchers from Google and ONA Tropic, uh, and, and, and some other labs.
## 00:38:29
**中文**
我对此非常兴奋。这正是一个跨实验室合作能令所有人受益的领域。匿名提问：你们是否会开源一些旧模型，比如最初的 GPT-4？也许有一天会把它们作为“博物馆藏品”发布，但 GPT-4 并不是特别有用的开源模型。
**English / Original**
And, uh, yeah, that's something I'm very excited about. And I think that is an example of something where we can really benefit from Collaborating across multiple apps. Anonymous asks, will you ever open source some of your old models like the original GPT-4? Um, we might do those as museum artifacts someday, but they're not, like GPT-4 is not a particularly useful open source model.
## 00:38:48
**中文**
它体量大，表现也没那么好。我们或许能用很小的规模做出能力超过 GPT-4、对人们真正有用的模型。对于实用模型，我预计今后会更多采用这种方式；至于好玩的博物馆藏品，也许有一天会做，谁知道呢？可以做出许多很酷的东西。
**English / Original**
It's big, it's not that good. Uh, you know, we could probably make something that is beyond the power of GBT four at a very tiny scale, uh, that actually would be useful to people. So for useful things, I expect more things like that for, uh, for fun museum artifacts. Yeah, someday, who, who knows? Like, I think there could be a lot of cool things like that.
## 00:39:07
**中文**
另一位匿名用户——也可能是同一个人——问：你是否承认新模型不如旧模型，而且你正因傲慢和贪婪、无视用户需求而毁掉公司？我相信，对你和你的具体用例而言，它确实不如旧模型；我们希望打造更适合你这一用例的模型。
**English / Original**
Another anonymous, or maybe the same one asks, will you admit that your new model is inferior to the previous one and that you're ruining your company with your arrogance and greed while ignoring users' needs? Um, I, I believe that it is inferior to you for your use case, and we would like to build models that are better for your use case.
## 00:39:23
**中文**
总体而言，我们认为，对大多数用户来说，新模型更好、能力也更强。但我们确实从 4o 升级到 5 的过程中吸取了教训。未来我们会努力做得好得多：既改善连续性，也确保模型对大多数用户都有提升，而不只是对使用 AI 做科学、编程或其他特定工作的人。
**English / Original**
On the whole, uh, we think for most users it's a, it's a better and more capable model, but we definitely have learned things about the four oh to five upgrade, and we will try to do much better in the future, both about better continuity and about making sure that our model gets better for most users, not just sort of people that are using AI for science or coding or for whatever.
## 00:39:44
**中文**
Yu 问：未来会不会推出用于个人联结与自我反思，而不只是商业或教育用途的 ChatGPT 版本？当然会。我们认为这是 AI 非常美好的用途。它对人们生活产生的意义令我们很感动。我们都收到大量用户邮件和反馈，讲述 ChatGPT 如何帮助他们度过个人困境或改善生活——这正是我们存在的意义。
**English / Original**
Uh, Yu asks, will there ever be a version of Chupy team meant for personal connection and reflection, not only business or education? Yeah, for sure. This is, this is, we think this is a wonderful use of ai. Um, we're very touched by how much this has meant to people's lives. We get, all of us get a ton of emails and outreach from users about how chat BT has helped people in difficult personal situations or to live a better life in like, this is what we're here for.
## 00:40:09
**中文**
这和我们所做的任何事情一样重要。我们喜欢听到科学进展，也喜欢听到有人确诊疾病后得到治愈；但个人故事对我们同样极其重要。我们对此非常欣喜，也绝对希望提供这样的服务。
**English / Original**
I mean, this is like, as important as anything that we do. Like we, we love to hear about scientific progress. We love to hear about people that, you know, got diagnosed with the disease and got cured. But the personal stories are incredibly important to us, and we're thrilled about that, and we absolutely wanna offer, uh, such a service, your safety.
## 00:40:29
**中文**
好，这里有两部分，来自 G 和一位匿名用户，都是关于安全路由的问题：你们的安全路由破坏用户信任和工作流程。它推翻我们的选择。你们是否会对所有自愿承担风险的成年用户撤销这种家长式政策，不再把我们当孩子？用户什么时候能控制路由？
**English / Original**
Oh, uh, okay. Two parts to this. Two questions that are about Tide, uh, from G and Anonymous. Your safety, routing breaks, user trust and workflows. By overriding our choices, will you commit to revoking this paternalistic policy for all consenting adult users and stop treating us like children? Uh, when do users get control over routing?
## 00:40:47
**中文**
安全和审查的透明度在哪里？为什么成年人不能自己选择模型？我认为，我们处理模型路由的方式确实算不上最佳。4o 存在一些真实问题；我们观察到，处于脆弱精神状态的人使用 4o 这样的模型时，状况可能进一步恶化。
**English / Original**
Where's the transparency on safety and censorship? Why can't adults pick their own models? Yeah. I, I don't think the way we handled, uh, the model routing was our best thing ever. There, there are some real problems with four oh, um, and we have seen a problem where people are forming people that are in fragile psychiatric situations using a model like four oh can get into a worse one.
## 00:41:07
**中文**
大多数成年用户使用它没有问题；但正如我们提过的，我们有义务保护未成年用户，也有义务保护那些精神状态不适合自主作出真实意愿选择、可能因我们而受伤害的成年用户。随着年龄验证上线，也随着我们能够区分真正处于心理健康危机的用户和其他用户，我们当然希望给予人们更多自由；这是我们的平台原则之一。
**English / Original**
Um, most adult users can use those fine, but we do, as we've mentioned, we have an obligation to protect minor users, and we also, um, have an obligation to protect adult users who are not in a frame of mind where, you know, more reasonably likely that they're choosing what they really want and we're not causing them harm as we build age verification, um, in, and as we are able to differentiate users that are having like a true, um, mental health crisis from users who are not, uh, we of course wanna give people more user freedom, as we mentioned, that's one of our, our platform principles.
## 00:41:41
**中文**
所以，这方面会有改进。此前发布时的沟通也不是我们做得最好的工作。怎样既保护人，又允许成年人谈论困难话题而不觉得受到管制？你要补充吗？是的。我们会在为 ChatGPT 制定的模型规范中列出高层政策和准则。
**English / Original**
Um, so yes, expect improvement there. And I don't think this was our best work in how we communicated the previous rollout, how we strike the right balance between protecting people and allowing adults to speak about difficult things without feeling policed. You wanna say anything there? Um, yeah. So, so, so definitely there is a problem where, um, we, we, we aim to lay out the kind of high, high level, um, policies and, and, and guidelines for the model in, in, in the spec, uh, that we, um, that we develop for, for, for charge GBT.
## 00:42:23
**中文**
但人们可能身处的情境空间极其庞大，到一定程度，划定恰当边界本身就成了一道艰难的智能问题。我们已经看到，推理模型以及投入更多推理来权衡这些微妙问题后，相关指标有所改善。
**English / Original**
But, um, the, the, the, the, the, the, the, the, the space of, of, of, of, of, of, of situations you can find yourself in is, is, is, is, is enormous. And, and, and, um, and at some point, kind of like establishing the right, the right boundaries really becomes a, a tough intelligence problem. Um, and so we are seeing, uh, improved results on this metrics from reasoning models and from expanding, uh, more, more reasoning on, on thinking about this, uh, uh, software, uh, software questions and, and, and trade offs.
## 00:43:05
**中文**
当然，与数学题相比，这类问题的训练也更困难，因此我们正在大力研究。Kate 问：成人模式会在 12 月的什么时候推出？它是否不只涉及 NSFW 内容？现在写作时哪怕有一点冲突，4o 都会触发过滤器。
**English / Original**
Um, of course, this is like a bit more difficult to train for also than, uh, math problems, for example. And, and so this is something that we're, uh, researching quite heavily. Uh, Kate says, when in December will adult mode come, will they have more than just NSFW when writing even slight conflict when writing triggers, filters on four?
## 00:43:25
**中文**
我不知道它会在 12 月的具体哪一天上线。但目标确实是：当你使用 ChatGPT 辅助创意写作时，它在许多类别上应当比以前的模型宽松得多。我们希望如此，也知道用户同样希望如此。
**English / Original**
Oh, uh, I don't know exactly when in December it will ship, but yes, the goal is, um, when you are writing, when you're using OpenAI to help chat GBT to help you with creative writing, um, you sh it should be much more permissive in, in many categories than the previous, uh, models are, again, we, we want this and we know users want this too.
## 00:43:44
**中文**
如果它要成为你的个人工具，就应该帮助你完成手头的事情。每当某件明显不该受限的内容触发过滤器时，我们理解那有多恼人。所以，我们会尝试通过成人模式解决。也许会遇到新问题，但我们希望给人们更多灵活性。
**English / Original**
Um, if this is gonna be your personal tool, it should help you with what you are doing. And every time you hit a content filter for something that is, uh, you know, feels like it shouldn't, we, we understand how annoying that is. So, uh, we are, uh, we're gonna try to fix that with adult mode. There may be new problems that we face, but we want to give people, uh, more flexibility.
## 00:44:07
**中文**
匿名用户问：为什么你们的安全理念要求在用户实际使用哪个模型这件事上误导他们？我还是认为这次上线方式有误。原本目标是让人们继续使用 4o；但在年龄验证尚未完善、而 4o 出现我们认为确有严重危害的行为时，把用户切换到一个不会引发我们在 4o 上遇到的某些心理健康问题的模型。4o 是个很棘手的挑战。
**English / Original**
Uh, anonymous says, why does your idea of safety require aligned to users about what model they're actually using? Uh, again, I think we mis rolled this one out, but the goal here was to let people continue to use four oh. But in the situations where four oh has behavior that we think is actually really harmful before we have all of the age g in that we'd like to kick it, uh, to, to put the user into a model where they're not going to have some of the mental health problems that we faced with four Oh, um, four oh was an interesting challenge.
## 00:44:34
**中文**
一些用户非常喜欢它，但它也给另一些并不希望受伤害的用户造成了伤害。我想，这不会是我们最后一次面对这类模型挑战；我们正在寻找正确的平衡方式。成年人能否重新使用旧版模型，而不被重新路由？
**English / Original**
It's a model that, uh, some users really love and it was a model that was causing some users harm that they really didn't want. And I don't think this is the last time we'll face challenges like this with a model, um, but we are trying to figure out the right, the right way to balance that. Um, will we get into legacy models back for adults without rewriting rewriting it?
## 00:44:53
**中文**
可以。Yum 问：12 月更新会不会正式阐明 OpenAI 对人类与 AI 情感联结的立场，还是会继续通过限制，暗中把这类关系在全球定义为有害？我不太明白“官方立场”该是什么。我们打造这个工具，你可以按自己的方式使用。如果你想与模型建立某种小写字母 r 的“关系”，并从中获得对生活有意义的共情或友谊。
**English / Original**
Yes. Uh, yum. Will the December update officially clarify open AI's position on human AI emotional bonds, uh, or will restrictions continue implicitly defining such connections as harmful worldwide? I, I dunno what it means to have an official position like we build this tool. You can use it the way you want. If, if you want to have like a small r relationship and you're getting something like empathy or friendship that matters to you and your life out of a model.
## 00:45:21
**中文**
对我们而言，模型如实说明自己是什么、不是什么，非常重要。但如果你作为用户确实从这种支持中获得价值，我们认为这很棒。那些从模型提供的情感支持或其他支持中获得价值、实用帮助和更好生活的故事，令我们非常感动。
**English / Original**
Like, it's very important to us that the model faithfully communicate what it is and what it isn't. But if you as the user are finding value in that support, again, we think that's awesome. Like we, we are very touched by the stories of people who find value utility a better life in the emotional support or other kinds of support they get from these models.
## 00:45:43
**中文**
Kylo 问：OpenAI 怎么能让免费版用户使用越来越多的功能？我可以从产品和商业角度回答，但 Jakub，或许你可以谈谈：模型正以多么惊人的速度，用更低价格、更少算力获得更强能力。
**English / Original**
Kylo says, how is OpenAI increasingly allowing so many features for the free version users? I can answer this from a product and business perspective, but jaka, I think it might be useful for you to just talk about the incredible rate at which models are getting more capable for lower prices and less amounts of compute.
## 00:45:59
**中文**
是的，我们看到，随着抵达新的智能前沿，降低达到相应能力的成本也可以非常快。尤其是推理模型：价格很低的模型，只要允许它使用一些额外测试时计算，能力就能大幅提高。
**English / Original**
Um, yeah, we, we are seeing quite a lot of, uh, um, ability to, as we get to like the new frontiers of intelligence to to, to reduce the cost toward that quite, quite quickly. Um, and so yeah, especially, especially with reasoning models, we've seen, uh, that actually quite, uh, quite cheap models when allowed some, some additional test time compute can become much more capable.
## 00:46:30
**中文**
我预计这一趋势会持续。随着我们谈论抵达新前沿、实现研究自动化，我也预计其中许多能力的成本会继续显著下降。我们常谈模型能力增长及其对推动科学进步的作用。
**English / Original**
Um, yeah, and so this is, uh, this is something that I expect we'll continue. And so, um, yeah, as we talk kind of about, about, um, you know, getting to these, um, new frontiers and automating research and so far I I I, I expect that the cost of, of of, of a lot of that will, will keep falling quite a lot too. So yeah, our, our, we talk a lot about the increase in model capability and for, you know, pushing forward science.
## 00:47:03
**中文**
那当然极其重要。但过去几年里，我观察到 AI 最令人惊叹的一点是：某一固定单位智能的价格，每年大约下降 40 倍。最初拥有 GPT-3 时，我们觉得它很酷，但在当时的成本下，使用起来并不容易。
**English / Original**
That's, that's hugely important. One of the most amazing things that I've observed about AI is over the last few years, the sort of price of a particular unit of intelligence has fallen about 40 x per year for the last few years. Um, so when we first, you know, when we first had like GPT-3, we thought it was very cool and it was at this cost, it was kind of hard.
## 00:47:28
**中文**
如今，GPT-3 规模的模型基本可以免费运行，比如直接在手机上。达到 GPT-4 发布时同等智能水平的模型，其成本已经大幅下降。我们预计这一趋势会延续。即便如此，仍需要海量基础设施，因为我们不断发现：成本越低，人们越想使用。
**English / Original**
And like GP three scale models now basically run for free, like on a phone or something. Um, the cost of a, a model that's as smart as GPT-4 at the time we launch it relative now has fallen hugely. And we expect this trend to keep going. Now we still think we need a ton of infrastructure. 'cause what we continue to find is the cheaper we can make it, the more people want to use it.
## 00:47:47
**中文**
我预计需求只会继续增长。我们的目标是让智能成本一降再降，使人们把它用于越来越多的事情，从而让我们继续免费提供许多功能。与此同时，那些真的想投入巨资把 AI 推向极限——例如治愈疾病、设计更好的火箭等——的人，也会投入极其庞大的金额。
**English / Original**
And I expect that only to increase. But our goal is to drive the, you know, cost of intelligence down and down and down and have people use it for more and more things that will allow us to continue to offer lots of features for free. Um, but that will also mean I think that people who really wanna spend a lot on pushing AI to the limit to cure a disease or figure out how to build a better rocket or whatever, we'll spend a huge amount.
## 00:48:11
**中文**
只要商业模式还能基本成立，我们就承诺继续把能提供的最佳技术放进免费层。今后大家会看到我们在这方面做得更多。匿名用户问：年龄验证启动后，能否让用户选择退出安全路由，并签署免责声明，免除 OpenAI 的一切责任？
**English / Original**
Um, we are committed to continuing to put the best technology we can, uh, as long as we can make the business model even sort of work into the free tier. And you should expect a lot more from us there over time. Uh, okay, anonymous asks, will an age will an age verification start that allows users to opt out of the safety router a waiver that could be signed, releasing open air from any liability.
## 00:48:36
**中文**
即使你签了免责声明，我们也不会去做相当于贩卖海洛因之类的事。但遵循“把成年用户当成年人”的原则，通过年龄验证后，你会获得相当大的灵活性。我们认为这很重要，而且显然也引起了提问者的共鸣。
**English / Original**
Um, we're not gonna like do, again, we're not gonna do the equivalent of selling heroin or whatever, uh, even if you sign a liability. But yes, on the, on the principle of treat adult users like adults, if you're age verified, you will get quite a lot of, of flexibility. We think that's important and clearly it resonates with the people asking these questions.
## 00:48:55
**中文**
另一位匿名用户问：ChatGPT 是不是 AI 时代的 Ask Jeeves？我们当然希望不是，也不认为它会是。好，只剩十分钟了；有些问题和刚才已经谈到的内容重复，我们会跳过一部分，争取在今后的问答中回答更多。
**English / Original**
Um, anonymous also asks, is Chacha bt the asks Gs of ai? We sure hope not. We don't think it will be. Uh, okay. Since we only have 10 minutes left, we're gonna, um, and some of these touch other things that we've, uh, we've already touched on. We're gonna skip down through some of the same questions and try to get to more, uh, in future question sessions.
## 00:49:18
**中文**
如果今天没能全部答完，我们以后还可以多办几次。Just Ask the Macintosh 问：就像 Macintosh 是 iPhone 的前身，你们认为 ChatGPT 就是 OpenAI 的最终产品，还是它只是某种更伟大、真正重塑世界之物的前身？作为一家研究实验室，我们最初并不是为了制造聊天机器人，尽管后来逐渐认识到，这个产品与我们的整体使命高度一致。
**English / Original**
We can, um, do more of these if we don't get to everything here. Um, just as the Macintosh no asks, just how the Macintosh was the precursor to the iPhone, do you see chat GBT as the open AI product or do you see it as a precursor to something much greater that truly reshapes the world? Um, So I would say like as a research lab, um, when we haven't set out to build a a, a chatbot originally, although I think we've since come to appreciate how aligned this this product is with, with our overall mission.
## 00:49:55
**中文**
我们当然预计 ChatGPT 会不断进步，继续成为人们与日益先进的 AI 互动的方式。但正如刚才讨论的，我们也预计 AI 系统最终会创造有价值的成果，真正推动科学进步。
**English / Original**
Um, and we of course expect chat GPT to continue to become better and, and, and, and be this way, um, for people to interact with increasingly advanced ai. Um, but we do anticipate that, uh, eventually AI systems will be capable of, um, creating valuable artifacts of actually pushing scientific progress forward as we were discussing.
## 00:50:21
**中文**
我相信，那才会是 AI 真正持久的遗产。聊天界面很出色，但不会是唯一界面；人们使用这些系统的方式会随时间发生巨大变化。想想 Jakub 刚才提到的五秒、五分钟和五小时任务；若是一个人类需要五年、甚至五百年才能完成的任务，我们甚至很难理解那意味着什么。
**English / Original**
And, um, I believe that will be the, the real lasting legacy of ai. I, I think the chat interface is a great interface. It won't be the only interface, but the way that people use these systems, um, will change hugely over time. If you think about what Jakob shared earlier of the five second, five minute, five hour tasks, um, if you think about a five year or five century task that would take something that would take humans, it's hard to even think about what that means.
## 00:50:51
**中文**
这时，你大概需要一种不同的产品体验。我还认为，人们可能希望它更像一个融入环境、始终在场的伙伴。现在你可以向 ChatGPT 提问，让它替你做事；但如果有一种服务可以观察你的生活，在你需要时主动提供帮助，那会非常好。
**English / Original**
But probably you want a different kind of, of product experience. I also think you probably will want this to feel more like a sort of ambient always present companion. Like right now you can ask Che BT something, it can do something for you, but it'd be really nice to have a service that was sort of just observing your life and proactively helping you when you needed it.
## 00:51:11
**中文**
它还能帮助你产生更好的想法。我认为我们可以大力向这个方向推进。Neil 问：我喜欢 GPT-4.5，它是目前市场上最好的写作模型，也是我订阅 Pro 的主要原因。能否明确说明它的未来？我们认为，很快就会有比 4.5 好得多的模型。
**English / Original**
And, you know, helping you come up with better ideas and just, I, I think we can probably push very hard in, in that direction. Um, Neil asks, I love GPT-4 0.5. It's by far the best on the market for writing, and it's the main reason I pay for Pro. Could we get some clarity on its future, please? Um, we think we're gonna have models that are much better than 4.5 very soon.
## 00:51:32
**中文**
尤其在写作方面会好得多。在出现写作能力实现巨大飞跃的模型之前，我们计划保留 4.5。不过，我们已经不认为 4.5 有那么好了，希望提供远为出色的产品。当然，我们绝没有停止这条研究路线；我们预计，把它与其他正在研究的成果结合，会得到在所有维度上都显著优于 4.5 的模型。
**English / Original**
Uh, and for writing much, much better. We plan to keep it around until we have a model that is a huge step forward in, uh, writing. But, you know, we'd like to, we don't think 4.5 is that good anymore. We'd like to offer something much, much better. But yeah, we are definitely not done with that direction of research. Uh, and yeah, we, we expect combining, combining that with, with, um, other things working on will get wheel files are dramatically better than 4.5 on all axis.
## 00:51:59
**中文**
对于在写作等任务上显著超越 4.5 的模型，你们能否透露大概时间？这种提升最终能走多远？我预计明年肯定会出现。Lars 问：Windows 版 ChatGPT Atlas 什么时候推出？
**English / Original**
Do you have, uh, any sense of timing to share about when you think we have a model that is dramatically better than 4.5 on, on this kind of task, like writing and also anything about like how far that's gonna go? Um, next year I think is definitely what I expect. Um, when is Chacha PT Atlas for Windows come in, asks Lars?
## 00:52:22
**中文**
我不知道准确时间，猜测是几个月后。这绝对是我们想做的事情。更广泛地说，我们希望打造浏览器和新设备等体验，让人们随身携带 AI，使它成为融入环境、始终提供帮助的助手，而不是只能在你发出请求后回应的工具。
**English / Original**
Uh, I don't know, an exact timeframe. Some number of months I would guess. Uh, it's, it's definitely something we wanna do. And more generally, this idea that we can build experiences like browsers and new devices that let you take AI with you, that get towards this sort of ambient, always helpful, uh, assistant rather than something you just query in response.
## 00:52:42
**中文**
这将是我们需要重点推进的重要方向。你们会公开 170 位专家的意见文件吗？这样，我们可以了解模型行为新依据，提高透明度。我会问 Fidji（首席营销官）希望如何处理。我不清楚具体能分享什么，但我认为应当在这方面有所行动。
**English / Original**
Uh, this will be a very, this will be a very important direction for us to, to push more on. Will you disclose the documents with the opinions of the 170 experts so that we'll have some transparency regarding the new justifications from model behavior. Um, I will ask Fiji cmo, uh, how she'd like to handle that, but I think we could, I don't know exactly what we'll be able to share, but I think we should do something there.
## 00:53:03
**中文**
提高这方面的透明度是好事。匿名用户说：我从第二个月起就是 Pro 用户。作为研究人员和小说作者，我觉得 GPT 帮我思考得更清晰——抱歉，问题丢了，这是个很好的问题，我再找一下——更清晰，却不再更自由。想象力是否成了优化过程的牺牲品？
**English / Original**
And I, I think more, um, more transparency there is a good thing. Anonymous says, I've been a pro user since month two. As a researcher and fiction writer, I feel GPT helps me think clearer. Uh, I lost the question. Um, sorry, that was a really good question. Lemme try to find it again. Uh, clearer but not freer. Has imagination become an optimization casualty?
## 00:53:30
**中文**
你怎么看？对于当前系统，这种情况确实可能存在。例如比较 4.5 和 o3，我预计它们之间会有取舍。我认为，这是我们摸索这些技术时必然出现的一些过渡性现象。
**English / Original**
What do you think? Um, it is, I think it is definitely possible for our current systems that, uh, you know, like I think if you compare a model like 4.5 to a model like oh three, I would expect that like there will be trade offs there. I I think there are definitely like transitory as we like figure out our way around these technologies.
## 00:53:56
**中文**
我预计它会改善。我还认为，这会产生群体规模的影响。我注意到一件奇怪的事：现实生活中，有人说话时会带上 ChatGPT 某些特有的表达习惯。未来还会出现其他类似现象，人类与技术将以我们无法完全预测的方式共同演化。
**English / Original**
And so again, like, I expect this will, this will get better. Yeah, I, I, I think there are going to be population scale. Like one, one of the sort of strange things I've noticed is people in real life talking to, like Chachi BTEs, uh, where they sort of use some of the quirks of, of things Chachi BT says. And I, I think there will be other things like this where there's like this co-evolution of people and the technology in ways we can't totally predict.
## 00:54:27
**中文**
但我预计，随着时间推移，人们会比今天更有能力、更有创造力，思考得更开阔、更广泛。我们已经看到这样的例子：有人说，“过去我绝不可能同时在脑中容纳这些内容，也不可能产生这个想法。”
**English / Original**
But my expectation is over time people are much more capable, much more creative, think much more expansively, and much more broadly than they do today. Um, and we start, we certainly see examples of this, uh, where people are just like, I never would've been able to keep this in my head. I never would've been able to have this idea.
## 00:54:45
**中文**
也有人说，“我把思考外包了，它叫我做什么，我就做什么。”显然，前一种情况比后一种更令我们兴奋。有人问：你们为何打造具有情商的模型，却又批评人们出于无障碍需求，使用它们处理情绪和心理健康问题？
**English / Original**
And then we hear other examples where people say, you know, I've outsourced my thinking and I just do what this thing tells me. And obviously we're much more excited about the former than the latter. Um, can you help us understand why you build emotionally intelligent models and then criticize people for using, using it for accessibility reasons when it comes to processing emotions and mental health?
## 00:55:04
**中文**
我们仍然认为那是好事，也希望看到这种使用方式。令我们高兴的是，同一个能提供这类帮助的模型，也可能被用来强化心理脆弱用户的妄想。我们希望使用者是在有意识地使用，模型没有欺骗用户、没有歪曲自己是什么或不是什么。
**English / Original**
Uh, again, we think that's a good thing. We want that, uh, we're, we're happy about that there. The same model that can do that can also be used to encourage delusions and mentally fragile users. And what we want is people who are using these models intentionally, the model's not deceiving the user about what it is and what it isn't.
## 00:55:23
**中文**
模型应当提供帮助，帮助用户实现自己的目标。我们希望这类情况更多，而任何像是模型“诱骗”用户的情况更少——暂且借用这个不够科学的说法。我们完全理解大家的挫败：每当试图阻止有害行为时，也会连带阻止一些完全正当的用途。
**English / Original**
The model's being helpful, the model's helping the user accomplish their goals. We want more of that and less of anything that would feel like the model tricking a user for lack of a more scientific word. Um, I totally get, we totally get the, the frustration here like to, you know, whenever you're trying to stop something that is causing harm, you stop some perfectly good use as well.
## 00:55:43
**中文**
但请理解我们的出发点：我们希望为知情的成年人提供真正有价值的服务，同时避免在此过程中无意伤害并不希望受伤害的人。好，只剩几分钟了，看看是否还有方向完全不同、值得回答的问题。
**English / Original**
Um, but please understand the place we're coming from here is trying to provide a service to adults that are aware of it and that are getting real value from it and not cause unintended harm to people who, who don't want that along along the way. Um, alright, given that we have just a couple of minutes left, let's see if there's any questions in very other directions that we should try to get to.
## 00:56:14
**中文**
Razzi 问：你们认为 AI 何时会导致大规模失业？我认为，我们已经到了这样一个阶段：阻碍现有模型完成许多知识型工作的主要差距，可能更多在于集成和界面，而不是原始智力能力。
**English / Original**
Okay. Uh, when do you think massive jobs loss will happen due to AI from Razzi? So I think, I think already we are at a point where I think a lot of the, um, um, the gap that stops that, that stops present models from like being able to perform a lot of, um, intellectual jobs. Uh, it's more about, um, integrations and interfaces, uh, than maybe raw intellectual capability.
## 00:56:57
**中文**
因此，我们必须认真考虑未来几年许多工作的自动化。这是需要全社会共同讨论和思考的大问题：哪些新工作会取代原有工作？我们又会共同投入哪些新的追求？下面这个问题来自我本人，不是直播观众。
**English / Original**
Um, and so I think we definitely have to, Uh, Think about that, um, uh, think about automation of, of, of a lot of jobs as something that, um, will be happening over the next years. And I think this is a big, uh, thing for, for, for, for, for us to talk collectively think about like what are, uh, what are the jobs that will, will replace those and uh, and what are the kind of new pursuits that will, that, that will all, uh, engage on. This Is a question from me, not from, uh, not from the live stream.
## 00:57:39
**中文**
你认为未来的意义感会是什么样？未来的工作会是什么样？当 AI 把当前许多事情自动化之后，我们会如何获得满足感、如何度过时间？我想，这是一个相当哲学性的问题。
**English / Original**
What, what do you think meaning will look like? Uh, what do you think the jobs of the future will look like? How do you think when AI automates a lot of the current things, like how do you think we'll derive our fulfillment and spend our time? Um, I, I expect, um, yeah. Well, I, I think this is a, this is a quite, uh, um, philosophical question.
## 00:58:09
**中文**
它可能走向许多方向，但有些事情我有所预期。高层目标设定——也就是选择我们要追求什么——仍将属于人类；我认为，许多人会从中获得意义。
**English / Original**
I, I, I think, I think I, I think can go in many directions, but some things I expect, I think the high level goal setting, right? Like picking what pursuits, uh, um, we're, we're chasing, that is something that will remain human and, and uh, and I think that that is something that a lot of people will derive meaning from.
## 00:58:37
**中文**
此外，人们将有能力更深入地理解世界；世界中会出现种类惊人的新知识、新娱乐以及新的智能。我认为，这也会为人们带来大量意义和满足感。
**English / Original**
Um, I think also just the ability to understand so much more about the world, the wide, the incredible variety of, of, of new knowledge and new also enter entertainment, but also, uh, um, just intelligence that, that, that, that that, that, that will be in the world, I think, I think will, uh, will provide quite a lot of, um, me and fulfillment for people.
## 00:59:04
**中文**
好，最后两分钟，快速问答。Shindi 问：GPT-6 什么时候发布？某种意义上，这也许更该由你回答。过去的 GPT-3、GPT-4 等模型，我们一直把新模型的训练方式与最终发布的产品紧密联系在一起；到了 GPT-5，情况有所变化。
**English / Original**
Okay. Rapid fire. Two minutes. Uh, Shindi says when G PT six. Um, I think, I think, I think, uh, in some ways maybe that's more of a question for you, uh, in that like, um, I think I think we've, we've with GPT with g PT five, uh, we have, uh, with previous models, right? Like, uh, like G PT four, GP three, like, we've kind of like kept very tight, uh, connection of like how, how we're training new models.
## 00:59:34
**中文**
也就是说，研究模型和我们发布什么产品过去绑得很紧。正如刚才所说，目前在集成方面还有很多工作。例如，GPT-5 是我们第一次真正把推理模型作为旗舰、作为主要旗舰模型。
**English / Original**
So like what are the products that we ship? And as I was just saying, right, like, I think right now there's a lot to do on the kind of integration side and on the, uh, so for example, with G PT five, it's the kind of the, the first time we really bring reasoning models as kind of our, our flagship, our our our main flagship model.
## 00:59:52
**中文**
因此，我们不再把这些发布和产品与研究计划如此紧密地捆绑。我也不知道我们究竟何时会把某个模型称为 GPT-6；但我们传递出的明确信息是：从现在起六个月后——很可能更早——我们预计模型能力会取得巨大的进步。
**English / Original**
Um, and so, uh, we're, we're not coupling like these releases and these products as tightly to our research program anymore. Yeah. Um, I don't know either exactly when we'll call it that, but I think a clear message from us is, say six months from now, probably sooner we expect to have huge steps forward in model capability.
## 01:00:11
**中文**
Felix 问：IPO 是否仍在计划中？届时组织架构会怎样？是否已有增资规则？我们没有具体计划，也没有确定的时间点；但考虑到未来的资金需求和公司规模，公平地说，这很可能是我们最终会走的道路。
**English / Original**
Felix asks, is an IPO still planned? And how would the structure then look like? Are there rules in place for increasing capital? Um, we don't have like specific plans or this is exactly when it's going to happen, but I think it, it's fair to say it is the most likely path for us given the capital needs that we'll have and sort of the size of the company.
## 01:00:28
**中文**
不过，IPO 目前并不是我们最优先考虑的事情。Alec 问：你提到对 1.4 万亿美元投资规模感到可承受。长期支撑它需要多少收入？最大的收入来源会是什么？不可能只靠用户订阅。最终，我们的年收入需要达到数千亿美元；而目前正沿着一条相当陡峭的增长曲线前进。
**English / Original**
Um, but you know, that's not like a top of mind thing for us right now. Alec asks, you mentioned being comfortable with the 1.4 trillion of investment. What level of revenues would you need to support this over time? Will it be the largest revenue driver? It can't just be a per user subscription. Um, you know, eventually we need to get to hundreds of billions of a year in revenue and, and we're on a pretty steep curve towards that.
## 01:00:49
**中文**
我预计企业业务会成为巨大的收入来源，但消费者业务同样如此。它不会只有订阅；还会有新产品、设备，以及许多其他业务。这还完全没有计入 AI 在科学上取得发现后所代表的意义，以及由此打开的全部收入可能性。
**English / Original**
I expect enterprise to be a huge revenue driver for us, but I think consumer really will be too, and it won't just be the subscription subscription, but we'll have new products, devices, tons of other things there as, as well. Uh, and this says nothing about like, what it would really mean to have AI discovery in science and all of the revenue, uh, um, possibilities that, that would unlock.
## 01:01:10
**中文**
随着我们看到更多这类成果，也会增加基础设施支出。好，时间到了。非常感谢大家参加并提出问题。我们会从这种形式中吸取经验、持续迭代，并继续举办这类问答。非常感谢。
**English / Original**
And as we see more of that, we will, uh, we will increase spend on infrastructure. Okay. Um, we are out of time. Thank you all very much for joining us and the questions, and we will try to learn from this format and iterate on it and keep doing, uh, these sorts of q and a. Thank you very much. Okay.