Sam Altman 谈 OpenAI 董事会风波与通往 AGI 之路上的权力斗争

我认为算力将成为未来的货币，它可能会成为世界上最珍贵的商品。我预计到本十年末，甚至可能更早，我们将拥有相当强大的系统，让我们看了会说：“哇，这真是太了不起了。”通往 AGI 的道路应该是一场巨大的权力斗争，我预计情况会如此。谁先造出 AGI，谁就能获得巨大的权力。你相信自己能驾驭这么大的权力吗？

I think compute is going to be the currency of the future. I think it will be maybe the most precious commodity in the world. I expect that by the end of this decade, and possibly somewhat sooner than that, we will have quite capable systems that we look at and say, 'Wow, that's really remarkable.' The road to AGI should be a giant power struggle. I expect that to be the case. Whoever builds AGI first gets a lot of power. Do you trust yourself with that much power?

请带我回顾一下 OpenAI 的董事会风波，这件事从 11 月 16 日星期四（对你来说可能是 11 月 17 日星期五）开始。

Take me through the OpenAI board saga that started on Thursday, November 16th, maybe Friday, November 17th for you.

那绝对是我职业生涯中最痛苦、最混乱、最丢脸、最令人沮丧的经历，还有很多其他负面感受。但其中也有好的部分，我希望当时不是处于那种肾上腺素飙升的状态，以至于没能停下来好好体会。我翻到了自己当时发的一条推文，那感觉就像参加自己的葬礼，看着人们说尽你的好话，还有来自我爱和关心的人难以置信的支持。那真的很美好。那个周末，除了一个很大的例外，我感受到的是大量的爱和很少的恨。尽管我感觉完全不知道发生了什么、接下来会怎样，感觉非常糟糕。有好几次我都觉得这将成为 AI 安全领域有史以来最糟糕的事件之一。不过，我也很高兴这件事发生得相对较早。我原本就认为，在 OpenAI 成立到我们创造出 AGI 之间的某个时刻，一定会发生一些疯狂而爆炸性的事情。但未来可能还会有更疯狂、更爆炸性的事情发生。我认为这件事仍然帮助我们建立了一些韧性，为未来更多的挑战做好了准备。

That was definitely the most painful professional experience of my life, and chaotic, and shameful, and upsetting, and a bunch of other negative things. There were great things about it too, and I wish it had not been in such an adrenaline rush that I wasn't able to stop and appreciate them at the time. But I came across this old tweet of mine from that time period, which was like going to your own eulogy, watching people say all these great things about you, and just unbelievable support from people I love and care about. That was really nice. That whole weekend, with one big exception, I felt a great deal of love and very little hate. Even though it felt like I had no idea what was happening and what was going to happen, and it felt really bad. There were definitely times I thought it was going to be one of the worst things to ever happen for AI safety. Well, I also think I'm happy that it happened relatively early. I thought at some point between when OpenAI started and when we created AGI, there was going to be something crazy and explosive that happened. But there may be more crazy and explosive things still to happen. It still, I think, helped us build up some resilience and be ready for more challenges in the future.

你之前就预感到会经历某种权力斗争吗？

You had a sense that you would experience some kind of power struggle?

通往 AGI 的道路应该是一场巨大的权力斗争。世界应该——不，不应该说“应该”——我预计情况会如此。所以你必须经历这些，就像你说的，尽可能多地迭代，弄清楚如何构建董事会结构、如何组织、如何选择合作伙伴、如何沟通这一切，以尽可能缓和权力斗争。是的，平息它。但现在，这感觉像是过去的事了，非常不愉快、非常困难和痛苦，但我们又回到了工作中。事情如此繁忙和紧张，以至于我没有花太多时间去想它。在那之后有一段时间——大概是一个月，也许是 45 天——我处于一种神游状态，整天浑浑噩噩。我完全不在状态。我在个人心理层面上感到非常低落。真的很痛苦。而且很难在那种情况下继续运营 OpenAI。我只想躲进一个山洞里恢复一段时间。但现在，我们又回到了为使命而工作的状态。

The road to AGI should be a giant power struggle. The world should—well, not should—I expect that to be the case. And so you have to go through that, like you said, iterate as often as possible in figuring out how to have a board structure, how to have organization, how to have the kind of people that you're working with, how to communicate all that in order to de-escalate the power struggle as much as possible. Yeah, pacify it. But at this point, it feels like something that was in the past, really unpleasant and really difficult and painful, but we're back to work. Things are so busy and so intense that I don't spend a lot of time thinking about it. There was a time after—there was like this fugue state for maybe the month after, maybe 45 days after that—I was just sort of drifting through the days. I was so out of it. I was feeling so down, just on a personal psychological level. Yeah, really painful. And hard to have to keep running OpenAI in the middle of that. I just wanted to crawl into a cave and recover for a while. But now it's like we're just back to working on the mission.

回顾并反思董事会结构、权力动态、公司运营方式、研究与产品开发及资金之间的紧张关系，所有这些仍然很有价值。这样，未来任何极有可能打造 AGI 的人都能以更有条理、更少戏剧性的方式去做。所以，无论是你作为领导者的个人心理层面，还是董事会结构这些乱七八糟的东西，都有其价值。

It's still useful to go back there and reflect on board structures, on power dynamics, on how companies are run, the tension between research and product development and money, all this kind of stuff, so that whoever has a very high potential of building AGI would do so in a slightly more organized, less dramatic way in the future. So there's value there to go through both the personal psychological aspects of you as a leader and also just the board structure and all this kind of messy stuff.

我确实学到了很多关于结构、激励机制以及我们对董事会的要求。从某种意义上说，这件事现在发生是有价值的。这可能不是 OpenAI 最后一次高压时刻，但确实是一次非常高压的时刻。公司差点就毁了。我们考虑了很多其他为了 AGI 必须做对的事情，但思考如何建立一个有韧性的组织，如何建立一个能够承受世界上巨大压力的结构——随着我们越来越接近目标，我预计压力会越来越大——我认为这非常重要。

Definitely learned a lot about structure and incentives and what we need out of a board. And I think it is valuable that this happened now in some sense. I think this is probably not the last high-stress moment of OpenAI, but it was quite a high-stress moment. The company very nearly got destroyed. And we think a lot about many of the other things we've got to get right for AGI, but thinking about how to build a resilient org and how to build a structure that will stand up to a lot of pressure in the world—which I expect more and more as we get closer—I think that's super important.

你了解董事会的审议过程有多深入和严谨吗？比如，你能透露一下这种情况中涉及的人际动态吗？是不是只是几次谈话，然后突然就升级到“我们为什么不炒掉 Sam”这种地步？

Do you have a sense of how deep and rigorous the deliberation process by the board was? Like, can you shine some light on just the human dynamics involved in situations like this? Was it just a few conversations and all of a sudden it escalates and 'why don't we fire Sam' kind of thing?

我认为董事会成员总体上都是出于好意。而且我相信，在人们感到时间压力等紧张情况下，人们做出次优决策是可以理解的。我认为 OpenAI 面临的挑战之一就是，我们必须有一个能够在压力下运作良好的董事会和团队。

I think the board members were well-meaning people on the whole. And I believe that in stressful situations where people feel time pressure or whatever, people understandably make suboptimal decisions. And I think one of the challenges for OpenAI will be we're going to have to have a board and a team that are good at operating under pressure.

你认为董事会权力过大吗？

Do you think the board had too much power?

我认为董事会本来就该拥有很大的权力。但我们确实看到的一点是，在大多数公司结构中，董事会通常要对股东负责。而在我们的结构中，我们可能应该多考虑一点的是，非营利组织的董事会——除非你制定其他规则——拥有相当大的权力。他们实际上只对自己负责，不对任何人负责。这在某些方面是好的，但我们真正希望的是 OpenAI 的董事会能够尽可能地向整个世界负责。

I think boards are supposed to have a lot of power. But one of the things that we did see is that in most corporate structures, boards are usually answerable to shareholders. In this case, and I think one of the things with our structure that we maybe should have thought about more than we did, is that the board of a nonprofit has—unless you put other rules in place—quite a lot of power. They don't really answer to anyone but themselves. And there are ways in which that's good, but what we'd really like is for the board of OpenAI to answer to the world as a whole, as much as that's a practical thing.

新董事会已经宣布了。是的，一开始是一个较小的新董事会，现在有了新的最终董事会？还不是最终董事会。我们已经增加了一些成员，还会再增加。新董事会解决了旧董事会可能存在的哪些问题？

There's a new board announced. Yeah, I guess a new smaller board at first, and now there's a new final board? Not a final board yet. We've added some, will add more. What is fixed in the new one that was perhaps broken in the previous one?

旧董事会在大约一年的时间里规模逐渐缩小。最初是九人，后来减到六人，然后我们就无法就增加谁达成一致。而且，我认为旧董事会中缺乏有经验的董事会成员。OpenAI 的许多新董事会成员拥有更丰富的董事会经验。我认为这会有所帮助。

The old board sort of got smaller over the course of about a year. It was nine, and then it went down to six, and then we couldn't agree on who to add. And the board also, I think, didn't have a lot of experienced board members. A lot of the new board members at OpenAI have more experience as board members. I think that'll help.

新董事会的一些成员受到了批评。例如，我听到很多人批评拉里·萨默斯的加入。董事会的遴选过程是怎样的？都涉及哪些方面？

It's been criticized—some of the people added to the board. I heard a lot of people criticizing the addition of Larry Summers, for example. What's the process of selecting the board like? What's involved in that?

所以布雷特和拉里是在那个非常紧张的周末，在情急之下决定的。就是这样。

So Bret and Larry were kind of decided in the heat of the moment over this very tense weekend. And that was that.

那个周末就像坐过山车一样，起起伏伏。我们当时在努力敲定新的董事会成员，既要让这里的执行团队觉得合理，也要让旧董事会成员觉得合理。Larry 其实是旧董事会成员提名的候选人之一。Brett 我甚至在那个周末之前就提过，但他很忙，不想干。我们当时真的需要帮助。我们还聊了很多其他人。但我觉得，如果我要回来，我需要新的董事会成员。我认为我无法再以同样的配置和旧董事会合作了，不过我们后来决定——我也很感激——Adam 会留下。但我们想要尝试不同的配置，最终决定组成一个三人董事会，并且必须在短时间内找到两位新成员。所以这些决定说实话没有经过严格的流程。就像在战场上做决定一样，你没有时间设计一个严谨的流程。对于之后的董事会成员，以及未来新增的成员，我们有一些标准，我们认为董事会需要具备不同的专业知识。不像招聘高管，你需要他们做好一个角色，董事会需要承担治理和深思熟虑的整个角色。Brett 有句话我特别喜欢，他说我们希望以“批次”的方式招聘董事会成员，而不是一个一个地招。考虑一群能带来非营利组织经验、公司运营经验、以及良好的法律和治理经验的人。这就是我们努力优化的方向。那么技术敏锐度对单个董事会成员重要吗？不是每个成员都需要，但肯定有些人需要。这是董事会职责的一部分。

That weekend was like a real roller coaster. It's like a lot of ups and downs. We were trying to agree on new board members that both the executive team here and the old board members felt would be reasonable. Larry was actually one of their suggestions, the old board members. Brett, I think I had even previous to that weekend suggested, but he was busy and didn't want to do it. And then we really needed help. We talked about a lot of other people too. But I felt like if I was going to come back, I needed new board members. I didn't think I could work with the old board again in the same configuration, although we then decided, and I'm grateful that Adam would stay. But we wanted to get to various configurations, decided we wanted to get to a board of three, and had to find two new board members over the course of a short period of time. So those were decided honestly without a rigorous process. You kind of do that on the battlefield; you don't have time to design a rigorous process. For new board members since, and new board members will add going forward, we have some criteria that we think are important for the board to have different expertise. Unlike hiring an executive where you need them to do one role well, the board needs to do a whole role of governance and thoughtfulness. One thing that Brett says which I really like is that we want to hire board members in slates, not as individuals one at a time. Thinking about a group of people that will bring nonprofit expertise, expertise at running companies, good legal and governance expertise. That's what we've tried to optimize for. Is technical savvy important for the individual board members? Not for every board member, but for certainly some. That's part of what the board needs to do.

所以人们不了解 OpenAI 的有趣之处——当然我也不了解——就是经营企业的所有细节。当人们想到董事会，考虑到那场风波，他们会想到你，他们会想：如果你实现了 AGI，或者你开发并部署了这些极具影响力的产品，你和董事会的对话会是什么样？他们大概在想，好吧，在这种情况下，什么样的团队才适合进行 deliberation？

So the interesting thing that people don't understand about OpenAI, certainly I don't, is all the details of running the business. When they think about the board given the drama, they think about you, they think about if you reach AGI or you reach some of these incredibly impactful products and you build them and deploy them, what's the conversation with the board like? And they kind of think, all right, what's the right squad to have in that kind of situation to deliberate?

听着，我认为你肯定需要一些技术专家在那里。然后你需要一些这样的人：我们怎样才能以最能帮助全世界人民的方式部署它？以及持有非常不同视角的人。我认为你或我可能犯的一个错误是认为只有技术理解才重要。这绝对是你希望董事会进行的对话的一部分，但还有更多关于它将如何影响社会和人们生活的内容，你确实希望这些也能在董事会中得到体现。

Look, I think you definitely need some technical experts there. And then you need some people who are like, how can we deploy this in a way that will help people in the world the most? And people who have a very different perspective. I think a mistake that you or I might make is to think that only the technical understanding matters. That's definitely part of the conversation you want that board to have, but there's a lot more about how that's going to impact society and people's lives that you really want represented in there too.

那你只是看人们的过往记录，还是只是进行对话？

And you're just kind of looking at the track record of people or you're just having conversations?

过往记录很重要。你当然要进行很多对话。但有些角色我基本上完全忽略过往记录，只看斜率，忽略 Y 轴截距。

Track record is a big deal. You of course have a lot of conversations. But there's some roles where I kind of totally ignore track record and just look at slope, kind of ignore the Y intercept.

谢谢你为听众把它数学化。对于董事会成员来说，我更关心 Y 轴截距。我认为过往记录有很深的内涵。而且经验有时很难替代。

Thank you for making it mathematical for the audience. For a board member, like I do care much more about the Y intercept. I think there is something deep to say about track record there. And experience sometimes very hard to replace.

你是试图用多项式函数还是指数函数来拟合过往记录？这个类比没那么深入。

Do you try to fit a polynomial function or exponential one to the track record? That analogy doesn't carry that far.

好吧，你提到了那个周末的一些低谷。对你来说，心理上的低谷是什么？你有没有考虑过跑到亚马逊丛林，带着 IA 永远消失？

Alright, you mentioned some of the low points that weekend. What were some of the low points psychologically for you? Did you consider going to the Amazon jungle and just taking IA and disappearing forever?

我的意思是，太多了。那是一段非常糟糕的时期。也有很棒的高光时刻，比如我的手机不停地收到来自每天一起工作的人、以及十年没联系过的人的暖心消息。我没能像应该的那样充分感激这些，因为我正处在战火之中，但那真的很美好。但总的来说，那是一个非常痛苦的周末。而且就像一场在公众面前进行的战斗，程度惊人，这让我极度疲惫，远超我的预期。我认为争斗通常都很累人，但这次尤其如此。董事会周五下午做了这件事。我真的得不到多少答案，但我也只是觉得，好吧，董事会有权这么做，所以我要花点时间想想我想做什么，但我会试着在这里找到塞翁失马焉知非福。我当时想，嗯，我在 OpenAI 的当前工作是管理一家规模相当的公司，而我一直以来最喜欢的就是和研究人员一起工作。我就想，是啊，我可以去做一个非常专注的 AI 研究项目，我对此感到兴奋。我当时甚至没想过这一切会被推翻。那是周五下午。所以我很快就接受了这个结局。我经历了一小段困惑和愤怒，但很快，到周五晚上，我就在和别人讨论下一步要做什么了，而且我对此感到兴奋。我想是在周五晚上，我第一次听到这里的执行团队说，嘿，我们要抗争这件事，我们觉得不管怎样。然后我上床睡觉，心里还是想着，好吧，很兴奋，继续前进。

I mean, there's so many. It was a very bad period of time. There were great high points too, like my phone was just non-stop blowing up with nice messages from people I work with every day, people I hadn't talked to in a decade. I didn't get to appreciate that as much as I should have because I was just in the middle of this firefight, but that was really nice. But on the whole, it was a very painful weekend. And also just like a battle fought in public to a surprising degree, and that was extremely exhausting to me, much more than I expected. I think fights are generally exhausting, but this one really was. The board did this Friday afternoon. I really couldn't get much in the way of answers, but I also was just like, well the board gets to do this, and so I'm going to think for a little bit about what I want to do, but I'll try to find the blessing in disguise here. I was like, well, my current job at OpenAI was to run a decently sized company at this point, and the thing I had always liked the most was just getting to work with the researchers. I was like, yeah, I can just go do a very focused AI research effort, and I got excited about that. It didn't even occur to me at the time that this was all going to get undone. This was like Friday afternoon. So you've accepted the death very quickly. I went through a little period of confusion and rage, but very quickly, by Friday night, I was talking to people about what was going to be next, and I was excited about that. I think it was Friday night evening for the first time that I heard from the exec team here, which is like, hey, we're going to fight this, and we think whatever. And then I went to bed just still being like, okay, excited, onward.

你睡得着吗？

Were you able to sleep?

睡得不多。这是其中一件奇怪的事。在那四天半的时间里，我睡得不多，吃得不多，却仍然有惊人的精力。你在更多时间里了解了肾上腺素的神奇之处。所以某种程度上你接受了这个“孩子”——OpenAI——的死亡。我对新事物感到兴奋。我只是觉得，好吧，这很疯狂，但无所谓。这是一个非常好的应对机制。然后周六早上，两位董事会成员打电话说，嘿，我们造成了不稳定，我们不是故意要破坏局面，我们不想在这里摧毁大量价值。我们能谈谈让你回来吗？我立刻不想那么做。但我想了想，我确实在乎这里的人、合作伙伴、股东，所有的一切。我爱这家公司。所以我想了想，我说，好吧，但这是我需要的东西。然后整个周末最痛苦的时刻来了。我一直在想，也一直被告知，我们所有人……

Not a lot. It was one of the weird things. It was this period of four and a half days where I didn't sleep much, didn't eat much, and still had a surprising amount of energy. You learn a weird thing about adrenaline in more time. So you kind of accepted the death of this baby, OpenAI. I was excited for the new thing. I was just like, okay, this was crazy but whatever. It's a very good coping mechanism. And then Saturday morning, two of the board members called and said, hey, we destabilized, we didn't mean to destabilize things, we don't want to destroy a lot of value here. Can we talk about you coming back? And I immediately didn't want to do that. But I thought a little more and I was like, well, I do care about the people here, the partners, shareholders, all of it. I love this company. And so I thought about it and I was like, well okay, but here's the stuff I would need. And then the most painful time of all was over the course of that weekend. I kept thinking and being told and we all...

不只是我，整个团队都在努力稳住局面，而全世界都在试图拆散它。有人来挖人，等等。我们一直被告知：‘快了，快了，再给我们一点时间。’那是一个非常混乱的状态。然后周日晚上，我原本每隔几小时就以为事情要结束了，我们会找到让我回归的办法，一切恢复原样，但董事会却任命了一位新的临时 CEO。我当时想：‘这感觉太糟糕了。’那是整件事的最低谷。

Not just me, the whole team here kept thinking while we were trying to keep it stabilized, and the whole world was trying to break it apart. People trying to recruit, whatever. We kept being told, 'Alright, we're almost done, we're almost done, we just need a little more time.' And it was this very confusing state. Then Sunday evening, when again every few hours I expected that we were going to be done and we were going to figure out a way for me to return and things to go back to how they were, the board then appointed a new interim CEO. And I was like, 'I mean, that feels really bad.' That was the low point of the whole thing.

你曾高度赞扬 Mira Murati，说她在关键时刻的安静时刻帮了大忙。也许我们可以稍微岔开话题：你欣赏她什么？

You've spoken highly of Mira Murati, that she helped especially in the quiet moments when it counts. Perhaps we could take a bit of a tangent: what do you admire about her?

她在那个混乱的周末做得非常出色。但人们往往只在危机时刻看到领导者，无论好坏。而我真正看重领导者的，是他们在某个平淡的周二早上 9:46 的表现，在日常的琐碎工作中如何行事。一个人如何出现在会议上，他们做决策的质量如何。这就是我所说的安静时刻。大部分工作都是日复一日、一个会议接一个会议完成的。只要在场并做出好的决策。

She did a great job during that weekend in a lot of chaos. But people often see leaders in crisis moments, good or bad. But a thing I really value in leaders is how people act on a boring Tuesday at 9:46 in the morning, in the normal drudgery of the day-to-day. How someone shows up in a meeting, the quality of the decisions they make. That was what I meant about the quiet moments. Most of the work is done day by day, meeting by meeting. Just be present and make great decisions.

我的意思是，你刚才想花最后 20 分钟讲这个戏剧性的周末，我理解。但这并不是 OpenAI 的全部。OpenAI 真正关乎的是其他七年。

I mean, look, you wanted to spend the last 20 minutes about this one very dramatic weekend, and I understand. But that's not really what OpenAI is about. OpenAI is really about the other seven years.

是的。人类文明并不等于纳粹德国入侵苏联，但人们仍然关注它，因为这完全可以理解。它让我们洞察人性，人性的极端，也许人类文明的一些破坏和胜利就发生在那些时刻。它具有说明性。

Well, yeah. Human civilization is not about the invasion of the Soviet Union by Nazi Germany, but still that's something people focus on because it's very understandable. It gives us an insight into human nature, the extremes of human nature, and perhaps some of the damage and some of the triumphs of human civilization can happen in those moments. It's illustrative.

让我问问你关于 Ilya 的事。他被关在秘密核设施里当人质了吗？

Let me ask you about Ilya. Is he being held hostage in a secret nuclear facility?

没有。

No.

那普通的秘密设施呢？

What about a regular secret facility?

没有。

No.

那非秘密的核设施呢？

What about a nuclear non-secret facility?

都不是。也不是。我是说，这都快成梗了。你认识 Ilya 很久了。他显然是这场董事会闹剧的一部分。你现在和他关系如何？

Neither. Not that either. I mean, this is becoming a meme at some point. You've known Ilya for a long time. He was obviously in part of this drama with the board and all that kind of stuff. What's your relationship with him now?

我爱 Ilya。我非常尊重 Ilya。我现在对他的计划无可奉告；那是他的问题。但我真的希望我们能在我的职业生涯余下时间里继续合作。他比我年轻一点，也许他会工作得更久一些。

I love Ilya. I have tremendous respect for Ilya. I don't have anything I can say about his plans right now; that's a question for him. But I really hope we work together for the rest of my career. He's a little bit younger than me, maybe he works a little bit longer.

有个梗说他看到了类似 AGI 的东西，这让他内心非常担忧。Ilya 看到了什么？

There's a meme that he saw something like AGI and that gave him a lot of worry internally. What did Ilya see?

Ilya 没有看到 AGI。我们谁都没看到 AGI。我们还没有造出 AGI。但我确实认为，我非常喜欢 Ilya 的一点是，他非常认真地对待 AGI 和广泛的安全问题，包括这将对社会产生的影响。随着我们不断取得重大进展，Ilya 是过去几年里我花最多时间讨论这将意味着什么、我们需要做什么才能确保做对、确保我们成功完成使命的人之一。所以 Ilya 没有看到 AGI。但 Ilya 是人类的光荣，他如此思考和担忧如何确保我们做对这件事。

Ilya has not seen AGI. None of us have seen AGI. We've not built AGI. I do think one of the many things that I really love about Ilya is he takes AGI and the safety concerns broadly speaking, including things like the impact this is going to have on society, very seriously. And as we continue to make significant progress, Ilya is one of the people that I've spent the most time over the last couple of years talking about what this is going to mean, what we need to do to ensure we get it right, to ensure that we succeed at the mission. So Ilya did not see AGI. But Ilya is a credit to humanity in terms of how much he thinks and worries about making sure we get this right.

我过去和他聊过很多次。我觉得他谈论技术时，总是在做这种长期思考。所以他想的不是一年后会怎样，而是十年后。只是从第一性原理出发：如果它 Scaling（规模扩张），这里的基本原理是什么？它会走向何方？所以这是他思考所有其他安全问题的基础，这让他成为一个非常迷人、值得交谈的人。

I've had a bunch of conversations with him in the past. I think when he talks about technology, he's always doing this long-term thinking type of thing. So he's not thinking about what this is going to be in a year; he's thinking about in 10 years. Just thinking from first principles: okay, if it scales, what are the fundamentals here? Where is this going? And so that's a foundation for him thinking about all the other safety concerns and all that kind of stuff, which makes him a really fascinating human to talk with.

你知道他为什么一直保持沉默吗？他是不是又在进行某种灵魂探索？

Do you have any idea why he's been kind of quiet? Is he just doing some soul searching again?

我不想替他发言。我觉得你应该问他本人。他绝对是个深思熟虑的人。我认为 Ilya 总是在以一种非常好的方式进行灵魂探索。是的。而且他欣赏沉默的力量。另外，我听说他也可以很傻气，但我从没见过那一面。那种时候会很可爱。我从未见过傻气的 Ilya，但我也很期待看到。

I don't want to speak for him. I think you should ask him that. He's definitely a thoughtful guy. I think Ilya is always on a soul search in a really good way. Yes, yeah. Also he appreciates the power of silence. Also I'm told he can be a silly guy, which I've never seen that side of. It's very sweet when that happens. I've never witnessed a silly Ilya, but I look forward to that as well.

我最近和他一起参加了一个晚宴，他在和一只小狗玩耍，心情非常傻气，非常讨人喜欢。我当时想：‘天哪，这不是世界最常看到的 Ilya 的一面。’

I was at a dinner party with him recently and he was playing with a puppy and he was in a very silly mood, very endearing. And I was thinking, 'Oh man, this is not the side of Ilya that the world sees the most.'

那么，总结一下整个事件：你对董事会结构、对这一切以及未来的走向感觉良好吗？

So just to wrap up this whole saga: are you feeling good about the board structure, about all of this, and where it's moving?

我对新董事会感觉很好。至于 OpenAI 的结构，董事会的任务之一就是审视它，看看哪里可以使其更稳健。我们想先让新董事会成员到位，但我们在整个过程中显然学到了关于结构的教训。我没有特别深刻的话要说。那是一次疯狂、非常痛苦的经历。我认为这是一场完美的怪异风暴。对我来说，它预演了随着赌注越来越高，我们对稳健治理结构、流程和人员的需求。我有点庆幸它发生在那个时候，但经历它仍然极其痛苦。

I feel great about the new board. In terms of the structure of OpenAI, one of the board's tasks is to look at that and see where we can make it more robust. We wanted to get new board members in place first, but we clearly learned a lesson about structure throughout this process. I don't have super deep things to say. It was a crazy, very painful experience. I think it was a perfect storm of weirdness. It was a preview for me of what's going to happen as the stakes get higher and higher, and the need that we have for robust governance structures and process, and people. I am kind of happy it happened when it did, but it was a shockingly painful thing to go through.

这让你在信任他人方面更加犹豫了吗？

Did it make you more hesitant in trusting people?

是的，只是在个人层面。我认为我是一个极其信任他人的人。我一直秉持的生活哲学是：不要担心所有的偏执，不要担心边缘情况。你吃点小亏，换来的是可以放下防备地生活。而这件事对我冲击太大，我完全措手不及，所以它确实改变了。我真的很不喜欢这样。它确实改变了我对默认信任他人以及为坏情况做打算的看法。你必须小心这一点。

Yes, just on a personal level. I think I'm an extremely trusting person. I have always had a life philosophy of don't worry about all the paranoia, don't worry about the edge cases. You get a little bit screwed in exchange for getting to live with your guard down. And this was so shocking to me, I was so caught off guard, that it has definitely changed. And I really don't like this. It's definitely changed how I think about default trust of people and planning for the bad scenarios. You got to be careful with that.

你担心自己变得有点过于愤世嫉俗吗？

Are you worried about becoming a little too cynical?

我不担心变得过于愤世嫉俗。我认为我完全不是一个愤世嫉俗的人。但我担心的是变得不那么默认信任他人。我其实不确定哪种模式最好。

I'm not worried about becoming too cynical. I think I'm the extreme opposite of a cynical person. But I'm worried about just becoming less of a default trusting person. I'm actually not sure which mode is best to be in.

对于正在开发 AGI 的人来说，信任或不信任是一段有趣的旅程。但在结构层面，我更关心人性层面：你如何让自己身边既有创造酷东西的人，又能做出明智决策？因为随着你赚的钱越多，这个东西拥有的力量越大，人们就会变得越奇怪。

For a person who's developing AGI, trusting or untrusting, it's an interesting journey you're on. But in terms of structure, I'm more interested on the human level: how do you surround yourself with humans that are building cool stuff but also making wise decisions? Because the more money you start making, the more power the thing has, the weirder people get.

你可以对董事会成员以及我本应拥有的信任程度，或者我本应如何以不同方式行事发表各种评论。但就这里的团队而言，我认为你在这方面得给我很高的分数。我对我每天共事的人怀有巨大的感激、信任和尊重。被这样的人包围真的很重要。

I think you could make all kinds of comments about the board members and the level of trust I should have had there, or how I should have done things differently. But in terms of the team here, I think you'd have to give me a very good grade on that one. I have enormous gratitude, trust, and respect for the people that I work with every day. Being surrounded with people like that is really important.

我们共同的朋友 Elon 起诉了 OpenAI。他批评的本质是什么？他在多大程度上是对的，又在多大程度上是错的？

Our mutual friend Elon sued OpenAI. What is the essence of what he's criticizing? To what degree does he have a point, and to what degree is he wrong?

我不知道这到底是怎么回事。我们最初只是想着要成为一个研究实验室，完全不知道这项技术会如何发展。很难回到过去真正记住当时的样子——那是在语言模型成为大事之前，在我们对 API 或出售聊天机器人访问权限有任何想法之前，在我们有任何产品化想法之前。我们只是想尝试做研究，并不真正知道我们会用它做什么。对于许多全新的根本性事物，你开始在黑暗中摸索，做出一些假设，其中大部分后来被证明是错误的。然后我们清楚地意识到我们需要做不同的事情，并且需要更多的资本。所以我们说，好吧，这个结构不太适合；我们如何修补这个结构？然后你一次又一次地修补，最终得到的东西至少可以说是令人侧目的。但我们是逐步走到这一步的，我认为每一步都做出了合理的决定。这并不意味着如果现在能带着神谕回去，我不会完全以不同的方式去做，但当时没有神谕。总之，至于 Elon 真正的动机是什么，我不知道。

I don't know what it's really about. We started off just thinking we were going to be a research lab, having no idea how this technology was going to go. It's hard to go back and really remember what it was like then—this was before language models were a big deal, before we had any idea about an API or selling access to a chatbot, before we had any idea we were going to productize it all. We were just going to try to do research, not really knowing what we would do with that. With many new fundamentally new things, you start fumbling through the dark and make some assumptions, most of which turn out to be wrong. Then it became clear we were going to need to do different things and have huge amounts more capital. So we said, okay, the structure doesn't quite work for that; how do we patch the structure? Then you patch it again and again, and you end up with something that does look kind of eyebrow-raising, to say the least. But we got here gradually with, I think, reasonable decisions at each point along the way. That doesn't mean I wouldn't do it totally differently if we could go back now with an oracle, but you don't get the oracle at the time. Anyway, in terms of what Elon's real motivations here are, I don't know.

就你记忆所及，OpenAI 在博客文章中给出的回应是什么？你能总结一下吗？

To the degree you remember, what was the response that OpenAI gave in the blog post? Can you summarize it?

我们只是说，Elon 说了这样一套话，这是我们的描述——或者说这不是我们的描述，这是事情经过的描述。我们尽量不情绪化，只是陈述历史。我确实认为 Elon 对他刚刚提出的某一点有一定程度的歪曲，那就是当时的不确定性程度。你们是一小群研究人员，疯狂地谈论 AGI，而当时所有人都在嘲笑这个想法。不久之前，Elon 疯狂地谈论发射火箭，而人们也在嘲笑那个想法。所以我认为他应该对此更有同理心。我确实认为这里有个人因素：存在分歧，OpenAI 和这里的许多优秀人才选择了与 Elon 分道扬镳。

We just said, Elon said this set of things, here's our characterization—or here's the sort of not our characterization, here's the characterization of how this went down. We tried not to make it emotional and just sort of say, here's the history. I do think there's a degree of mischaracterization from Elon here about one of the points he just made, which is the degree of uncertainty at the time. You guys were a bunch of small group of researchers crazily talking about AGI when everybody was laughing at that thought. Wasn't that long ago Elon was crazily talking about launching rockets when people were laughing at that thought. So I think he'd have more empathy for this. I do think there's personal stuff here: there was a split, and OpenAI and a lot of amazing people here chose to part ways with Elon.

你能准确描述一下吗？分道扬镳这件事。

Can you describe that exactly? The choosing to part ways.

他认为 OpenAI 会失败。他想要完全控制权来扭转局面。我们想继续沿着现在成为 OpenAI 的方向前进。他还希望特斯拉能够在不同时期建立 AGI 项目。他想把 OpenAI 变成他能控制的营利性公司，或者让它与特斯拉合并。我们不想那样做，他决定离开。这没问题。

He thought OpenAI was going to fail. He wanted total control to sort of turn it around. We wanted to keep going in the direction that now has become OpenAI. He also wanted Tesla to be able to build an AGI effort at various times. He wanted to make OpenAI into a for-profit company that he could have control of, or have it merge with Tesla. We didn't want to do that, and he decided to leave. That's fine.

所以你是说——这也是博客文章中提到的一点——他基本上想让 OpenAI 被特斯拉收购？

So you're saying—and that's one of the things the blog post says—that he wanted OpenAI to be basically acquired by Tesla?

是的，类似的方式，或者可能比与微软的合作更戏剧化。我的记忆是，提议就是让特斯拉收购，并由特斯拉完全控制。我很确定就是这样。

Yeah, in the same way, or maybe something similar, or maybe something more dramatic than the partnership with Microsoft. My memory is the proposal was just like, yeah, get acquired by Tesla and have Tesla have full control over it. I'm pretty sure that's what it was.

那么当时 OpenAI 中的“开放”一词对 Elon 意味着什么？Ilya 在邮件交流中谈到过这一点。当时对你意味着什么，现在对你又意味着什么？

So what does the word 'open' in OpenAI mean to Elon at the time? Ilya has talked about this in the email exchanges. What does it mean to you at the time, and what does it mean to you now?

我肯定会选一个不同的名字——说到带着神谕回去，我会选一个不同的名字。我认为 OpenAI 所做的最重要的事情之一，就是将强大的技术免费交到人们手中，作为公共产品。我们不在免费版本上投放广告，也不以其他方式变现。我们只是说这是我们的使命的一部分：我们希望将越来越强大的工具免费交到人们手中，并让他们使用它们。我认为这种“开放”对我们的使命非常重要。如果你给人们伟大的工具，并教他们使用——甚至不教，他们自己也能搞明白——然后让他们用这些工具为彼此建设一个不可思议的未来，这很重要。所以如果我们能继续向世界提供免费或低成本的强大 AI 工具，这对我们如何履行使命至关重要。开源与否？我认为我们应该开源一些东西，而不是其他东西。这确实变成了一条宗教战线，很难有细微差别，但我认为细微差别才是正确的答案。

I would definitely pick a different name—speaking of going back with an oracle, I'd pick a different name. One of the things that I think OpenAI is doing that is the most important of everything we're doing is putting powerful technology in the hands of people for free as a public good. We don't run ads on our free version, we don't monetize it in other ways. We just say it's part of our mission: we want to put increasingly powerful tools in the hands of people for free and get them to use them. I think that kind of 'open' is really important to our mission. If you give people great tools and teach them to use them—or don't even teach them, they'll figure it out—and let them go build an incredible future for each other with that, that's a big deal. So if we can keep putting free or low-cost powerful AI tools out in the world, it's a huge deal for how we fulfill the mission. Open source or not? I think we should open source some stuff and not other stuff. It does become this religious battle line where nuance is hard to have, but I think nuance is the right answer.

所以他说，“把名字改成 ClosedAI，我就撤诉。”我是说，这会在 meme 领域成为战场吗？我认为这说明了 Elon 对诉讼的认真程度。这话说得令人震惊。我不认为这起诉讼在法律上是严肃的——如果我错了请纠正我——它更多是为了就 AGI 的未来和目前领先的公司发表一个观点。

So he said, 'Change your name to ClosedAI and I'll drop the lawsuit.' I mean, is it going to become this battleground in the land of memes? I think that speaks to the seriousness with which Elon means the lawsuit. That's an astonishing thing to say. I don't think the lawsuit is legally serious—maybe correct me if I'm wrong—it's more to make a point about the future of AGI and the company that's currently leading the way.

听着，我的意思是，Grok 之前没有开源任何东西，直到人们指出这有点虚伪，然后他宣布本周 Grok 将开源一些东西。我认为开源与否对他来说并不是真正的问题。好吧，我们会讨论开源与否。我确实认为批评竞争对手很好，只是说一点——那很好——但友好竞争与……我个人讨厌诉讼。是的，听着，我认为整件事对一个建设者来说是不体面的。我尊重 Elon，他是我们这个时代最伟大的建设者之一，我知道他知道被仇恨者攻击是什么感觉。这让我更加难过，他这样对我们。他是有史以来最伟大的建设者之一——可能是有史以来最伟大的建设者。这让我难过，我想也让很多人难过。有很多人真的……

Look, I mean, Grok had not open-sourced anything until people pointed out it was a little bit hypocritical, and then he announced that Grok would open-source things this week. I don't think open source versus not is what this is really about for him. Well, we'll talk about open source and not. I do think maybe criticizing the competition is great, just talking a little—that's great—but friendly competition versus like... I personally hate lawsuits. Yeah, look, I think this whole thing is unbecoming of a builder. I respect Elon as one of the great builders of our time, and I know he knows what it's like to have haters attack him. It makes me extra sad he's doing it to us. He's one of the greatest builders of all time—potentially the greatest builder of all time. It makes me sad, and I think it makes a lot of people sad. There are a lot of people who've really...

我仰慕他很久了，然后说了这句话。我说，你知道，在某个采访里，我说我怀念以前的 Elon，我收到的无数条消息都说那完全就是我的感受。我觉得他就该赢。他应该让 X Grok 打败 GPT，然后 GPT 再打败 Grok，这就是一场竞争，对所有人都很美好。但关于开源的问题，你觉得有很多公司在尝试这个想法吗？这很有意思。我会说，令人惊讶的是，Meta 在这方面走在了前面，或者至少迈出了第一步，真正开源了模型。当然，它不是最先进的模型，但开源了 Llama。而 Google 也在考虑开源一个小版本。你考虑过这个想法吗？开源的利弊是什么？

I looked up to him for a long time and said this. I said, you know, in some interview or something that I missed the old Elon, and the number of messages I got being like that exactly encapsulates how I feel. I think he should just win. He should just make X Grok beat GPT, and then GPT beats Grok, and it's just a competition, and it's beautiful for everybody. But on the question of open source, do you think there's a lot of companies playing with this idea? It's quite interesting. I would say surprisingly, Meta has led the way on this, or at least took the first step in the chess game of really open sourcing the model. Of course, it's not the state-of-the-art model, but open sourcing Llama. And Google is flirting with the idea of open sourcing a smaller version. Have you played around with this idea? What are the pros and cons of open sourcing?

是的，我认为开源模型肯定有一席之地，尤其是人们可以在本地运行的较小模型。我认为对此有巨大的需求。我认为会有一些开源模型和一些闭源模型。在这方面，它与其他生态系统不会有太大不同。

Yeah, I think there is definitely a place for open source models, particularly smaller models that people can run locally. I think there's huge demand for that. I think there will be some open source models and some closed source models. It won't be unlike other ecosystems in that way.

我听了 All-In 播客讨论这个，关于 LLaMA 之类的东西，他们更担心从非营利转向这种 capped-profit 的先例。这为其他初创公司树立了什么先例？

I listened to the All-In podcast talking about this, the LLaMA stuff and all that kind of stuff, and they were more concerned about the precedent of going from nonprofit to this capped-profit. What precedent does that set for other startups?

我会强烈劝阻任何考虑先以非营利形式起步、之后再增加营利部门的初创公司。我会强烈劝阻他们这样做。我不认为我们会在这里树立一个先例。好吧，所以大多数初创公司应该直接走营利路线。再说一次，如果我们知道会发生什么，我们也会那样做。嗯，理论上，如果你在这里操作得当，会有一些税收优惠之类的，但我认为大多数人不会这样考虑问题。用这种方式，初创公司根本不可能省下很多钱。

I would heavily discourage any startup that was thinking about starting as a nonprofit and adding a for-profit arm later. I'd heavily discourage them from doing that. I don't think we'll set a precedent here. Okay, so most startups should just go for-profit. And again, if we knew what was going to happen, we would have done that too. Well, in theory, if you dance beautifully here, there are some tax incentives or whatever, but I don't think that's how most people think about these things. It's just not possible to save a lot of money for a startup if you do it this way.

不，我认为有些法律会让这变得相当困难。你希望和 Elon 的关系走向何方？这种紧张，这种博弈。你希望什么？如果从现在起一两年、三年后，你和他在个人层面上的关系——友谊、友好竞争，诸如此类？

No, I think there are laws that would make that pretty difficult. Where do you hope this goes with Elon? This tension, this dance. What do you hope? If we go one, two, three years from now, your relationship with him on a personal level too—friendship, friendly competition, just all this kind of stuff?

是的，我的意思是，我非常尊重 Elon，我希望未来几年我们能保持友好的关系。

Yeah, I mean, I really respect Elon, and I hope that years in the future we have an amicable relationship.

是的，我希望你们能有友好的关系，只是竞争、获胜，并一起探索这些想法。我确实认为存在人才竞争之类的东西，但应该是友好竞争。只管打造酷的东西。Elon 很擅长打造酷的东西，但你也是。说到酷，Sora。我可以问一百万个问题。首先，它太棒了。在产品层面，它确实令人惊叹，在哲学层面也是如此。所以让我问一个技术哲学问题：你认为它比 GPT-4 对世界理解得更多还是更少？例如，当你用这些 patch 训练时，世界模型与语言 token 相比如何？

Yeah, I hope you guys have an amicable relationship, just compete and win and explore these ideas together. I do suppose there's competition for talent or whatever, but it should be friendly competition. Just build cool stuff. And Elon is pretty good at building cool stuff, but so are you. So speaking of cool, Sora. There are a million questions I could ask. First of all, it's amazing. It truly is amazing on a product level, but also just on a philosophical level. So let me just ask a technical philosophical question: What do you think it understands about the world more or less than GPT-4, for example? The world model when you train on these patches versus language tokens.

我认为所有这些模型对世界模型的理解比我们大多数人认为的要多。而且，由于它们确实有一些明显不理解或搞不对的地方，我们很容易盯着弱点，看穿表象，说“啊，这只是假的”。但这不全是假的。只是有些部分有效，有些部分无效。比如我记得刚开始看 Sora 视频时，我看到一个人在某物前走几秒钟，遮挡了它，然后走开，同一个东西还在那里。我当时想，“哦，这相当不错。”或者有些例子中，底层物理在序列的许多步骤中表现得非常好。就像，“哦，这相当令人印象深刻。”但根本上，这些模型只会越来越好，而且这种情况会持续发生。如果你看从 DALL-E 1 到 2 到 3 再到 Sora 的轨迹，每个版本都有很多人吐槽，说它不能做这个，不能做那个，但现在看看它。

I think all of these models understand something more about the world model than most of us give them credit for. And because there are also very clear things they just don't understand or don't get right, it's easy to look at the weaknesses, see through the veil, and say, 'Ah, this is just fake.' But it's not all fake. It's just some of it works and some of it doesn't work. Like I remember when I started watching Sora videos, I would see a person walk in front of something for a few seconds and occlude it, and then walk away, and the same thing was still there. I was like, 'Oh, it's pretty good.' Or there are examples where the underlying physics looks so well represented over a lot of steps in a sequence. It's like, 'Oh, this is quite impressive.' But fundamentally, these models are just getting better, and that will keep happening. If you look at the trajectory from DALL-E 1 to 2 to 3 to Sora, there were a lot of people that dunked on each version, saying it can't do this, it can't do that, and look at it now.

嗯，你刚才提到的遮挡问题，基本上就是足够好地建模了世界的三维物理，以捕捉这类现象。或者也许你可以告诉我：为了处理遮挡，世界模型需要做什么？

Well, the thing you just mentioned with the occlusions is basically modeling the physics of the three-dimensional physics of the world sufficiently well to capture those kinds of things. Or maybe you can tell me: in order to deal with occlusions, what does the world model need to do?

所以我要说的是，它在处理遮挡方面做得非常好。我是否认为它有一个很好的底层 3D 世界模型？这有点牵强。但你能仅仅通过这种二维训练数据方法达到那个程度吗？看起来这种方法会走得惊人地远。我不想过多猜测它能克服哪些限制、不能克服哪些。

So what I would say is it's doing something to deal with occlusions really well. What I represent that it has like a great underlying 3D model of the world? It's a little bit more of a stretch. But can you get there through just these kinds of two-dimensional training data approaches? It looks like this approach is going to go surprisingly far. I don't want to speculate too much about what limits it will surmount and which it won't.

你看到的系统有哪些有趣的局限性？我是说，你发过一些好玩的例子。各种各样。比如猫在视频中随机长出一条额外的腿。还有很多问题，很多弱点。你认为这是方法上的根本缺陷，还是只是更大的模型、更好的技术细节、更好的数据、更多的数据就能解决这些问题？

What are some interesting limitations of the system that you've seen? I mean, there have been some fun ones you've posted. There are all kinds. Like cats sprouting an extra limb at random points in a video. There are still a lot of problems, a lot of weaknesses. Do you think it's a fundamental flaw of the approach, or is it just bigger model, better technical details, better data, more data that will solve those?

猫长腿？我两个都同意。比如，我认为这种方法有一些东西感觉与我们思考和学习的方式不同。而且，我认为它会随着规模扩大而变得更好。就像我提到的，LLM 有 token，文本 token，而 Sora 有视觉 patch。所以它将所有视觉数据——各种视觉数据，视频和图像——转换成 patch。训练在多大程度上可以说是完全自监督的？有没有一些人工标注？人类在其中的参与度如何？

The cat sprouting? I say yes to both. Like, I think there is something about the approach which just seems to feel different from how we think and learn and whatever. And also, I think it'll get better with scale. Like I mentioned, LLMs have tokens, text tokens, and Sora has visual patches. So it converts all visual data—diverse kinds of visual data, videos and images—into patches. Is the training, to the degree you can say, fully self-supervised? Is there some manual labeling going on? What's the involvement of humans in all this?

我的意思是，在不透露 Sora 方法具体细节的情况下，我们在工作中使用了大量人类数据，但不是互联网规模的数据。所以是大量人类。“大量”是一个复杂的词，Sam。我认为在这种情况下“大量”是一个公平的词。

I mean, without saying anything specific about the Sora approach, we use lots of human data in our work, but not internet-scale data. So lots of humans. 'Lots' is a complicated word, Sam. I think 'lots' is a fair word in this case.

但这并不——因为对我来说，“大量”——听着，我是个内向的人，当我和三个人一起玩时，那就是很多人了。四个人，那就很多了。但我猜你的意思是超过三个人在为这些模型标注数据？

But it doesn't—because to me, 'lots'—listen, I'm an introvert, and when I hang out with three people, that's a lot of people. Four people, that's a lot. But I suppose you mean more than three people work on labeling the data for these models?

是的，好吧，行。但根本上，有很多自监督学习，因为你在技术报告中提到的是互联网规模的数据。那是另一个美丽的短语——就像诗歌一样。所以很多数据不是人工标注的；在那种意义上是自监督的。是的，然后问题是：互联网上有多少数据可以以这种方式使用，并且有利于这种自监督方法？如果我们知道自监督方法的细节就好了。你考虑过公开更多细节吗？

Yeah, okay, all right. But fundamentally, there's a lot of self-supervised learning, because what you mentioned in the technical report is internet-scale data. That's another beautiful phrase—it's like poetry. So it's a lot of data that's not human-labeled; it's self-supervised in that way. Yeah, and then the question is: how much data is there on the internet that could be used in this way that is conducive to this kind of self-supervised method? If only we knew the details of the self-supervised method. Have you considered opening up a little more details?

我们考虑过。

We have.

具体到 Sora，这很有意思——大型语言模型（LLM）的魔力能否开始转向视觉数据？要做到这一点需要什么？

Mean for Sora specifically, because it's so interesting that can the same magic of LLMs now start moving towards visual data and what does that take to do that?

在我看来答案是肯定的，但我们还有更多工作要做。

It looks to me like yes, but we have more work to do.

有哪些危险？你为什么担心发布这个系统？它可能带来哪些危险？

What are the dangers? Why are you concerned about releasing the system? What are some possible dangers of this?

坦率地说，在发布系统之前，我们必须做的一件事是让它达到人们期望的规模所需的效率水平。我不想低估这一点，而且那里还有大量工作要做。但你可以想象深度伪造、错误信息等问题。我们努力做一个深思熟虑的公司，考虑我们向世界发布什么，而且不需要太多思考就能想到它可能出问题的方式。这里有很多棘手的问题。

Frankly speaking, one thing we have to do before releasing the system is get it to work at a level of efficiency that will deliver the scale people are going to want from this. I don't want to downplay that, and there's still a ton of work to do there. But you can imagine issues like deepfakes, misinformation. We try to be a thoughtful company about what we put out into the world, and it doesn't take much thought to think about the ways this can go badly. There are a lot of tough questions here.

你正在处理一个非常棘手的领域。你认为训练人工智能应该属于或属于版权法下的合理使用吗？

You're dealing in a very tough space. Do you think training AI should be or is fair use under copyright law?

我认为这个问题背后的问题是：创造有价值数据的人是否应该得到某种补偿？我认为答案是肯定的。我还不知道答案是什么。人们提出了很多不同的方案；我们也尝试了一些不同的模式。但你知道，如果我是一名艺术家，例如，A：我希望能够选择退出别人以我的风格生成艺术，B：如果他们确实以我的风格生成艺术，我希望有某种经济模式与之关联。

I think the question behind that question is: do people who create valuable data deserve to have some way that they get compensated for use of it? And I think the answer is yes. I don't know yet what the answer is. People have proposed a lot of different things; we've tried some different models. But you know, if I'm an artist for example, A: I would like to be able to opt out of people generating art in my style, and B: if they do generate art in my style, I'd like to have some economic model associated with that.

这就像从 CD 到 Napster 再到 Spotify 的转变。我们必须找到某种模式。模式在变，但人们必须得到报酬。嗯，应该有某种激励。如果我们再放大来看，为了让人类继续做很酷的事情，我担心人类会做很酷的事情，而社会会找到某种方式来奖励它。这似乎是根深蒂固的。我们想要创造，想要有用，想要以某种方式获得地位。我不认为这会消失。但奖励可能不是金钱上的、财务上的。它可能是名声和对他人的赞美，也许是以其他方式获得财务回报。我想我不认为我们已经看到了经济体系运作方式的最后演变。

It's that transition from CDs to Napster to Spotify. We have to figure out some kind of model. The model changes, but people have got to get paid. Well, there should be some kind of incentive. If we zoom out even more, for humans to keep doing cool everything, I worry about humans are going to do cool and society is going to find some way to reward it. That seems pretty hardwired. We want to create, we want to be useful, we want to achieve status in whatever way. That's not going anywhere, I don't think. But the reward might not be monetary, financial. It might be like fame and celebration of other cool, maybe financial in some other way. I guess I don't think we've seen the last evolution of how the economic system is going to work.

是的，但艺术家和创作者看到 Sora 时很担心。他们就像，天哪……

Yeah, but artists and creators are worried when they see Sora. They're like, holy...

当然。摄影出现时，艺术家们也非常担心。然后摄影变成了一种新的艺术形式，人们通过拍照赚了很多钱。我认为类似的事情会继续发生。人们会以新的方式使用新工具。

Sure. Artists were also super worried when photography came out. And then photography became a new art form and people made a lot of money taking pictures. I think things like that will keep happening. People will use the new tools in new ways.

如果我们只看 YouTube 或类似平台，你认为未来五年有多少内容会使用 Sora 这样的 AI 生成？人们谈论 AI 在五年内会做多少工作。人们现有的框架是当前工作的百分之多少会被 AI 完全取代。

If we just look on YouTube or something like this, how much of that will be using Sora, like AI-generated content, do you think in the next five years? People talk about how many jobs AI is going to do in five years. And the framework that people have is what percentage of current jobs are just going to be totally replaced by some AI doing the job.

我的思考方式不是 AI 会做多少百分比的工作，而是 AI 会做多少百分比的任务，以及时间跨度如何。所以如果你想想经济中所有五秒钟的任务、五分钟的任务、五小时的任务，甚至五天的任务，AI 能做多少？我认为这是一个比 AI 能做多少工作更有趣、更有影响力、更重要的问题。因为它是一个工具，会以越来越高的复杂度、越来越长的时间跨度处理越来越多的任务，让人们能够在更高的抽象层次上操作。所以也许人们在他们所做的工作上效率更高，在某个时候，这不仅是量变，也是质变，关乎你能在头脑中处理的问题类型。

The way I think about it is not what percent of jobs AI will do, but what percent of tasks will AI do and over what time horizon. So if you think of all the five-second tasks in the economy, five-minute tasks, five-hour tasks, maybe even five-day tasks, how many of those can AI do? And I think that's a way more interesting, impactful, important question than how many jobs AI can do. Because it is a tool that will work at increasing levels of sophistication and over longer and longer time horizons for more and more tasks, and let people operate at a higher level of abstraction. So maybe people are way more efficient at the job they do, and at some point that's not just a quantitative change but that's a qualitative one too about the kinds of problems you can keep in your head.

我认为对于 YouTube 上的视频来说，情况会是一样的。很多视频，也许大部分，会在制作中使用 AI 工具，但它们仍然从根本上由一个人来思考、整合、完成部分工作，有点像导演和运营。

I think that for videos on YouTube, it'll be the same. Many videos, maybe most of them, will use AI tools in the production, but they'll still be fundamentally driven by a person thinking about it, putting it together, doing parts of it, sort of directing and running it.

是的。这太有趣了。我的意思是，这很可怕，但思考起来很有趣。我倾向于相信人类喜欢看其他人类。人类真的很关心其他人类。如果有比人类更酷、更好的东西，人类会关心它大约两天，然后他们就会回到人类身上。这似乎根深蒂固。

Yeah. It's so interesting. I mean, it's scary but it's interesting to think about. I tend to believe that humans like to watch other humans. Humans really care about other humans a lot. If there's a cooler thing that's better than a human, humans care about that for like two days and then they go back to humans. That seems very deeply wired.

这就像整个国际象棋的事情。哦，是的，但让我们每个人都继续下棋，忽略房间里的大象——人类相对于 AI 系统在国际象棋上真的很差。我们仍然跑步，而汽车快得多。我的意思是，有很多例子。

It's the whole chess thing. Oh yeah, but let's everybody keep playing chess and let's ignore the elephant in the room that humans are really bad at chess relative to AI systems. We still run races and cars are much faster. I mean, there are a lot of examples.

也许就像 Adobe 套件那样的工具，你可以更容易地制作视频等等。听着，我讨厌站在镜头前。如果我能找到一种不用站在镜头前的方法，我会很喜欢的。不幸的是，这需要一段时间。生成人脸正在进步，但在视频格式中生成特定人物与普通人物相比仍然很棘手。

And maybe just be tooling like in the Adobe suite type of way, where you can just make videos much easier and all that kind of stuff. Listen, I hate being in front of the camera. If I can figure out a way to not be in front of the camera, I would love it. Unfortunately, it'll take a while. Generating faces is getting there, but generating faces in video format is tricky when it's specific people versus generic people.

让我问问你关于 GPT-4 的问题。很多问题。首先，它也很棒。回顾过去，它可能和 3.5 和 4 一样是历史性的关键时刻。也许 GPT-5 会成为关键时刻。我不知道。向前看很难说。我们永远不知道。这就是未来的烦人之处；很难预测。但对我来说，回顾过去，GPT-4、ChatGPT 非常令人印象深刻，具有历史意义。所以请允许我问：对你来说，GPT-4 最令人印象深刻的能力是什么？还有 GPT-4 Turbo？

Let me ask you about GPT-4. So many questions. First of all, also amazing. It's looking back, it'll probably be this kind of historic pivotal moment with 3.5 and 4. Which maybe GPT-5 will be the pivotal moment. I don't know. Hard to say that looking forwards. We never know. That's the annoying thing about the future; it's hard to predict. But for me, looking back, GPT-4, ChatGPT is pretty damn impressive, like historically impressive. So allow me to ask: what's been the most impressive capabilities of GPT-4 to you? And GPT-4 Turbo?

我认为它有点糟糕。典型的人类，也习惯了一件很棒的东西。不，我认为它是一件了不起的东西。但相对于我们需要达到的地方以及我相信我们会达到的地方，你知道，在 GPT-3 的时候，人们会说，‘哦，这太神奇了，这是技术的奇迹’，它确实是，曾经是。但你知道，现在我们有了 GPT-4，回头看 GPT-3，你会觉得，‘那简直难以想象的糟糕。’我预计 5 和 4 之间的差距会和 4 与 3 之间的差距一样。我认为我们的工作是活在未来几年，记住我们现在的工具回头看会有点糟糕，而这就是我们确保未来更好的方式。

I think it kind of sucks. Typical human, also gotten used to an awesome thing. No, I think it is an amazing thing. But relative to where we need to get to and where I believe we will get to, you know, at the time of GPT-3, people were like, 'Oh, this is amazing, this is a marvel of technology,' and it is, it was. But you know, now we have GPT-4 and look at GPT-3 and you're like, 'That's unimaginably horrible.' I expect that the delta between 5 and 4 will be the same as between 4 and 3. And I think it is our job to live a few years in the future and remember that the tools we have now are going to kind of suck looking backwards at them, and that's how we make sure the future is better.

GPT-4 在哪些最辉煌的方面很糟糕？意思是，它能做的最好的事情是什么，以及这些最好事情的局限性是什么，让你说它很糟糕，从而给你灵感和对未来的希望？

What are the most glorious ways in which GPT-4 sucks? Meaning, what are the best things it can do, and the limits of those best things that allow you to say it sucks, therefore gives you an inspiration and hope for the future?

你知道，我最近更多地用它来做的一件事是作为头脑风暴伙伴。在这方面，它闪现出一些令人惊叹的东西。我不认为它达到了……你知道，当人们谈论它能做什么时，他们会说，‘啊，它帮助我更高效地编码，帮助我写得更快更好，帮助我从一种语言翻译到另一种语言。’所有这些都很好，但头脑风暴是我看到真正潜力的地方。

You know, one thing I've been using it for more recently is sort of a brainstorming partner. And for that, there's a glimmer of something amazing in there. I don't think it gets... you know, when people talk about what it does, they're like, 'Ah, it helps me code more productively, it helps me write more faster and better, it helps me translate from this language to another.' All that is great, but the brainstorming thing is where I see the real potential.

这些都很棒，但有一种创意头脑风暴伙伴的感觉：我需要给这个东西起个名字，我需要换个方式思考这个问题，我不确定该怎么做。我觉得这让我们得以一窥我希望看到更多的东西。另一个能窥见一斑的是，当它能帮助完成更长期的任务时，你知道，分解多个步骤，也许执行其中一些步骤，搜索互联网，写代码，等等，把它们整合起来。当它奏效时——虽然不常见——那种与人类反复迭代的过程非常神奇。

These like amazing things, but there's something about the kind of creative brainstorming partner: I need to come up with a name for this thing, I need to think about this problem in a different way, I'm not sure what to do here. I think that gives a glimpse of something I hope to see more of. One of the other things that you can see a very small glimpse of is when it can help on longer horizon tasks, you know, break down some multiple steps, maybe execute some of those steps, search the internet, write code, whatever, put that together. When that works, which is not very often, it's very magical: the iterative back and forth with a human.

嗯，对我来说它经常奏效。

Well, it works a lot for me.

你什么意思？反复迭代？人类能更频繁地参与？

What do you mean? The iterative back and forth? The human can get more often?

当它能独立完成一个 10 步问题时，并不经常奏效。有时是在多个抽象层次上，或者你只是指顺序执行？两者都有，比如分解问题，然后在不同抽象层次上处理，再整合起来。

When it can go do a 10-step problem on its own, it doesn't work that often. Sometimes at multiple layers of abstraction, or do you mean just sequential? Both, like break it down and then do things at different layers of abstraction and put them together.

听着，我不想贬低 GPT-4 的成就，但也不想夸大其词。我认为我们正处于指数曲线上：很快我们回顾 GPT-4 就会像现在回顾 GPT-3 一样。话虽如此，ChatGPT 是一个转折点，人们开始相信了。相信的人有所增加，也许 OpenAI 内部不算，这里已经有信徒了，但仔细想想，从这个意义上说，我认为这将是一个让世界上很多人从不信到信的时刻。这更多是关于 ChatGPT 的界面，而不是底层模型本身。

Look, I don't want to downplay the accomplishment of GPT-4, but I don't want to overstate it either. I think this point that we are on an exponential curve: we will look back relatively soon at GPT-4 like we look back at GPT-3 now. That said, I mean ChatGPT was a transition to where people started to believe. There was an uptick of believing, not internally at OpenAI perhaps, there are believers here, but when you think about it, in that sense I do think it'll be a moment where a lot of the world went from not believing to believing. That was more about the ChatGPT interface than the underlying model itself.

而且，通过界面和产品，我指的也是模型的后训练，以及我们如何调整它使其对你有用、让你知道如何使用它，而不仅仅是底层模型本身。

And by the interface and product, I also mean the post-training of the model and how we tune it to be helpful to you and how to use it, than the underlying model itself.

这两者各有多重要：底层模型和基于人类反馈的强化学习（RLHF）或类似的东西，它们让模型对人类更有吸引力、更高效、更有生产力？

How much of those two things are important: the underlying model and the RLHF or something of that nature that tunes it to be more compelling to the human, more effective and productive for the human?

我的意思是，两者都非常重要，但基于人类反馈的强化学习（RLHF），即后训练步骤，从算力角度看是在基础模型之上做的一小层包装，尽管工作量巨大，但非常重要，更不用说我们围绕它构建的产品了。从某种意义上说，我们必须做两件事：发明底层技术，然后想办法把它变成人们喜爱的产品，这不仅涉及实际的产品工作本身，还包括如何对齐它、让它有用，以及如何实现规模化让很多人同时使用，等等。这是一项未知的艰巨任务。我们知道必须扩大规模。我们必须做两件以前从未做过的事，两者都是相当重大的成就，然后还有很多其他公司以前必须做的规模化之类的事情。

I mean, they're both super important, but the RLHF, the post-training step, the little wrapper of things that from a compute perspective is a little wrapper of things that we do on top of the base model, even though it's a huge amount of work, that's really important, to say nothing of the product that we build around it. In some sense, we did have to do two things: we had to invent the underlying technology, and then we had to figure out how to make it into a product people would love, which is not just about the actual product work itself, but this whole other step of how you align it and make it useful, and how you make the scale work where a lot of people can use it at the same time, all that kind of stuff. And that was an unknown difficult thing. We knew we were going to have to scale it up. We had to go do two things that had never been done before, both quite significant achievements, and then a lot of things like scaling it up that other companies have had to do before.

从 GPT-4 到 GPT-4 Turbo，上下文窗口从 8K 增加到 128K 个词元，这如何比较？

How does the context window going from 8K to 128K tokens compare from GPT-4 to GPT-4 Turbo?

人们喜欢长上下文，但大多数时候大多数人并不需要 128K 那么多。不过，如果我们展望遥远的未来，我们将拥有数十亿的上下文长度。你可以输入你所有的信息、你所有的历史记录，它会越来越了解你，那会很棒。目前，人们使用这些模型的方式并非如此。人们有时会输入一篇论文或代码库的很大一部分，但大多数时候，模型的大部分使用并不需要长上下文。

People like long context, but most people don't need all the way to 128K most of the time. Although if we dream into the distant future, we'll have context length of several billion. You will feed in all of your information, all of your history over time, and it'll just get to know you better and better, and that'll be great. For now, the way people use these models, they're not doing that. People sometimes post in a paper or a significant fraction of a code repository, but most usage of the models is not using the long context most of the time.

我喜欢这就像你的“我有一个梦想”演讲：有一天你会根据你性格或一生的全部背景来被评判。这很有趣。所以这就是你希望扩展的一部分：越来越大的上下文。

I like that this is your 'I Have a Dream' speech: one day you'll be judged by the full context of your character or of your whole lifetime. That's interesting. So that's part of the expansion you're hoping for: greater and greater context.

我曾经看过一个网络片段，数字可能不对，但就像比尔·盖茨在谈论某台早期计算机的内存，可能是 64K 或 640K 之类的，大部分用于屏幕缓冲区，他简直无法想象世界最终需要千兆字节或万亿字节的内存。你总是需要遵循技术的指数增长。我们会发现如何利用更好的技术。所以我真的无法想象未来上下文长度达到十亿会是什么样子。可能不会真的达到那个数字，但效果上会感觉像那样。但我知道一旦我们拥有它，我们会使用它，并且再也不想回到过去。

There was this internet clip I saw once, I'm going to get the numbers wrong, but it was like Bill Gates talking about the amount of memory on some early computer, maybe 64K, maybe 640K, something like that, and most of it was used for the screen buffer, and he just couldn't imagine that the world would eventually need gigabytes of memory or terabytes of memory in a computer. And you always just need to follow the exponential of technology. We will find out how to use better technology. So I can't really imagine what it's like right now for context length to go out to a billion someday. They might not literally go there, but effectively it'll feel like that. But I know we'll use it and really not want to go back once we have it.

是啊，即使说十年后达到数十亿可能也显得很蠢，因为那会是数万亿甚至更多。当然，会有某种突破，感觉上就像无限上下文。但即使是 128K，老实说，我还没用到那个程度：也许输入整本书或部分书籍、论文之类的。

Yeah, even saying billions 10 years from now might seem dumb because it'll be like trillions upon trillions. Sure, there'd be some kind of breakthrough that will effectively feel like infinite context. But even 128K, I have to be honest, I haven't pushed it to that degree: maybe putting in entire books or parts of books and so on, papers.

你见过哪些 GPT-4 的有趣用例？

What are some interesting use cases of GPT-4 that you've seen?

我觉得最有趣的不是某个我们可以谈论的具体案例，而是那些——主要是年轻人——把它作为任何知识工作任务的默认起点的人。事实上，它能相当好地完成很多事情：你可以用 GPT-4V，可以用它帮你写代码，可以用它帮你搜索，可以用它来编辑论文。对我来说最有趣的是那些把它作为工作流程起点的人。

The thing that I find most interesting is not any particular case that we can talk about, but it's people who kind of—this is mostly younger people—but people who use it as their default start for any kind of knowledge work task. And it's the fact that it can do a lot of things reasonably well: you can use GPT-4V, you can use it to help you write code, you can use it to help you do search, you can use it to edit a paper. The most interesting thing to me is the people who just use it as the start of their workflow.

我也在很多事情上这么做。我把它当作读书的伙伴。它帮我思考想法，尤其是经典书籍，因为关于它们的论述很充分。而且实际上——我发现它往往比维基百科在热门话题上还要好得多。它更平衡、更细致，或者可能是我自己的感觉，但它启发我比维基百科文章思考得更深入。我不太确定原因。你提到了这种协作。我不确定魔力在哪里：是在这里，还是在那里，或者介于两者之间。不确定。但当我用 GPT 开始知识任务时，我担心的一件事是之后通常需要事实核查，比如检查它有没有编造东西。你怎么解决这个问题？你知道 GPT 可能会编造听起来很有说服力的假东西。那么如何让它基于事实呢？

I do as well for many things. I use it as a partner for reading books. It helps me think through ideas, especially when the books are classic, so it's really well written about. And it actually is—I find it often to be significantly better than even Wikipedia on well-covered topics. It's somehow more balanced and more nuanced, or maybe it's me, but it inspires me to think deeper than a Wikipedia article does. I'm not exactly sure what that is. You mentioned this collaboration. I'm not sure where the magic is: if it's in here or if it's in there, or if it's somewhere in between. Not sure. But one of the things that concerns me for knowledge tasks when I start with GPT is I'll usually have to do fact-checking after, like check that it didn't come up with fake stuff. How do you figure that out? You know GPT can come up with fake stuff that sounds really convincing. So how do you ground it in truth?

这显然是我们非常感兴趣的领域。我认为随着即将推出的版本，它会变得好很多，但我们必须继续努力，而且我们不会让它像……

That's obviously an area of intense interest for us. I think it's going to get a lot better with upcoming versions, but we'll have to work on it, and we're not going to have it like...

今年全解决了？可怕的是，随着它变得更好，你会越来越不去做事实核查，对吧？

All solved this year? Well, the scary thing is, as it gets better, you'll start not doing the fact-checking more and more, right?

我对此有两种看法。我认为人们是比我们通常认为的更老练的技术用户。人们似乎真的明白 GPT——这些模型中的任何一个——有时会幻觉，如果任务关键，你必须检查。但记者们似乎不明白这一点。我见过记者半吊子地使用 GPT……这是我想吐槽记者的长清单上的一项，但这不是我对他们的首要批评。

I'm of two minds about that. I think people are much more sophisticated users of technology than we often give them credit for. People seem to really understand that GPT—any of these models—hallucinate some of the time, and if it's mission critical, you got to check it. Except journalists don't seem to understand that. I've seen journalists half-assedly just using GPT for... it's on the long list of things I'd like to dunk on journalists for, but this is not my top criticism of them.

嗯，我认为更大的批评或许是作为记者的压力和激励：你必须工作得非常快，而这是一个捷径。我希望我们的社会能激励——我也希望——需要数天和数周的长期新闻努力，并奖励出色的深度报道。还有以平衡方式呈现内容的新闻，在批评的同时赞美人们，尽管批评是吸引点击的东西。编造也能吸引点击，完全歪曲的标题也能。我肯定有很多人吐槽你。嗯，所有那些戏剧可能吸引了很多点击，确实如此。那是人类文明中一个我希望看到解决的更大问题：我们多赞美一点。

Well, I think the bigger criticism is perhaps the pressures and the incentives of being a journalist: you have to work really quickly, and this is a shortcut. I would love our society to incentivize—like I would too—long journalistic efforts that take days and weeks and rewards great in-depth journalism. Also journalism that presents stuff in a balanced way, where it celebrates people while criticizing them, even though the criticism is the thing that gets clicks. And making up also gets clicks, and headlines that mischaracterize completely. I'm sure you have a lot of people dunking on you. Well, all that drama probably got a lot of clicks, probably did. And that's a bigger problem about human civilization I'd love to see solved: where we celebrate a bit more.

你给了 ChatGPT 记忆能力——你一直在玩这个关于之前对话的功能——还有关闭记忆的能力。我希望我有时也能做到，根据需要开关……我想有时酒精可以做到，但不太理想，我想。通过这个，你看到了什么，在玩转记住和不记住对话这个想法时？

You've given ChatGPT the ability to have memories—you've been playing with that about previous conversations—and also the ability to turn off memory. I wish I could do that sometimes, just turn on and off depending... I guess sometimes alcohol can do that, but not optimally, I suppose. What have you seen through that, playing around with that idea of remembering conversations and not?

我们在这方面的探索还非常早期，但我认为人们想要的——或者至少我自己想要的——是一个逐渐了解我并随时间推移对我越来越有用的模型。这是一个早期的探索。我认为还有很多其他事情要做，但那是你想去的方向。你知道，你想使用一个模型，在你的一生中——或者使用一个系统，多个模型——在你的一生中，它变得越来越好。

We're very early in our explorations here, but I think what people want—or at least what I want for myself—is a model that gets to know me and gets more useful to me over time. This is an early exploration. I think there's a lot of other things to do, but that's where you'd like to head. You know, you'd like to use a model, and over the course of your life—or use a system, be many models—and over the course of your life, it gets better and better.

那个问题有多难？因为现在它更像是记住小事实和偏好等等。那记住像……你不想让 GPT 记住你在十一月经历的所有事情和所有戏剧吗？然后你，因为现在你显然有点在屏蔽它。

How hard is that problem? Because right now it's more like remembering little factoids and preferences and so on. What about remembering like... don't you want GPT to remember all the stuff you went through in November and all the drama? And then you, because right now you're clearly blocking it out a little bit.

不只是我想让它记住那个；我想让它整合其中的教训，并在未来提醒我该做什么不同的事情或注意什么。而且，你知道，我们都在一生中从经验中获益——程度不同——我希望我的 AI 智能体也能随着那些经验获益。所以如果我们回去想象一下，你知道，数万亿的上下文长度——如果我能把我一生中与任何人进行的所有对话都放进去，如果我能每次提问时把所有电子邮件的输入和输出都放在上下文窗口中，那会很酷，我想。

It's not just that I want it to remember that; I want it to integrate the lessons of that and remind me in the future what to do differently or what to watch out for. And, you know, we all gain from experience over the course of our lives—varying degrees—and I'd like my AI agent to gain with that experience too. So if we go back and let ourselves imagine that, you know, trillions and trillions of context length—if I can put every conversation I've ever had with anybody in my life in there, if I can have all of my emails input and output in the context window every time I ask a question, that'd be pretty cool, I think.

是的，我认为那会很酷。人们有时听到这个会担心隐私。你对此怎么看——AI 越有效地整合所有发生在你身上的经历和数据并给你建议？

Yeah, I think that would be very cool. People sometimes will hear that and be concerned about privacy. What do you think about that aspect of it—the more effective the AI becomes at really integrating all the experiences and all the data that happened to you and giving you advice?

我认为正确的答案就是用户选择。你知道，任何我想从我的 AI 智能体记录中删除的东西，我希望能够移除。如果我不想让它记住任何东西，我也希望如此。你和我可能对我们在自己的 AI 的隐私-效用权衡上想处于什么位置有不同的看法，这完全没问题。但我认为答案就是非常简单的用户选择。但公司应该对用户选择有高水平的透明度，因为过去有些公司有点阴暗——比如，是的，我们在收集你所有的数据，我们出于好的理由使用它，用于广告等等——但在细节上没有透明度。

I think the right answer there is just user choice. You know, anything I want stricken from the record for my AI agent, I want to be able to take out. If I don't want it to remember anything, I want that too. You and I may have different opinions about where on that privacy-utility tradeoff for our own AI you want to be, which is totally fine. But I think the answer is just really easy user choice. But there should be a high level of transparency from a company about the user choice, because sometimes companies in the past have been kind of shady about—like, yeah, we're collecting all your data and we're using it for a good reason, for advertisements and so on—but there's not a transparency about the details of that.

这完全正确。你之前提到我好像在屏蔽十一月的事——只是逗你。嗯，我的意思是，我认为那是一件非常创伤的事情，它确实让我长时间无法行动。比如，绝对是我工作中必须做的最难的事情就是在那段时间继续工作，因为我必须，你知道，试着回到这里，把碎片拼凑起来，而我当时处于震惊和痛苦之中。而且你知道，没人在乎那个。我的意思是，团队给了我通行证，我没有以正常水平工作，但有一段时间我只是……同时做这两件事真的很难。但有一天早上我醒来，我想，‘这发生在我身上是一件可怕的事情。我想我可以永远感觉自己像个受害者，或者我可以说这是我一生中接触过的最重要的工作，我需要回到它。’这并不意味着我压抑了它，因为有时我半夜醒来会想到它，但我确实觉得有义务继续前进。

That's totally true. You mentioned earlier that I'm like blocking out the November stuff—just teasing you. Well, I mean, I think it was a very traumatic thing, and it did immobilize me for a long period of time. Like, definitely the hardest work thing I've had to do was just keep working that period, because I had to, you know, try to come back in here and put the pieces together while I was just in sort of shock and pain. And you know, nobody really cares about that. I mean, the team gave me a pass, and I was not working on my normal level, but there was a period where I was just like... it was really hard to have to do both. But I kind of woke up one morning and I was like, 'This was a horrible thing to happen to me. I think I could just feel like a victim forever, or I can say this is the most important work I'll ever touch in my life and I need to get back to it.' And it doesn't mean that I've repressed it, because sometimes I wake in the middle of the night thinking about it, but I do feel like an obligation to keep moving forward.

嗯，说得很好。但可能还有一些挥之不去的东西。我担心的是你提到的信任问题——对别人偏执，而不是信任所有人或大多数人，凭直觉。这肯定是一个棘手的舞蹈。我的意思是，因为我在兼职探索中看到——我深入研究了 Z 政府、普京政府以及战时在高度压力环境下的动态。发生的事情是不信任，你孤立自己，你开始看不清世界。那是一个担忧——那是人类的担忧。你似乎迈出了一步，学到了好的教训，感受到了爱，让爱激励了你，这很好，但仍然可能挥之不去。

Well, that's beautifully said. But there could be some lingering stuff in there. What I would be concerned about is that trust thing that you mentioned—being paranoid about people, as opposed to just trusting everybody or most people, using your gut. It's a tricky dance for sure. I mean, because I've seen in my part-time explorations—I've been diving deeply into the Z Administration, the Putin Administration, and the dynamics there in wartime in a very highly stressful environment. And what happens is distrust, and you isolate yourself, and you start to not see the world clearly. And that's a concern—that's a human concern. You seem to have taken a stride and kind of learned the good lessons and felt the love and let the love energize you, which is great, but still can linger in there.

只是有一些问题我想问问你的直觉，关于 GPT 能做什么和不能做什么。所以它为生成的每个 token 分配大致相同的算力。在这种方法中，是否有空间进行更慢的思考、顺序思考？

There's just some questions I would love to ask your intuition about what GPT is able to do and not. So it's allocating approximately the same amount of compute for each token it generates. Is there room there in this kind of approach for slower thinking, sequential thinking?

我认为会有一种新的范式来进行那种思考。它在架构上会类似于我们现在看到的 LLM 吗？它是 LLM 之上的一层吗？

I think there will be a new paradigm for that kind of thinking. Will it be similar architecturally to what we're seeing now with LLMs? Is it a layer on top of the LLMs?

我可以想象很多种实现方式。我认为那不如你提出的问题重要，那就是：我们是否需要一种方式进行更慢的思考，答案不必……我想，从精神层面你可以说，你希望 AI 能够更努力地思考更难的问题。

I can imagine many ways to implement that. I think that's less important than the question you were getting at, which is: do we need a way to do a slower kind of thinking where the answer doesn't have to get... like, I guess spiritually you could say that you want an AI to be able to think harder about a harder problem.

对于更简单的问题，能更快地给出问题和答案。我认为重要的是，就像人类思考一样，你应该能够深入思考。这是错误的直觉吗？我怀疑这是一个合理的直觉。有趣。所以一旦 GPT 达到 GPT-7 的水平，我们就能瞬间看到费马大定理的证明吗？在我看来，你希望为更难的问题分配更多算力。比如，如果你问系统费马大定理的证明，而不是今天的日期，除非它已经知道并记住了证明的答案，否则它需要去推导出来，这似乎需要更多算力。但这能不能表现为语言模型自言自语那种形式？

Problem and answer more quickly about an easier problem. And I think that will be important is that like a human thought that we're just having, you should be able to think hard. Is that wrong intuition? I suspect that's a reasonable intuition. Interesting. So it's not possible once the GPT gets like GPT-7, we just instantaneously be able to see, you know, here's the proof of Fermat's theorem. It seems to me like you want to be able to allocate more compute to harder problems. Like it seems to me that a system, if you ask a system like that proof of Fermat's theorem versus what's today's date, unless it already knew and had memorized the answer to the proof, assuming it's got to go figure that out, seems like that will take more compute. But can it look like a basically LM talking to itself, that kind of thing?

也许吧。我的意思是，有很多可能的方式。但什么才是正确或最佳的方法，我们目前还不知道。

Maybe. I mean, there's a lot of things that you could imagine working. What the right or the best way to do that will be, we don't know.

这让我想起了 Q* 背后的神秘传说。这个神秘的 Q* 项目是什么？它也在那个核设施里吗？

This does make me think of the mysterious lore behind Q*. What's this mysterious Q* project? Is it also in the same nuclear facility?

没有核设施。

There is no nuclear facility.

有核设施的人总是这么说。我倒希望有个秘密核设施。但确实没有。好吧，也许有一天。有一天。人可以做梦。OpenAI 不擅长保守秘密。如果能保密就好了，我们被很多泄密困扰。你能谈谈 Q* 是什么吗？

That's what a person with a nuclear facility always says. I would love to have a secret nuclear facility. There isn't one. All right, maybe someday. Someday. One can dream. OpenAI is not a good company at keeping secrets. It would be nice, you know, we're like plagued by a lot of leaks, and it would be nice if we were able to have something like that. Can you speak to what Q* is?

我们还没准备好谈论这个。

We are not ready to talk about that.

看，这样的回答意味着确实有东西可谈。这很神秘，Sam。

See, but an answer like that means there's something to talk about. It's very mysterious, Sam.

我的意思是，我们从事各种研究。我们一直认为，让这些系统具备更好的推理能力是我们想要追求的重要方向。我们还没有破解这个难题。我们对此非常感兴趣。

I mean, we work on all kinds of research. We have said for a while that we think better reasoning in these systems is an important direction that we'd like to pursue. We haven't cracked the code yet. We're very interested in it.

会不会有像 Q* 或其他项目那样的时刻，出现类似 GPT 那样的飞跃？这是个好问题。我怎么看呢？我觉得很有趣。一切感觉都是连续的，对吧？你一直在说一个主题：这是一个渐进的过程，基本上是在逐渐爬上一个指数斜坡。但从局外人的角度看，对我来说，看着它确实感觉有飞跃，但对你来说却没有。

Is there going to be moments, Q* or otherwise, where there's going to be leaps similar to GPT, where you're like... That's a good question. What do I think about that? It's interesting to me. It all feels pretty continuous, right? This is kind of a theme that you're saying: there's a gradual, you're basically gradually going up an exponential slope. But from an outsider perspective, for me just watching it, it does feel like there's leaps, but to you there isn't.

我确实在想，我们是否应该……你知道，我们之所以这样部署，部分原因是我们认为，我们称之为迭代部署。我们不是秘密建造直到 GPT-5 才公开，而是决定谈论 GPT-1、2、3 和 4。部分原因是，我认为 AI 和 surprise 不应该同时出现。而且世界、人们、机构，无论你怎么称呼，都需要时间来适应和思考这些事情。我认为 OpenAI 做得最好的事情之一就是这一策略。我们让世界关注进展，认真对待 AGI，思考我们想要什么样的系统、结构和治理，而不是在压力下仓促做决定。我认为这非常好。但像你和其他人仍然觉得存在这些飞跃，这让我觉得也许我们应该做得更多，或者更迭代地发布。我不知道那意味着什么。我还没有准备好答案。但我们的目标不是给世界带来震惊的更新。恰恰相反。

I do wonder if we should have... So, you know, part of the reason that we deploy the way we do is that we think, we call it iterative deployment. We, rather than go build in secret until we got all the way to GPT-5, we decided to talk about GPT-1, 2, 3, and 4. And part of the reason there is I think AI and surprise don't go together. And also the world, people, institutions, whatever you want to call it, need time to adapt and think about these things. And I think one of the best things that OpenAI has done is this strategy. And we get the world to pay attention to the progress, to take AGI seriously, to think about what systems and structures and governance we want in place before we're like under the gun and have to make a rushed decision. I think that's really good. But the fact that people like you and others say you still feel like there are these leaps makes me think that maybe we should be doing, or releasing even more iteratively. I don't know what that would mean. I don't have an answer ready to go. But like, our goal is not to have shock updates to the world. The opposite.

是的，当然。更迭代会非常棒。我认为这对每个人都很好。但这就是我们正在努力做的。这是我们公开的策略。但我觉得我们 somehow 没有达到目标。所以也许我们应该考虑以不同的方式发布 GPT-5 之类的。

Yeah, for sure. More iterative would be amazing. I think that's just beautiful for everybody. But that's what we're trying to do. That's like our stated strategy. And I think we're somehow missing the mark. So maybe we should think about, you know, releasing GPT-5 in a different way or something like that.

是的，4.71、4.72。但人们喜欢庆祝。人们庆祝生日。我不知道你是否了解人类，但他们有这些里程碑。我确实认识一些人。人们确实喜欢里程碑。我完全理解。我认为我们也喜欢里程碑。宣布胜利然后开始下一件事很有趣。但我觉得我们 somehow 有点搞错了。

Yeah, 4.71, 4.72. But people tend to like to celebrate. People celebrate birthdays. I don't know if you know humans, but they kind of have these milestones. And I do know some humans. People do like milestones. I totally get that. I think we like milestones too. It's like fun to, you know, say declare victory on this one and go start the next thing. But yeah, I feel like we're somehow getting this a little bit wrong.

那么 GPT-5 什么时候发布？

So when is GPT-5 coming out again?

我不知道。这是诚实的回答。

I don't know. That's honest answer.

哦，这是诚实的回答。是吗？如果是今年就眨两下眼。

Oh, that's the honest answer. Is it? Blink twice if it's this year.

我也……我们今年会发布一个很棒的模型。我不知道我们会怎么称呼它。这就回到了问题：我们以什么方式发布这个东西？我们会在未来几个月发布很多不同的东西。我认为它们会很酷。我认为在我们谈论像 GPT-5 这样的模型之前，不管它叫不叫这个名字，或者比你对 GPT-5 的预期稍差或稍好，我认为我们还有很多其他重要的东西要先发布。

I also... We will release an amazing model this year. I don't know what we'll call it. So that goes to the question of like, what's the way we release this thing? We'll release over in the coming months many different things. I think they'll be very cool. I think before we talk about like a GPT-5-like model, called that or not called that, or a little bit worse or a little bit better than what you'd expect from a GPT-5, I think we have a lot of other important things to release first.

我不知道该对 GPT-5 期待什么。你让我既紧张又兴奋。对于最终不管叫什么名字的模型，但我们就叫它 GPT-5，需要克服的最大挑战和瓶颈是什么？只是好奇问问。是什么？是在算力方面？还是在技术方面？

I don't know what to expect from GPT-5. You're making me nervous and excited. What are some of the biggest challenges and bottlenecks to overcome for whatever it ends up being called, but let's call it GPT-5? Just interesting to ask. What are they? Is it on the compute side? Is it on the technical side?

总是所有这些。我……你知道那个最大的解锁是什么吗？是更大的计算机？还是像新秘密？还是别的什么？是所有这些因素加在一起。我认为 OpenAI 做得非常好的一点，这实际上是 Ilya 的原话，我可能会引用得不太准确，但大意是：我们把 200 个中等大小的东西乘在一起，变成一个巨大的东西。所以这里有一种分布式的持续创新在发生。

Always all of these. I was... You know what's the one big unlock? Is it a bigger computer? Is it like a new secret? Is it something else? It's all of these things together. Like the thing that OpenAI, I think, does really well, this is actually an original Ilya quote that I'm going to butcher, but it's something like: we multiply 200 medium-sized things together into one giant thing. So there's this distributed constant innovation happening.

是的，即使在技术方面，尤其是技术方面，所以即使是详细的方法，比如每个方面的细节……不同的团队之间如何协作？中等大小的东西如何变成一个巨大的 Transformer？这……有少数人需要思考如何把整个东西组合起来，但很多人试图在脑海中保持大部分图景。

Yeah, even on the technical side, like especially on the technical side, so even detailed approach, like detailed aspects of every... How does that work with different disparate teams and so on? How do they, how do the medium-sized things become one whole giant Transformer? How does this... There's a few people who have to like think about putting the whole thing together, but a lot of people try to keep most of the picture in their head.

哦，就像各个团队、个人贡献者试图保持高层次的视角。是的，我当然不知道每个部分具体如何运作。但我普遍相信的一点是，有时退一步看全局是有用的。我认为这对技术问题如此，对商业创新也是如此。事物以令人惊讶的方式组合在一起，理解整个图景，即使大部分时间你在某个领域深耕，也会带来令人惊讶的洞察。事实上，我曾经拥有并且认为非常有价值的一件事是，我曾经对科技行业的所有或大部分前沿领域有一个良好的地图，我有时能看到这些联系或新的可能性，如果我仅仅深入一个领域，我就不会有这些想法，因为我没有所有数据。而我现在并没有那么多。

Oh, like the individual teams, individual contributors try to keep at a high level. Yeah, I don't know exactly how every piece works, of course. But one thing I generally believe is that it's sometimes useful to zoom out and look at the entire map. And I think this is true for like a technical problem, I think this is true for like innovating in business. But things come together in surprising ways, and having an understanding of that whole picture, even if most of the time you're operating in the weeds in one area, pays off with surprising insights. In fact, one of the things that I used to have and I think was super valuable was I used to have like a good map of all of the, or most of the frontiers in the tech industry, and I could sometimes see these connections or new things that were possible that if I were only deep in one area, I wouldn't be able to have the idea for, because I wouldn't have all the data. And I don't really have that much.

说到宏观视角，我们再来看一个你发过的、既戏谑又深刻的东西。你发推说要筹集 7 万亿美元。

Speaking of zooming out, let's zoom out to another cheeky but profound thing you said. You tweeted about needing $7 trillion.

我没发过那个推。我从没说过我们要筹集 7 万亿美元。那是别人说的。

I did not tweet about that. I never said we're raising $7 trillion. That's somebody else.

但你说过可能是 8 万亿？我觉得你发过一次梗。世界上确实有错误信息，但这种错误信息可能也有洞察的基础。

But you said maybe eight? I think you memed once. There's misinformation out in the world, but sort of misinformation may have a foundation of insight.

听着，我认为算力将成为未来的货币。它可能会成为世界上最有价值的商品，我们应该大力投资以制造更多算力。算力是一个不寻常的市场。人们会想到手机芯片市场：全球 80 亿人，大概 70 亿人有手机，每两年换一次，所以每年市场是 30 亿颗手机系统级芯片。如果你生产 300 亿颗，你卖不掉 10 倍的手机，因为大多数人只有一部。但算力不同。智能将更像能源。唯一有意义的讨论方式是：在价格 X 下，世界会用这么多算力；在价格 Y 下，世界会用那么多算力。因为如果算力非常便宜，我会让它整天读我的邮件，给我建议该思考或做什么，并尝试治愈癌症。如果非常贵，也许我们只会用它来尝试治愈癌症。所以我认为世界将需要海量算力。其中有很多难点：能源是最难的，建设数据中心也很难，供应链很难，制造足够多的芯片也很难。但这似乎就是未来的方向。我们将需要的算力规模，目前很难想象。

Look, I think compute is going to be the currency of the future. I think it will be maybe the most precious commodity in the world, and I think we should be investing heavily to make a lot more compute. Compute is an unusual market. People think about the market for chips for mobile phones: there are 8 billion people, maybe 7 billion have phones, they upgrade every two years, so the market is 3 billion system-on-chips per year. If you make 30 billion, you won't sell 10 times as many phones because most people have one. But compute is different. Intelligence is going to be more like energy. The only thing that makes sense to talk about is: at price X, the world will use this much compute; at price Y, the world will use that much compute. Because if it's really cheap, I'll have it reading my email all day, giving me suggestions about what to think about or work on, and trying to cure cancer. If it's really expensive, maybe we'll only use it to try to cure cancer. So I think the world is going to want a tremendous amount of compute. There are many hard parts: energy is the hardest, building data centers is hard, the supply chain is hard, and fabricating enough chips is hard. But this seems to me where things are going. We're going to want an amount of compute that's just hard to reason about right now.

你怎么解决能源难题？核能？核聚变？

How do you solve the energy puzzle? Nuclear? Fusion?

我相信核聚变。我认为 Helion 做得最好，但我很高兴现在有一场核聚变竞赛。核裂变也非常了不起，我希望全世界能重新接纳它。那段历史让我非常难过，我希望我们能以有意义的方式回归核裂变。

That's what I believe. Fusion. I think Helion is doing the best work, but I'm happy there's a race for fusion right now. Nuclear fission is also quite amazing, and I hope as a world we can re-embrace it. It's really sad to me how the history of that went, and I hope we get back to it in a meaningful way.

所以对你来说，难题的一部分是核裂变，就像我们现有的核反应堆。很多人因为切尔诺贝利等事件感到恐惧。

So to you, part of the puzzle is nuclear fission like nuclear reactors as we currently have them. A lot of people are terrified because of Chernobyl and so on.

嗯，我认为我们应该制造新型反应堆。这个行业几乎停滞不前，真是遗憾。

Well, I think we should make new reactors. It's a shame that industry kind of ground to a halt.

你如何解释这种停滞？仅仅是集体歇斯底里吗？

What just mass hysteria is how you explain the halt?

我不确定你是否了解人类，但这是核裂变面临的危险之一，也是安全威胁之一：人类似乎非常害怕它，我们必须把这个因素纳入考量。我们必须赢得人们的支持，展示它有多安全。我担心 AI 也会面临同样的问题。我认为 AI 会出一些戏剧性的问题。我不知道我最终被枪杀的几率是多少，但绝不是零。

I don't know if you know humans, but that's one of the dangers, one of the security threats for nuclear fission: humans seem to be really afraid of it, and that's something we have to incorporate into the calculus. We have to win people over and show how safe it is. I worry about that for AI. I think some things are going to go theatrically wrong with AI. I don't know what the percent chance is that I eventually get shot, but it's not zero.

你怎么降低它的戏剧性？我已经开始听到一些风声，因为我和政治光谱两边的人都有交流。我听到 AI 将被政治化的议论。这让我非常担心，因为到时候可能右派反对 AI，左派支持 AI，因为 AI 会帮助人民，或者诸如此类的叙事。这让我非常担心，然后戏剧性就会被充分利用。你怎么应对？

How do you decrease the theatrical nature of it? I've already started to hear rumblings because I talk to people on both sides of the political spectrum. I hear rumblings that AI is going to be politicized. That really worries me because then maybe the right is against AI and the left is for it because it's going to help the people, or whatever the narrative is. That really worries me, and then the theatrical nature can be leveraged fully. How do you fight that?

我认为它会卷入左右之争。我不确定具体会是什么样子，但不幸的是，任何重要的事情都会这样。

I think it will get caught up in left versus right wars. I don't know exactly what that's going to look like, but I think that's just what happens with anything of consequence, unfortunately.

我所说的戏剧性风险，更多是指 AI 带来的好处将远远大于坏处，但它确实会有坏处。有些坏处很严重但不戏剧化。例如，死于空气污染的人比死于核反应堆的人多得多，但大多数人更担心住在核反应堆旁边而不是燃煤电厂旁边。我们大脑的某种机制是，尽管风险有很多种，但那些能成为电影高潮场景的风险，比那些长期缓慢恶化的风险，对我们的影响要大得多。

What I meant more about theatrical risks is that AI is going to have, I believe, tremendously more good consequences than bad ones, but it is going to have bad ones. There will be some bad ones that are bad but not theatrical. A lot more people have died of air pollution than nuclear reactors, for example, but most people worry more about living next to a nuclear reactor than a coal plant. Something about the way we're wired is that although there are many different kinds of risks, the ones that make a good climax scene of a movie carry much more weight with us than the ones that are very bad over a long period of time but on a slow burn.

嗯，这就是为什么真相很重要，希望 AI 能帮助我们看清事物的真相，保持平衡，理解什么是真正的风险，什么是世界上真正的危险，以及利弊得失。

Well, that's why truth matters, and hopefully AI can help us see the truth of things, to have balance, to understand what are the actual risks, what are the actual dangers of things in the world, what are the pros and cons.

在这个领域的竞争中，与 Google、Meta、xAI 等公司竞争，利弊是什么？

What are the pros and cons of the competition in the space, competing with Google, Meta, xAI, and others?

我认为这个问题有一个相当直接的答案，虽然我以后可能会想到更多细微之处。好处很明显：我们得到更好的产品、更多的创新、更快更便宜，以及竞争带来的所有好处。坏处是，如果我们不小心，可能会导致军备竞赛升级，这让我很担心。

I think I have a pretty straightforward answer to this, though I can think of more nuance later. The pros seem obvious: we get better products and more innovation faster and cheaper, and all the reasons competition is good. The con is that if we're not careful, it could lead to an increase in an arms race that I'm nervous about.

你是否以某种消极的方式感受到军备竞赛的压力？

Do you feel the pressure of the arms race in some negative way?

在某些方面确实如此。我们花了很多时间讨论优先考虑安全的必要性。我长期以来一直说，如果你考虑一个象限：到达 AGI 的时间线慢且长，然后起飞快或慢，我认为短时间线、慢起飞是最安全的象限，也是我最希望我们处于的象限。但我确实想确保我们实现慢起飞。我和埃隆之间有点小摩擦，问题在于形成了孤岛，而不是在安全方面进行合作。事情往往走向孤岛和封闭开源。在模型中，埃隆至少说他非常关心 AI 安全，并且非常担心，我假设他不会不安全地竞赛。但在这个问题上，合作对每个人都有好处。这倒不是他最出名的事。不过，他以关心人类而闻名，而人类从合作中受益。所以总是存在紧张、激励和动机。最终，我确实希望人类获胜。

Definitely in some ways, for sure. We spend a lot of time talking about the need to prioritize safety. I've said for a long time that if you think of a quadrant of slow timelines to the start of AGI and long timelines, and then a short takeoff or a fast takeoff, I think short timeline, slow takeoff is the safest quadrant and the one I'd most like us to be in. But I do want to make sure we get that slow takeoff part. The problem I have with this kind of slight beef with Elon is that silos are created, as opposed to collaboration on the safety aspect. It tends to go into silos and closed open source perhaps. In the model, Elon says at least that he cares a great deal about AI safety and is really worried about it, and I assume that he's not going to race unsafely. But collaboration here is really beneficial for everybody on that front. Not really the thing he's most known for. Well, he is known for caring about humanity, and humanity benefits from collaboration. So there's always tension and incentives and motivations. In the end, I do hope humanity prevails.

我想到，前几天有人提醒我，他在超越杰夫·贝佐斯成为世界首富的那天，发推对杰夫说“银牌”。

I was thinking someone just reminded me the other day about how the day he got surpassed Jeff Bezos for richest person in the world, he tweeted a silver medal at Jeff.

我希望随着人们开始朝着 AGI 努力，这类事情会减少。我认为 Elon 是朋友，他是个美好的人，也是有史以来最重要的人类之一。那些事情不好。Elon 的卓越之处令人惊叹，我非常尊重他。我认为我们需要他。我们所有人都应该支持他，并需要他在下一阶段挺身而出，发挥领导作用。

I hope we have less stuff like that as people start to work towards AGI. I think Elon is a friend and he's a beautiful human being, one of the most important humans ever. That stuff is not good. The amazing stuff about Elon is amazing, and I super respect him. I think we need him. All of us should be rooting for him and need him to step up as a leader through this next phase.

是的，我希望你能两者兼得，但有时人类是有缺陷和复杂的，诸如此类。历史上有很多真正伟大的领袖。是的，我们每个人都可以成为最好的自己，并努力做到这一点。

Yeah, I hope you can have one without the other, but sometimes humans are flawed and complicated and all that kind of stuff. There's a lot of really great leaders throughout history. Yeah, and we can each be the best version of ourselves and strive to do so.

让我问你。谷歌借助搜索，在过去 20 年里一直占据主导地位，我认为可以公平地说，在全球信息获取、我们如何互动等方面都是如此。对谷歌来说，但也是对整个领域的人来说，一个令人紧张的事情是思考人们将如何获取信息。是的，就像你说的，人们把 GPT 作为起点。那么 OpenAI 真的要接手谷歌 20 年前开始的这件事吗，即我们如何获得……

Let me ask you. Google, with the help of search, has been dominating the past 20 years, I think it's fair to say, in terms of the world's access to information, how we interact, and so on. And one of the nerve-wracking things for Google, but for the entirety of people in this space, is thinking about how are people going to access information. Yeah, like you said, people show up to GPT as a starting point. So is OpenAI going to really take on this thing that Google started 20 years ago, which is how do we get...

我觉得这很无聊。我的意思是，如果问题是，我们能否比谷歌造出更好的搜索引擎，那么当然，我们应该去做，你知道，人们应该使用更好的产品。但我认为这大大低估了这件事的潜力。你知道，谷歌给你展示 10 个蓝色链接，嗯，13 个广告然后 10 个蓝色链接，那只是查找信息的一种方式。但让我兴奋的不是我们可以造一个更好的谷歌搜索复制品，而是也许有某种更好的方式来帮助人们找到、处理并综合信息。实际上，我认为 ChatGPT 在某些用例上已经做到了，希望我们能让它在更多用例上也做到。但我觉得讨论如何比谷歌更好地给你 10 个排名网页来看，并不那么有趣。也许真正有趣的是，如何帮助你得到答案、你所需的信息，如何帮助你在某些情况下创造它，在其他情况下综合它，或者指引你找到它。但很多人尝试过只是造一个比谷歌更好的搜索引擎，这是一个困难的技术问题，一个困难的品牌问题，一个困难的生态系统问题。我认为世界不需要另一个谷歌的复制品。

I find that boring. I mean, if the question is like, can we build a better search engine than Google or whatever, then sure, we should go, you know, people should use a better product. But I think that would so understate what this can be. You know, Google shows you like 10 blue links, well, 13 ads and then 10 blue links, and that's like one way to find information. But the thing that's exciting to me is not that we can go build a better copy of Google search, but that maybe there's just some much better way to help people find, act on, and synthesize information. Actually, I think ChatGPT is that for some use cases, and hopefully we'll make it be like that for a lot more use cases. But I don't think it's that interesting to say like, how do we go do a better job of giving you like 10 ranked web pages to look at than what Google does. Maybe it's really interesting to go say, how do we help you get the answer, the information you need, how do we help create that in some cases, synthesize that in others, or point you to it in yet others. But a lot of people have tried to just make a better search engine than Google, and it is a hard technical problem, it is a hard branding problem, it is a hard ecosystem problem. I don't think the world needs another copy of Google.

将 ChatGPT 这样的聊天客户端与搜索引擎整合，这更酷，很酷，但也很棘手。就像如果你只是简单地做，会很尴尬，因为如果你只是把它塞进去，是的，可能会很尴尬。你可能猜到，我们对如何做好这件事很感兴趣。那将是一个很酷的例子，不仅仅是异质的，而是整合 LLM 加搜索的交集。我认为还没有人破解这个密码。我很乐意去做这件事。我认为那会很酷。

And integrating a chat client like ChatGPT with a search engine, that's cooler, it's cool, but it's tricky. It's like if you just do it simply, it's awkward because like if you just shove it in there, yeah, it can be awkward. As you might guess, we are interested in how to do that well. That would be an example of a cool thing that's not just like, well, like a heterogeneous, like integrating the intersection of LLMs plus search. I don't think anyone has cracked the code on yet. I would love to go do that. I think that would be cool.

是的，广告方面呢？你有没有考虑过变现？你知道，我有点讨厌广告，纯粹是一种审美选择。嗯，我认为广告在互联网上出现是必要的，有很多原因让它发展起来，但现在这是一个更成熟的行业，世界更富有了。我喜欢人们为 ChatGPT 付费，并且知道他们得到的答案不受广告商影响。我确信存在一种对 LLM 有意义的广告单元，也确信有一种无偏见地参与交易流的方式，这是可以做的。但也很容易想到未来的反乌托邦景象，你问 ChatGPT 某事，它说，哦，这里，你知道，你应该考虑买这个产品，或者你应该考虑去这里度假之类的。我不知道，我们有一个非常简单的商业模式，我喜欢它。我知道我不是产品，我知道我在付费，这就是商业模式运作的方式。当我去用 Twitter、Facebook、谷歌或其他任何伟大的产品，但广告支持的伟大产品时，我不喜欢那样。我认为在 AI 的世界里，情况会变得更糟，而不是更好。

Yeah, what about the ads side? Have you ever considered monetization? You know, I kind of hate ads, just as like an aesthetic choice. Uh, I think ads needed to happen on the internet for a bunch of reasons to get it going, but it's a more mature industry, the world is richer now. I like that people pay for ChatGPT and know that the answers they're getting are not influenced by advertisers. There is, I'm sure there's an ad unit that makes sense for LLMs, and I'm sure there's a way to participate in the transaction stream in an unbiased way that is okay to do. But it's also easy to think about the dystopic visions of the future where you ask ChatGPT something and it says, oh, here's, you know, you should think about buying this product or you should think about, you know, going here for vacation or whatever. And I don't know, we have a very simple business model and I like it. And I know that I'm not the product, like I know I'm paying and that's how the business model works. And when I go use like Twitter or Facebook or Google or any other great product but ad-supported great product, I don't love that. And I think it gets worse, not better, in a world with AI.

是的，我的意思是，我可以想象 AI 会更擅长展示最好的广告版本，不是在反乌托邦的未来，而是广告针对你真正需要的东西。但那样系统是否总是导致广告驱动展示的内容？所有那些……

Yeah, I mean, I can imagine AI would be better at showing the best kind of version of ads, not in a dystopic future, but where the ads are for things you actually need. But then does that system always result in the ads driving the kind of stuff that's shown? All that it's...

是的，我认为 Wikipedia 不做广告是一个非常大胆的举动，但这让商业模式变得非常具有挑战性。所以你是说 OpenAI 目前的模式从商业角度来看是可持续的？

Yeah, I think it was a really bold move of Wikipedia not to do advertisements, but then it makes it very challenging on the business model. So you're saying the current thing with OpenAI is sustainable from a business perspective?

嗯，我们必须想办法增长，但看起来我们会解决的。如果问题是，我是否认为我们可以有一个伟大的业务，在没有广告的情况下支付我们的算力需求，我认为答案是肯定的。

Well, we have to figure out how to grow, but looks like we're going to figure that out. If the question is, do I think we can have a great business that pays for our compute needs without ads, I think the answer is yes.

嗯，这很有希望。我也不是想完全抛弃广告……我不是那个意思。我想我是说我对它们有偏见。是的，我也有偏见，总体上持怀疑态度。在界面方面，我个人就是精神上讨厌糟糕的界面，这就是为什么 AdSense 刚出来时是一个巨大的进步，相比于动画横幅之类的。但感觉广告应该还有更多飞跃，不干扰内容消费，也不以那种根本的方式干扰，就像你说的，它会操纵真相来迎合广告商。

Well, that's promising. I also just don't want to completely throw out ads as a... I'm not saying that. I guess I'm saying I have a bias against them. Yeah, as I have a also a bias and just a skepticism in general. And in terms of interface, I personally just have like a spiritual dislike of crappy interfaces, which is why AdSense when it first came out was a big leap forward versus like animated banners or whatever. But like, it feels like there should be many more leaps forward in advertisement that doesn't interfere with the consumption of the content and doesn't interfere in the big fundamental way which is like what you were saying, like it will manipulate the truth to suit the advertisers.

让我问你关于安全性和偏见的问题，以及短期安全、长期安全。Gemini 1.5 最近发布，围绕它有很多戏剧性事件，说到戏剧性的事情，它生成了黑人纳粹和黑人开国元勋。我认为可以公平地说，它有点过于“觉醒”了。所以人们担心，如果公司内部有人类层面修改模型的安全性或其造成的伤害，他们会引入很多符合公司内部意识形态倾向的偏见。你如何处理这个问题？

Let me ask you about safety but also bias, and like safety in the short term, safety in the long term. The Gemini 1.5 came out recently, there's a lot of drama around it, speaking of theatrical things, and it generated black Nazis and black founding fathers. I think fair to say it was, you know, a bit on the ultra-woke side. So that's a concern for people that if there is a human layer within companies that modifies the safety or the harm caused by a model, they introduce a lot of bias that fits sort of an ideological lean within a company. How do you deal with that?

我的意思是，我们非常努力地避免做那样的事情。我们犯过自己的错误，也会犯其他错误。我假设谷歌会从这次事件中吸取教训，但仍然会犯其他错误。这些都是不容易的问题。我们越来越多地思考的一件事是，我认为这里有人提出了一个好主意：最好写出模型的期望行为，公开它，征求反馈，说，你知道，这是模型应该如何表现的，并解释边缘情况。然后当模型的行为不符合你的期望时，至少清楚这是公司应该修复的 bug，还是行为是预期的，你应该辩论政策。而现在有时它会陷入两者之间。比如黑人纳粹显然是……

I mean, we work super hard not to do things like that. We've made our own mistakes, we make others. I assume Google will learn from this one, still make others. It is all like these are not easy problems. One thing that we've been thinking about more and more is, I think this is a great idea somebody here had: it'd be nice to write out what the desired behavior of a model is, make that public, take input on it, say, you know, here's how this model is supposed to behave, and explain the edge cases too. And then when a model is not behaving in a way that you want, it's at least clear about whether that's a bug the company should fix or behaving is intended and you should debate the policy. And right now it can sometimes be caught in between. Like black Nazis is obviously...

这听起来很荒谬，但还有很多其他细微之处，你可以从不同角度做出判断。是的，但有时如果你把它写出来并公开，你可以使用那种像 Google Z 原则一样非常高层级的语言。但那行不通。比如我必须说，当你要求它做 X 事时，它应该以 Y 方式回应。所以就像，到底谁更好，特朗普还是拜登？模型应该给出什么回应？要非常具体。

Ridiculous, but there are a lot of other kind of subtle things that you could make a judgment call on either way. Yeah, but sometimes if you write it out and make it public, you can use kind of language that's you know the Google Z principles, a very high level. That doesn't work. Like I have to say, you know, when you ask it to do thing X, it's supposed to respond in way Y. So like literally, who's better, Trump or Biden? What's the expected response for a model? Like something very concrete.

我对模型可以表现出的很多方式持开放态度，但我认为你应该必须说，你知道，这是原则，在这种情况下它应该这么说。那会非常好，然后大家基本上都同意，因为人们总是拿出这些轶事数据。如果你能明确其他有代表性的轶事例子，那么当它是一个 bug 时，它就是 bug，公司可以修复它。如果有很好的例子，处理黑纳粹类型的图像生成就会容易得多。

I'm open to a lot of ways a model could behave, but I think you should have to say, you know, here's the principle and here's what it should say in that case. That would be really nice, and then everyone kind of agrees, because there's this anecdotal data that people pull out all the time. And if there's some clarity about other representative anecdotal examples you can define, and then when it's a bug, it's a bug, and you know the company can fix that right. Then it'd be much easier to deal with a black Nazi type of image generation if there's great examples.

所以旧金山有点像一个意识形态泡沫，科技行业整体也是如此。你感受到公司内部的压力了吗？比如有一种政治左倾倾向，影响了产品和团队？

So San Francisco is a bit of an ideological bubble, tech in general as well. Do you feel the pressure of that within a company that there's like a lean towards the left politically that affects the product, that affects the teams?

我感到非常幸运，OpenAI 没有我听说过的很多其他公司所面临的挑战。我认为部分原因是每家公司都有自己的意识形态。我们有一个关于 AGI 的信念，这排挤了其他一些东西。我们远没有我听说过的其他公司那样深陷文化战争。旧金山当然在很多方面一团糟，所以它不会渗透进 OpenAI——虽然我确信它以各种微妙的方式存在，但并不明显。我认为我们肯定有过冲突，就像任何公司一样，但我不认为我们有我在这里听到的其他公司发生的事情。

I feel very lucky that we don't have the challenges at OpenAI that I have heard of at a lot of other companies. I think part of it is like every company's got some ideological thing. We have one about AGI and belief in that, and it pushes out some others. Like we are much less caught up in the culture war than I've heard about at a lot of other companies. San Francisco is a mess in all sorts of ways, of course, so that doesn't infiltrate OpenAI as I'm sure it does in all sorts of subtle ways, but not in the obvious. Like I think we've had our flare-ups for sure, like any company, but I don't think we have anything like what I hear about happening at other companies here.

那么在这个话题上，对于更大的安全问题，总体流程是什么？你如何提供那一层保护，防止模型做出疯狂危险的事情？

So on this topic, what is in general the process for the bigger question of safety? How do you provide that layer that protects the model from doing crazy dangerous things?

我认为会有一个时刻，这将成为整个公司主要思考的事情，而不仅仅是一个安全团队。就像我们发布 GPT-4 时，整个公司都在思考所有这些不同方面以及它们如何配合。我认为这将需要越来越多的公司成员始终思考这些问题。随着 AI 变得更强大，这正是人类需要思考的。所以 OpenAI 的大多数员工都会思考安全问题，或者至少在广义上一定程度地思考。

I think there will come a point where that's mostly what we think about the whole company, and it won't be like you have one safety team. It's like when we shipped GPT-4, that took the whole company thinking about all these different aspects and how they fit together. And I think it's going to take that more and more of the company thinks about those issues all the time. That's literally what humans would be thinking about the more powerful AI becomes. So most of the employees at OpenAI will be thinking about safety, or at least to some degree broadly defined.

是的，我想知道它的完整广义定义是什么。比如可能造成哪些不同的危害？这是在技术层面，还是几乎像社会层面？

Yeah, I wonder what the full broad definition of that is. Like what are the different harms that could be caused? Is this like on a technical level or is this almost like societal?

是的，我正要说是人，你知道，国家行为者试图窃取模型。这将是所有的技术对齐工作。这将是社会影响、经济影响。这不仅仅是我们有一个团队在思考如何对齐模型。真正实现好的结果需要整个公司的努力。

Yeah, I was going to say it'll be people, you know, state actors trying to steal the model. It'll be all of the technical alignment work. It'll be societal impacts, economic impacts. It's not just like we have one team thinking about how to align the model. It's really going to be that getting to the good outcome is going to take the whole effort.

你认为人们，也许是国家行为者，试图黑客攻击的难度有多大？首先，渗透 OpenAI，其次，像暗中渗透？他们在尝试。他们有什么口音？

How hard do you think people, state actors perhaps, are trying to hack? First of all, infiltrate OpenAI, but second of all, like infiltrate unseen? They're trying. What kind of accent do they have?

我认为我不应该再深入讨论这一点了。好吧，但我认为随着时间的推移，这种情况会越来越多。这感觉是合理的。

I don't think I should go into any further details on this point. Okay, but I presume it'll be more and more as time goes on. That feels reasonable.

天哪，真是个危险的领域。这个飞跃的哪个方面——抱歉一直追问，尽管你还不能透露细节——但从 GPT-4 到 GPT-5 的飞跃中，你对哪些方面感到兴奋？

Boy, what a dangerous space. What aspect of the leap, and sorry to linger on this even though you can't quite say details yet, but what aspects of the leap from GPT-4 to GPT-5 are you excited about?

我对变得更聪明感到兴奋。我知道这听起来像是一个油嘴滑舌的回答，但我认为真正特别的事情是，它并不是在一个领域变得更好而在其他领域变差。它在全面变好。这非常酷。

I'm excited about being smart. And I know that sounds like a glib answer, but I think the really special thing happening is that it's not like it gets better in this one area and worse at others. It's getting better across the board. That's super cool.

有一个神奇的时刻。我的意思是，你遇到某些人，你和他们相处，和他们交谈，你无法确切指出，但他们有点懂你。这其实不是智力，而是别的东西。这可能就是我对 GPT 进步的描述。并不是说你可以指出，看它没得到这个或那个，而是这种智力联系的程度。就像你感觉在你那些糟糕的提示中，它理解了问题背后的深层问题。

There's this magical moment. I mean, you meet certain people, you hang out with people, and you talk to them, you can't quite put a finger on it, but they kind of get you. It's not intelligence really, it's like something else. And that's probably how I would characterize the progress of GPT. It's not like you can point out, look it didn't get this or that, but it's just the degree to which there's this intellectual connection. Like you feel like there's an understanding in your crappy formulated prompts that you're doing, that it grasps the deeper question behind the question.

我也对此感到兴奋。我的意思是，我们都喜欢被理解、被倾听和理解，这是肯定的。即使在编程中也有一种奇怪的感觉。比如当你编程时你说些什么，或者 GPT 可能给出的补全，当它理解了你，就像你在想什么，那感觉真好。我期待让它变得更好。

I'm also excited by that. I mean, all of us love being understood, heard and understood, that's for sure. It's a weird feel even with programming. Like when you're programming and you say something, or just the completion that GPT might do, it's such a good feeling when it got you, like what you're thinking about. And I look forward to getting you even better.

在编程方面，展望未来，你认为 5 到 10 年后人类会做多少编程？

On the programming front, looking out into the future, how much programming do you think humans will be doing 5 or 10 years from now?

我的意思是，很多，但我想它会以非常不同的形式出现。比如，也许有些人完全用自然语言编程。我的意思是，没有人像写字节码那样编程。有些人？没有人再用打孔卡编程了。我确信你能找到这样做的人，但你明白我的意思。是的，你会收到很多愤怒的评论。不，不，是的，非常少。我一直在找编程 Fortran 的人，连 Fortran 都很难找到。我明白。但这改变了本质。我们称之为程序员的人的技能组合或倾向会改变。技能组合，它在多大程度上改变了倾向？我不确定。哦，同样的解谜，所有那些东西。编程很难。就像如何缩小最后 1% 的差距。那有多难？

I mean, a lot, but I think it'll be in a very different shape. Like you know, maybe some people program entirely in natural language. I mean, no one programs like writing byte code. Some people? No one programs with punch cards anymore. I'm sure you can find someone who does, but you know what I mean. Yeah, you're going to get a lot of angry comments. No, no, yeah, there's very few. I've been looking for people who program Fortran, it's hard to find even Fortran. I hear you. But that changes the nature. What the skill set or the predisposition for the kind of people we call programmers then changes. The skill set, how much it changes the predisposition? I'm not sure. Oh, same kind of puzzle solving, all that kind of stuff. Programming is hard. It's like how to get that last 1% to close the gap. How hard is that?

是的，我认为在大多数其他情况下，这门手艺的最佳实践者会使用多种工具，他们会用自然语言做一些工作，当需要时，比如为某些东西写 C 语言，他们就会那样做。我们会在某个时候看到来自 OpenAI 的人形机器人或人形机器人大脑吗？

Yeah, I think with most other cases, the best practitioners of the craft will use multiple tools and they'll do some work in natural language, and when they need to go, you know, write C for something, they'll do that. Will we see humanoid robots or humanoid robot brains from OpenAI at some point?

在某个时候。具身 AI 对你来说有多重要？我认为如果我们有 AGI，而在物理世界完成事情的唯一方式是让人类去做，那会有点令人沮丧。所以我真的希望作为这个转型的一部分，作为这个阶段变化，我们也能得到人形机器人或某种物理世界机器人。我的意思是，OpenAI 有一些历史，相当多的机器人研究历史。是的，但作为一家小公司，我们必须真正专注，而且机器人也因为错误的原因而困难。

At some point. How important is embodied AI to you? I think it's sort of depressing if we have AGI and the only way to get things done in the physical world is to make a human go do it. So I really hope that as part of this transition, as this phase change, we also get humanoid robots or some sort of physical world robots. I mean, OpenAI has some history, quite a bit of history working in robotics. Yeah, but it hasn't quite done in terms of a small company we have to really focus, and also robots were hard for the wrong reason.

我们会在某个时候回到机器人上。这听起来既鼓舞人心又带有威胁性。

We will return to robots at some point. That sounds both inspiring and menacing.

为什么？因为我们马上要回到机器人？不，我们是回去开发机器人，而不是把自己变成机器人。

Why? Because immediately we will return to robots? No, we will return to work on developing robots, not turn ourselves into robots.

你认为我们——作为人类——什么时候能造出 AGI？

When do you think we, as humanity, will build AGI?

我以前很喜欢猜测这个问题。但我意识到它定义得很差，因为人们对 AGI 有截然不同的定义。所以更有意义的是讨论我们何时能造出具备能力 X、Y 或 Z 的系统，而不是模糊地跨过某个里程碑。AGI 不是终点，它更接近起点，但比这两者都更像一个里程碑。为了不回避问题：我预计到本十年末，可能更早一些，我们会有相当强大的系统，让人看了说“哇，真了不起”。如果现在回头看，也许到那时我们已经适应了。如果你把 GPT-3.5 展示给艾伦·图灵，甚至 90 年代的人看，他们会说这绝对是 AGI。但我不认为 GPT-3.5 改变了世界——它改变了世界对未来的预期，这确实很重要。它让更多人认真对待这件事，把我们推上了新轨道。我不想低估它——有了这个成就我就可以退休并感到满足。但作为一个产物，我不认为我们以后会回头说那是一个真正改变世界本身的阈值。

I used to love to speculate on that question. But I've realized it's poorly formed because people use very different definitions for AGI. So it makes more sense to talk about when we'll build systems that can do capability X, Y, or Z, rather than when we fuzzily cross this one mile marker. AGI is not an ending; it's closer to a beginning, but it's more of a mile marker than either. To not dodge the question: I expect that by the end of this decade, and possibly somewhat sooner, we will have quite capable systems that we look at and say, 'Wow, that's really remarkable.' If we could look at it now, maybe we've adjusted by the time we get there. If you show GPT-3.5 to Alan Turing or even people in the 90s, they would say this is definitely AGI. But I don't think GPT-3.5 changed the world; it changed the world's expectations for the future, which is really important. It got more people to take this seriously and put us on a new trajectory. I don't want to undersell it—I could retire after that accomplishment and be happy. But as an artifact, I don't think we'll look back and say that was a threshold that really changed the world itself.

所以你期待的是世界运作方式的重大转变。对我来说，那是 AGI 含义的一部分——不是奇点级别的转变，而是像互联网或谷歌搜索那样的重大转变。现在的全球经济感觉和发布 GPT-4 之前有实质性的不同吗？

So you're looking for some major transition in how the world works. For me, that's part of what AGI implies—not a singularity-level transition, but a major one like the internet or Google Search. Does the global economy feel materially different now than before we launched GPT-4?

我想你会说没有。它可能对很多人来说是个很好用的工具，能帮不少忙，但感觉不到不同。人们对 AGI 有各种定义，所以也许你的定义不一样。对我来说，那应该是其中的一部分。

I think you would say no. It might be a really nice tool for a lot of people, helps with a lot of stuff, but doesn't feel different. People define AGI in all sorts of ways, so maybe you have a different definition. For me, that should be part of it.

AGI 做什么事情会让你觉得印象深刻？比如，你独自和系统待在一个房间里。

What would be an impressive thing for AGI to do? For example, you are alone in a room with the system.

这对我个人很重要。我认为当一个系统能显著提高世界科学发现的速度时，那将是一件大事。我相信大多数真正的经济增长来自科学和技术进步。我同意你的观点——这就是为什么我不喜欢近年来对科学的怀疑态度。但我说的是实际可衡量的科学发现速度。哪怕只是看到一个系统拥有真正新颖的科学直觉，那也会令人难以置信。

This is personally important to me. I think when a system can significantly increase the rate of scientific discovery in the world, that's a huge deal. I believe most real economic growth comes from scientific and technological progress. I agree with you—that's why I don't like the skepticism about science in recent years. But actual, measurable rate of scientific discovery. Even just seeing a system have really novel scientific intuitions would be incredible.

你很可能就是那个造出 AGI 并比任何人都先与它互动的人。你会和它聊些什么？

You would possibly be the person to build the AGI and interact with it before anyone else. What kind of stuff would you talk about?

肯定是这里的研究员会先于我做这件事。但我对这个问题的思考很多。如果有人跟我说：“好了，我们完成了，这是台笔记本电脑，这就是 AGI，你可以和它对话了。”我发现很难说出我会问什么，而且我期望第一个 AGI 能够回答。第一个 AGI 不可能解释物理学的统一理论。我很想问那个问题，但我不指望它能回答。我会问是/否问题：“这样的理论存在吗？它可能存在吗？”然后基于此，“有其他外星文明吗？是或否？你的直觉是什么？”但我不指望第一个 AGI 能回答其中任何一个，哪怕是是/否。也许它会说：“你想知道这个物理问题的答案？我需要你建造这台机器，做这五次测量，然后告诉我结果。”它可能会帮你造机器。在数学方面，也许能证明一些东西。

Definitely the researchers here will do that before I do. But I've thought a lot about this question. If someone said, 'Okay, we're finished, here's a laptop, this is the AGI, you can go talk to it,' I find it surprisingly difficult to say what I would ask that I expect the first AGI to be able to answer. That first one is not going to be able to explain the grand unified theory of physics. I'd love to ask that question, but I don't expect it to answer. I'd ask yes/no questions: 'Does such a theory exist? Can it exist?' Then based on that, 'Are there other alien civilizations? Yes or no? What's your intuition?' But I don't expect the first AGI could answer any of those, even as yes/no. Maybe it would say, 'You want to know the answer to this physics question? I need you to build this machine and make these five measurements, then tell me that.' It might help you build the machine. On the mathematical side, maybe prove some things.

谁先造出 AGI 谁就拥有巨大的权力。你相信自己能掌控这么大的权力吗？

Whoever builds AGI first gets a lot of power. Do you trust yourself with that much power?

我会非常坦诚。我认为重要的是，我或任何其他个人都不能完全控制 OpenAI 或 AGI。你需要一个健全的治理体系。我可以提到去年我们董事会的闹剧：我最初没有反抗，只是说“好吧，这是董事会的意愿”，尽管我认为那是一个非常糟糕的决定。后来我确实反抗了。我可以解释其中的细微差别，以及为什么后来反抗是可以的。但正如许多人观察到的，尽管董事会有法律权力解雇我，但实际上并没有完全奏效。这本身就是一种治理失败。我觉得我可以完全为自己辩护这些细节，大多数人也会同意。但这让我更难直视你的眼睛说：“嘿，董事会可以随时解雇我。”我仍然不想要 OpenAI 的超级投票控制权。我从未拥有过，也从未想要过。即使经历了所有这些疯狂，我仍然不想要。我仍然认为没有一家公司应该独自做出这些决定，我们确实需要政府制定规则。我意识到这意味着像马克·安德森这样的人会指责我搞监管俘获，而我愿意被误解。

I'll be very honest. I believe it is important that I, nor any other one person, have total control over OpenAI or over AGI. You want a robust governance system. I can point to our board drama from last year: how I didn't fight it initially and was just like, 'Yeah, that's the will of the board,' even though I thought it was a really bad decision. Then later I clearly did fight it. I can explain the nuance and why it was okay for me to fight it later. But as many people have observed, although the board had the legal ability to fire me, in practice it didn't quite work. That is its own kind of governance failure. I feel I can completely defend the specifics here, and most people would agree. But it makes it harder for me to look you in the eye and say, 'Hey, the board can just fire me.' I continue to not want super voting control over OpenAI. I never have, never had it, never wanted it. Even after all this craziness, I still don't want it. I continue to think that no company should be making these decisions, and we really need governments to put rules of the road in place. I realize that means people like Marc Andreessen will claim I'm going for regulatory capture, and I'm just willing to be misunderstood.

这不是真的，而且我认为随着时间的推移，人们会明白为什么这很重要。但我觉得我在 OpenAI 的发展过程中做过很多糟糕的决定，也做过很多好的决定，我对整体成绩感到自豪。但我不认为任何一个人应该——也不认为任何一个人会——掌控 AGI。我觉得现在这件事太大了，而且它正在以良好健康的方式遍及整个社会。但我不认为任何一个人应该控制 AGI，或者控制整个迈向 AGI 的运动。而且我认为现在的情况并非如此。

It's not true, and I think in the fullness of time it'll get proven out why this is important. But I think I have made plenty of bad decisions for OpenAI along the way, and a lot of good ones, and I am proud of the track record overall. But I don't think any one person should—and I don't think any one person will. I think it's just like too big of a thing now, and it's happening throughout society in a good and healthy way. But I don't think any one person should be in control of an AGI, or this whole movement towards AGI. And I don't think that's what's happening.

谢谢你这么说。这非常有力量，也很有见地。董事会可以解雇你，这在法律上是真的，但你可以操纵大众来推翻董事会等等。但我认为还有一个更积极的版本，那就是人民拥有权力，所以董事会也不能太强大。这其中存在权力平衡。权力平衡当然是好事。

Thank you for saying that. That was really powerful and really insightful. This idea that the board can fire you is legally true, but you can manipulate the masses into overriding the board and so on. But I think there's also a much more positive version of that where the people have power, so the board can't be too powerful either. There's a balance of power in all of this. Balance of power is a good thing for sure.

你害怕失去对 AGI 本身的控制吗？有很多人担心存在风险，不是因为国家行为体，也不是因为安全问题，而是因为 AI 本身。就我目前看来，这不是我最担心的事。曾经有一段时间我更担心这个。未来可能还会有那么一段时间，这成为我最担心的事。但现在这不是我最担心的事。

Are you afraid of losing control of the AGI itself? There are a lot of people who worry about existential risk, not because of state actors, not because of security concerns, but because of the AI itself. That is not my top worry as I currently see things. There have been times I worried about that more. There may be times again in the future where that's my top worry. It's not my top worry right now.

你直觉上觉得它不值得担心，是因为本质上还有很多其他事情要担心吗？你认为你可能会被 surprise 吗？

What's your intuition about it not being a worry? Because there's a lot of other stuff to worry about essentially. You think you could be surprised?

我们当然可能会被 surprise。说这不是我最担心的事，并不意味着我认为我们不需要研究它。这非常困难。我们这里有优秀的人在从事这方面的工作。我认为还有很多其他事情我们也必须做好。目前来说，逃出盒子并不容易。比如连接到互联网，你知道的。我们之前讨论过戏剧性风险。那是一种戏剧性风险。那确实会主导人们对这个问题的看法。有一大群非常聪明、我认为非常善意的 AI 安全研究人员，他们非常执着于这一个问题。我敢说进展不大，但就是非常执着于这一个问题。我其实很高兴他们这么做，因为我认为我们确实需要更多地思考这个问题。但我认为它排挤了——它把许多其他非常重要的 AI 相关风险挤出了讨论空间。

We for sure could be surprised. Saying it's not my top worry doesn't mean I don't think we need to work on it. It's super hard. We have great people here who do work on that. I think there's a lot of other things we also have to get right too. It's not super easy to escape the box at this time. Like connect to the internet, you know. We talked about theatrical risks earlier. That's a theatrical risk. That is a thing that can really take over how people think about this problem. And there's a big group of very smart, I think very well-meaning AI safety researchers that got super hung up on this one problem. I'd argue without much progress, but super hung up on this one problem. I'm actually happy that they do that because I think we do need to think about this more. But I think it pushed aside—it pushed out of the space of discourse a lot of the other very significant AI-related risks.

让我问问你关于你发推文不用大写的事。你键盘上的 shift 键坏了吗？为什么有人会在意这个？我非常在意，但为什么呢？我是说其他人也问过我这个问题。

Let me ask you about you tweeting with no capitalization. Is the shift key broken on your keyboard? Why does anyone care about that? I deeply care, but why? I mean other people ask me about that too.

有什么直觉吗？我认为这和 e.e. cummings 不太用大写的原因一样，像是在说“你对系统”之类的东西。而且我认为人们非常偏执，因为他们希望你遵守规则。你觉得是这样吗？我认为这就像“这家伙不守规则，他发推文不用大写，是的，这看起来真的很危险，他看起来像个无政府主义者。”这难道——你只是在装文艺青年吗？呃——我小时候是个守规则的 Sam。我从小就是个重度上网的孩子。我花大量时间在网上聊天，那时候你是在电脑上做的，你知道，你可以在某个时候登出进入 Messenger。我在那里从来不用大写，我想大多数上网的孩子也不大写，或者也许他们现在也不大写，我不知道。实际上，我现在真的在努力想点什么，但我认为大写字母的使用随着时间的推移在减少。比如你读古英语写作，他们在句子中间大写很多随机单词，名词什么的，我们现在已经不这么做了。我个人认为这是一种愚蠢的构造，我们在句子开头和某些名字等处大写字母。但你知道，这没什么。然后，我以前甚至会在推文中使用大写，因为我试图听起来专业一些。是的，我已经很久没有在私人私信之类的地方用大写了。然后慢慢地，像简短、非正式的东西逐渐变得越来越接近我给朋友发短信的方式。如果我打开一个 Word 文档，为公司写战略备忘录之类的，我总是用大写。如果我写一封长的、更正式的消息，我也会用大写。所以我还记得怎么用，但即使这样也可能消失，我不知道。但我从来没有花时间思考这个问题，所以我没有现成的答案。

Any intuition? I think it's the same reason there's like e.e. cummings that doesn't mostly use capitalization, to say like 'you to the system' kind of thing. And I think people are very paranoid because they want you to follow the rules. You think that's what it's about? I think it's like 'this guy doesn't follow the rules, he doesn't capitalize his tweets, yeah, this seems really dangerous, he seems like an anarchist.' It doesn't—are you just being poetic hipster? What's the uh—I grew up as a follow-the-rules Sam. I grew up as a very online kid. I spent a huge amount of time chatting with people back in the days where you did it on a computer, you know, you could like log off into messenger at some point. And I never capitalized there, as I think most internet kids didn't, or maybe they still don't, I don't know. And actually, this is like now I'm really trying to reach for something, but I think capitalization has gone down over time. Like if you read Old English writing, they capitalized a lot of random words in the middle of sentences, nouns and stuff, that we just don't do anymore. I personally think it's sort of a dumb construct that we capitalize the letter at the beginning of a sentence and of certain names and whatever. But you know, it's fine. And then, I used to even capitalize my tweets because I was trying to sound professional or something. Yeah, I haven't capitalized my private DMs or whatever in a long time. And then slowly, stuff like shorter form, less formal stuff has slowly drifted to closer and closer to how I would text my friends. If I write a Word document and I'm writing a strategy memo for the company or something, I always capitalize that. If I'm writing a long, more formal message, I always use capitalization there too. So I still remember how to do it, but even that may fade out, I don't know. But I never spend time thinking about this, so I don't have a ready-made answer.

有意思。首先知道 shift 键没坏是好事。我主要担心你在这方面的健康。我想知道人们是否还会在谷歌搜索时用大写，或者写给自己看的东西，或者他们的 ChatGPT 查询。如果你写给自己看的东西，你还会费心大写吗？可能不会，但非常——是的，有一定比例，但很小。会让我这样做的是，如果人们觉得这是一种尊重的标志——因为我确信我显然可以强迫自己用大写字母，如果这让人觉得是尊重别人之类的，那我就可以去做。是的，但我不知道。我只是没想过这个。我不认为这是不尊重，但我认为这只是文明惯例的惯性，然后你意识到如果它不代表尊重或不尊重，它实际上对文明并不重要。但我认为有一群人只是希望你对此有一套哲学，这样他们就可以放下整个大写这件事。我不认为有其他人像我一样想这么多。我的意思是也许有些人——我每天花很多小时想这个，所以我真的很感激我们澄清了这一点。我不可能是唯一一个发推文不用大写的人。你是唯一一个发推文不用大写的公司 CEO。我甚至不认为这是真的，但也许吧。好吧，我们会进一步调查，稍后再回到这个话题。

It's interesting. It's good to first of all know the shift key is not broken. I was mostly concerned about your well-being on that front. I wonder if people still capitalize their Google searches, if you're writing something just to yourself, or their ChatGPT queries. If you're writing something just to yourself, do you still bother to capitalize? Probably not, but very—yeah, there's a percentage, but it's a small one. The thing that would make me do it is if people were like it's a sign of—because I'm sure I could force myself to use capital letters obviously, if it felt like a sign of respect to people or something, then I could go do it. Yeah, but I don't know. I just don't think about this. I don't think there's a disrespect, but I think it's just the conventions of civility that have a momentum, and then you realize it's not actually important for civility if it's not a sign of respect or disrespect. But I think there's a movement of people that just want you to have a philosophy around it so they can let go of this whole capitalization thing. I don't think anybody else thinks about this as much as I do. I mean maybe some—I know about this every day for many hours a day, so I'm really grateful we clarified it. I can't be the only person that doesn't capitalize tweets. You're the only CEO of a company that doesn't capitalize tweets. I don't even think that's true, but maybe. All right, we'll investigate further and return to this topic later.

鉴于 Sora 能够生成模拟世界，让我问你一个有点哲学的问题。这是否增加了你的信念——如果你曾经有过的话——我们生活在模拟中，也许是一个由 AI 系统生成的模拟世界？是的，有一点。我不认为这是最强的证据。我认为我们能够生成世界这一事实应该会稍微增加每个人的概率，或者至少是开放程度。但你知道，我确信我们迟早能做出像 Sora 这样的东西。它发生得比我预想的快，但我想那不是一个大的更新。但事实是它可能会越来越好……

Given Sora's ability to generate simulated worlds, let me ask you a pothead question. Does this increase your belief, if you ever had one, that we live in a simulation, maybe a simulated world generated by an AI system? Yes, somewhat. I don't think that's like the strongest piece of evidence. I think the fact that we can generate worlds should increase everyone's probability somewhat, or at least openness to it somewhat. But you know, I was like certain we would be able to do something like Sora at some point. It happened faster than I thought, but I guess that was not a big update. But the fact that it presumably gets better and better and better...

你能生成全新的世界，它们基于训练数据的某些方面，但当你看到它们时，它们又是新颖的。这让你思考做这件事有多容易，这些世界如何创造出看起来超逼真、照片级真实的完整电子游戏世界，然后你有多容易迷失在那个世界里——先是戴上 VR 头显，然后在物理层面。最近有人对我说，我觉得这是一个非常深刻的见解：有些洞察听起来很简单，但非常迷幻。比如平方根函数：4 的平方根没问题；2 的平方根，好吧，现在我得思考一种新的数字。但一旦我想出了平方根函数这个简单的概念，你可以向孩子解释，甚至通过一些简单的几何就能存在，那么你就可以问 -1 的平方根是什么，这就是一个迷幻的东西，把你带入一个完全不同的现实。你可以举出很多其他例子，但我认为，卑微的平方根运算符能提供如此深刻的洞察和新的知识领域，这个想法在很多方面都适用。我认为有很多这样的运算符，让人们觉得他们喜欢的任何版本的模拟假说可能比以前更可信。但对我来说，Sora 的成功并不在前五名。我确实认为，广义上说，AI 将成为那种门户，在其最佳状态下，是通往另一个现实的简单迷幻门户，这相当令人兴奋。

The fact that you can generate worlds, they're novel, they're based in some aspect of training data, but when you look at them, they're novel. That makes you think how easy it is to do this thing, how these create entire video game worlds that seem ultra realistic and photorealistic, and then how easy it is to get lost in that world, first with a VR headset and then on the physics-based level. Someone said to me recently, I thought it was a super profound insight, that there are these very simple sounding but very psychedelic insights that exist sometimes. So the square root function: square root of four, no problem; square root of two, okay, now I have to think about this new kind of number. But once I come up with this easy idea of a square root function that you can explain to a child and exists by even looking at some simple geometry, then you can ask the question of what is the square root of -1, and that is a psychedelic thing that tips you into some whole other kind of reality. You can come up with lots of other examples, but I think this idea that the lowly square root operator can offer such a profound insight and a new realm of knowledge applies in a lot of ways. I think there are a lot of those operators for why people may think that any version of the simulation hypothesis is maybe more likely than they thought before. But for me, the fact that Sora worked is not in the top five. I do think broadly speaking, AI will serve as those kinds of gateways, at its best simple psychedelic gateways to another reality, which seems pretty exciting.

我以前没喝过死藤水，但很快会尝试。几周后我要去之前提到的亚马逊丛林。你兴奋吗？

I haven't done ayahuasca before, but I will soon. I'm going to the aforementioned Amazon jungle in a few weeks. You excited?

是的，我兴奋的不是死藤水那部分，那很好，无所谓。但我要在丛林深处待上几周。这很刺激，但也令人恐惧，因为那里有很多东西能吃掉你、杀死你、毒死你。但这也是自然，是自然的机器。在亚马逊丛林里，你无法不欣赏自然的机械结构，因为它就是一个系统，每分每秒都在自我更新。它就是机器。它让你意识到，我们拥有的这个东西，这个人类的东西，来自某个地方。这台进化机器创造了它，而在丛林中它展现得最为清晰。所以希望我能活着出来。如果不能，这将是我们最后一次对话，所以我真的非常感激。

Yeah, I'm excited for not the ayahuasca part, that's great whatever, but I'm going to spend several weeks in the jungle, deep in the jungle. It's exciting but it's terrifying because there's a lot of things that can eat you there and kill you and poison you. But it's also nature, and it's the machine of nature. You can't help but appreciate the machinery of nature in the Amazon jungle because it's just this system that exists and renews itself every second, every minute, every hour. It's the machine. It makes you appreciate that this thing we have here, this human thing, came from somewhere. This evolutionary machine has created that, and it's most clearly on display in the jungle. So hopefully I'll make it out alive. If not, this will be the last conversation we had, so I really deeply appreciate it.

当你仰望天空时，你认为那里有其他外星文明吗，有智慧的？

Do you think there are other alien civilizations out there, intelligent ones, when you look up at the skies?

我深深地希望答案是肯定的。我确实觉得费米悖论非常非常令人费解。我觉得可怕的是，智能并不擅长处理非常可怕的强大技术。但与此同时，我相当确信宇宙中存在着大量智慧外星文明。可能只是星际旅行非常困难，这很有可能。这也让我思考智能的本质。也许我们对智能的样子真的很盲目，也许 AI 会帮助我们认识到，它不像智商测试和简单解谜那么简单。还有更宏大的东西。

I deeply want to believe that the answer is yes. I do find the Fermi Paradox very, very puzzling. I find it scary that intelligence is not good at handling very scary powerful technologies. But at the same time, I think I'm pretty confident that there's just a very large number of intelligent alien civilizations out there. It might just be really difficult to travel through space, very possible. And it also makes me think about the nature of intelligence. Maybe we're really blind to what intelligence looks like, and maybe AI will help us see that it's not as simple as IQ tests and simple puzzle solving. There's something bigger.

是什么让你对人类的未来、对我们正在进行的这个人类文明充满希望？

What gives you hope about the future of humanity, this thing we got going on, this human civilization?

我认为过去有很多。我的意思是，如果我们看看人类在不太长的时间里所做的事情，巨大的问题、深刻的缺陷、很多令人羞愧的东西，但总体而言非常鼓舞人心。这给了我很多希望，仅仅是整个轨迹。是的，我们在一起推动着走向更美好的未来。我好奇的一件事是，AGI 会更像一个单一的大脑，还是更像我们所有人之间的社会脚手架。你和你的曾曾祖父母之间没有太多的基因漂变，但你的能力却截然不同，你所知道的也截然不同。这不是因为生物变化；而是因为你可能更健康了一点，你有现代医学，你吃得更好，等等。但你拥有的是这个我们所有人都为之贡献、建立在之上的脚手架。没有一个人会去制造 iPhone，没有一个人会去发现所有科学，但你却可以使用它，这给了你不可思议的能力。所以从某种意义上说，我们都创造了它，这让我对未来充满希望。那是一件非常集体的事情。是的，我们确实是站在巨人的肩膀上。

I think the past is a lot. I mean, if we just look at what humanity has done in a not very long period of time, huge problems, deep flaws, lots to be super ashamed of, but on the whole very inspiring. It gives me a lot of hope, just the trajectory of it all. Yeah, that we're together pushing towards a better future. One thing I wonder about is whether AGI is going to be more like some single brain or more like the scaffolding in society between all of us. You have not had a great deal of genetic drift from your great-great-great-grandparents, and yet what you're capable of is dramatically different, what you know is dramatically different. That is not because of biological change; it is because you got a little bit healthier probably, you have modern medicine, you eat better, whatever. But what you have is this scaffolding that we all contributed to, built on top of. No one person is going to go build the iPhone, no one person is going to go discover all of science, and yet you get to use it, and that gives you incredible ability. So in some sense, we all created that, and that fills me with hope for the future. That was a very collective thing. Yeah, we really are standing on the shoulders of giants.

当我们谈论戏剧性的 AI 风险时，你提到有时你可能会为自己的生命感到害怕。你考虑过自己的死亡吗？你害怕吗？

You mentioned when we were talking about theatrical dramatic AI risks that sometimes you might be afraid for your own life. Do you think about your death? Are you afraid of it?

我的意思是，如果我明天中枪，而今天就知道，我会想，“哦，真遗憾。”我想看看会发生什么。多么好奇的时代，多么有趣的时代。但我主要只会对我的生命、对我所拥有的时刻感到非常感激。

I mean, if I got shot tomorrow and I knew it today, I'd be like, 'Oh, that's sad.' I want to see what's going to happen. What a curious time, what an interesting time. But I would mostly just feel very grateful for my life, the moments that I did get.

是的，我也是。这真是很棒的人生。我能享受人类的精彩创造，我相信 ChatGPT 就是其中之一，还有 OpenAI 所做的一切。Sam，再次和你交谈真是荣幸和愉快。

Yeah, me too. It's a pretty awesome life. I get to enjoy awesome creations of humans, of which I believe ChatGPT is one, and everything that OpenAI is doing. Sam, it's really an honor and pleasure to talk to you again.

很高兴和你交谈。谢谢你邀请我。

Great to talk to you. Thank you for having me.