The Gentle Singularity 温柔的奇点
We are past the event horizon1; the takeoff has started. Humanity is close to building digital superintelligence, and at least so far it’s much less weird than it seems like it should be. 我们已经越过了事件视界；起飞已经开始。人类即将建造出数字超级智能，至少到目前为止，它远比我们想象的要平淡无奇。
Robots are not yet walking the streets, nor are most of us talking to AI all day. People still die of disease, we still can’t easily go to space, and there is a lot about the universe we don’t understand. 机器人尚未走上街头，大多数人也没有整天与 AI 对话。人们仍会因病去世，我们仍无法轻松进入太空，宇宙中还有许多我们无法理解的事物。
And yet, we have recently built systems that are smarter than people in many ways, and are able to significantly amplify the output of people using them. The least-likely part of the work is behind us; the scientific insights that got us to systems like GPT-4 and o3 were hard-won, but will take us very far. 然而，我们最近构建的系统在许多方面已经超越了人类智慧，并能显著提升使用者的工作效率。最不可能实现的环节已成为过去；像 GPT-4 和 o3 这样的系统背后是来之不易的科学洞见，但它们将引领我们走得更远。
AI will contribute to the world in many ways, but the gains to quality of life from AI driving faster scientific progress and increased productivity will be enormous; the future can be vastly better than the present. Scientific progress is the biggest driver of overall progress; it’s hugely exciting to think about how much more we could have. 人工智能将以多种方式为世界做出贡献，但由 AI 推动的科学进步加速和生产力提升所带来的生活质量改善将是巨大的；未来可能比现在好得多。科学进步是所有进步的最大驱动力；想到我们能拥有的更多可能性，就令人无比振奋。
In some big sense, ChatGPT is already more powerful than any human who has ever lived. Hundreds of millions of people rely on it every day and for increasingly important tasks; a small new capability can create a hugely positive impact; a small misalignment2 multiplied by hundreds of millions of people can cause a great deal of negative impact. 从某种重大意义上说，ChatGPT 已经比历史上任何人类都更强大。每天有数亿人依赖它完成越来越重要的任务；一个小小的新功能就能产生极其积极的影响；而一个轻微的偏差乘以数亿人使用，也可能造成巨大的负面影响。
2025 has seen the arrival of agents that can do real cognitive work; writing computer code will never be the same. 2026 will likely see the arrival of systems that can figure out novel insights. 2027 may see the arrival of robots that can do tasks in the real world. 2025 年将迎来能够真正从事认知工作的智能体；编写计算机代码的方式将永远改变。2026 年可能会迎来能够发现新颖见解的系统。2027 年或将出现能在现实世界中执行任务的机器人。
A lot more people will be able to create software, and art. But the world wants a lot more of both, and experts will probably still be much better than novices, as long as they embrace the new tools. Generally speaking, the ability for one person to get much more done in 2030 than they could in 2020 will be a striking change, and one many people will figure out how to benefit from. 将会有更多人能够开发软件和创作艺术。但世界对这两者的需求远未饱和，只要专家们善于运用新工具，他们的水平仍将远高于新手。总体而言，到 2030 年，个人所能完成的工作量将比 2020 年有显著提升——这种剧变将使许多人找到获益之道。
In the most important ways, the 2030s may not be wildly different. People will still love their families, express their creativity, play games, and swim in lakes. 就最本质的层面而言，2030 年代或许不会天翻地覆。人们依然会深爱家人、释放创意、享受游戏、在湖中畅游。
But in still-very-important-ways, the 2030s are likely going to be wildly different from any time that has come before. We do not know how far beyond human-level intelligence we can go, but we are about to find out. 但在同样至关重要的维度上，2030 年代很可能迥异于过往任何时代。我们尚不清楚人工智能会超越人类智慧多远，但答案即将揭晓。
In the 2030s, intelligence and energy—ideas, and the ability to make ideas happen—are going to become wildly abundant. These two have been the fundamental limiters on human progress for a long time; with abundant intelligence and energy (and good governance), we can theoretically have anything else. 2030 年代，智力与能源——即构想能力与实现构想的能力——将呈现爆发式增长。这两者长期制约着人类发展进程；在充沛的智力、能源（以及良好治理）支撑下，理论上我们可以实现任何其他目标。
Already we live with incredible digital intelligence, and after some initial shock, most of us are pretty used to it. Very quickly we go from being amazed that AI can generate a beautifully-written paragraph to wondering when it can generate a beautifully-written novel; or from being amazed that it can make live-saving medical diagnoses to wondering when it can develop the cures; or from being amazed it can create a small computer program to wondering when it can create an entire new company. This is how the singularity goes: wonders become routine, and then table stakes. 如今我们已与不可思议的数字智能共生共存，经历最初震撼后，大多数人已习以为常。我们以惊人的速度从对 AI 能生成优美段落的惊叹，转为期待它何时能创作出优美小说；从对它能做出救命的医学诊断的惊奇，到期盼它何时能研发治愈方案；从对它能编写小程序感到不可思议，到渴望它何时能创立一家全新公司。这就是奇点降临的方式：奇迹变为常态，继而成为标配。
We already hear from scientists that they are two or three times more productive than they were before AI. Advanced AI is interesting for many reasons, but perhaps nothing is quite as significant as the fact that we can use it to do faster AI research. We may be able to discover new computing substrates, better algorithms, and who knows what else. If we can do a decade’s worth of research in a year, or a month, then the rate of progress will obviously be quite different. 科学家们已表示，在 AI 辅助下他们的工作效率比从前提升了两到三倍。先进 AI 之所以意义非凡，很大程度上在于它能加速 AI 研究本身。我们或许能借此发现新型计算基质、更优算法，以及难以预见的突破。若能将十年研究浓缩为一年甚至一个月，进步的速度显然将截然不同。
From here on, the tools we have already built will help us find further scientific insights and aid us in creating better AI systems. Of course this isn’t the same thing as an AI system completely autonomously updating its own code, but nevertheless this is a larval version of recursive self-improvement. 从此刻起，我们已经构建的工具将帮助我们获得更深入的科学研究洞见，并助力我们打造更优秀的人工智能系统。这虽然与人工智能系统完全自主更新其代码不是一回事，但已然是递归式自我改进的雏形。
There are other self-reinforcing loops at play. The economic value creation has started a flywheel of compounding infrastructure buildout to run these increasingly-powerful AI systems. And robots that can build other robots (and in some sense, datacenters that can build other datacenters) aren’t that far off. 其他自我强化的循环也在发挥作用。经济价值创造已经启动了一个飞轮效应，加速建设支撑这些日益强大人工智能系统的复合型基础设施。而能够制造其他机器人的机器人（某种意义上，数据中心也能复制自身）已不再遥远。
If we have to make the first million humanoid robots the old-fashioned way, but then they can operate the entire supply chain—digging and refining minerals, driving trucks, running factories, etc.—to build more robots, which can build more chip fabrication facilities, data centers, etc, then the rate of progress will obviously be quite different. 倘若我们不得不以传统方式制造最初的百万台人形机器人，之后它们便能接管整个供应链——开采冶炼矿物、驾驶卡车、运营工厂等等——来制造更多机器人，这些机器人继而可以建设更多的芯片制造工厂、数据中心等，那么进步的速度显然会大不相同。
As datacenter production gets automated, the cost of intelligence should eventually converge to near the cost of electricity. (People are often curious about how much energy a ChatGPT query uses; the average query uses about 0.34 watt-hours, about what an oven would use in a little over one second, or a high-efficiency lightbulb would use in a couple of minutes. It also uses about 0.000085 gallons of water; roughly one fifteenth of a teaspoon.) 随着数据中心生产自动化，智能的成本最终应趋近于接近电力的成本。（人们常好奇 ChatGPT 每次查询消耗多少能量；平均每次查询约耗电 0.34 瓦时，相当于烤箱运行略超一秒的能耗，或是高效节能灯泡几分钟的用电量。此外每次查询消耗约 0.000085 加仑水，大约十五分之一茶匙的量。）
The rate of technological progress will keep accelerating, and it will continue to be the case that people are capable of adapting to almost anything. There will be very hard parts like whole classes of jobs going away, but on the other hand the world will be getting so much richer so quickly that we’ll be able to seriously entertain new policy ideas we never could before. We probably won’t adopt a new social contract all at once, but when we look back in a few decades, the gradual changes will have amounted to something big. 技术进步的速度将持续加快，人类几乎能适应任何变化的特性仍将延续。尽管会出现诸如整类工作岗位消失等严峻挑战，但另一方面，世界财富将以惊人速度增长，使我们能够认真探讨从前根本无法实施的新政策构想。我们或许不会一次性采纳全新的社会契约，但几十年后回望，这些渐进变化将累积成重大变革。
If history is any guide, we will figure out new things to do and new things to want, and assimilate new tools quickly (job change after the industrial revolution is a good recent example). Expectations will go up, but capabilities will go up equally quickly, and we’ll all get better stuff. We will build ever-more-wonderful things for each other. People have a long-term important and curious advantage over AI: we are hard-wired to care about other people and what they think and do, and we don’t care very much about machines. 如果历史可资借鉴，我们将迅速适应新工具并找到新的追求与渴望（工业革命后的职业变迁就是个近例）。期望会提升，但能力提升的速度同样快，我们终将获得更好的东西。人类将持续为彼此创造愈发美妙的事物。相比人工智能，人类拥有一个长期重要且独特的优势：我们天生在意他人及其所思所为，而对机器则漠不关心。
A subsistence farmer from a thousand years ago would look at what many of us do and say we have fake jobs, and think that we are just playing games to entertain ourselves since we have plenty of food and unimaginable luxuries. I hope we will look at the jobs a thousand years in the future and think they are very fake jobs, and I have no doubt they will feel incredibly important and satisfying to the people doing them. 千年前的生存型农民若看到现代人的工作，会认为我们在从事虚假职业，觉得我们只因食物充足且享有难以想象的奢侈品，便终日玩游戏自娱。我期待未来千年后的人们回看今日职业时，也会觉得它们非常虚假，而毫无疑问，那些工作对当时从事者而言，必会显得无比重要且令人满足。
The rate of new wonders being achieved will be immense. It’s hard to even imagine today what we will have discovered by 2035; maybe we will go from solving high-energy physics one year to beginning space colonization the next year; or from a major materials science breakthrough one year to true high-bandwidth brain-computer interfaces the next year. Many people will choose to live their lives in much the same way, but at least some people will probably decide to “plug in”. 新奇迹涌现的速度将超乎想象。如今的我们甚至难以构想 2035 年会有怎样的发现——或许某年刚解决高能物理难题，次年就开启太空殖民；也许某年取得材料科学重大突破，次年就实现真正的高带宽脑机接口。许多人会选择延续原有的生活方式，但至少有一部分人很可能会决定“接入”数字世界。
Looking forward, this sounds hard to wrap our heads around. But probably living through it will feel impressive but manageable. From a relativistic perspective, the singularity happens bit by bit, and the merge happens slowly. We are climbing the long arc of exponential technological progress; it always looks vertical looking forward and flat going backwards, but it’s one smooth curve. (Think back to 2020, and what it would have sounded like to have something close to AGI by 2025, versus what the last 5 years have actually been like.) 展望未来，这种变化令人难以理解。但亲历其中时，或许会感到震撼却仍可适应。从相对论视角看，奇点是逐步发生的，融合过程缓慢推进。我们正沿着指数级技术进步的漫长弧线攀爬——前瞻时总觉陡峭如垂直，回望时却显得平缓，但始终是一条平滑曲线。（不妨回想 2020 年：若当时有人预言 2025 年将出现接近通用人工智能的技术，对比过去五年的真实发展历程，就能体会到这种认知差异。）
There are serious challenges to confront along with the huge upsides. We do need to solve the safety issues, technically and societally, but then it’s critically important to widely distribute access to superintelligence given the economic implications. The best path forward might be something like: 在拥有巨大优势的同时，我们也面临着严峻挑战。技术上和社会层面都需要解决安全问题，但鉴于其经济影响，广泛分配超级智能的使用权限至关重要。前进的最佳路径或许类似于：
-
Solve the alignment problem, meaning that we can robustly guarantee that we get AI systems to learn and act towards what we collectively really want over the long-term (social media feeds are an example of misaligned AI; the algorithms that power those are incredible at getting you to keep scrolling and clearly understand your short-term preferences, but they do so by exploiting something in your brain that overrides your long-term preference). 解决对齐问题，意味着我们能强有力地保证 AI 系统长期学习并践行人类集体的真实意愿（社交媒体信息流就是 AI 未对齐的例子；驱动这些平台的算法极其擅长让你不断滑动屏幕并精准捕捉短期偏好，但它们是通过利用你大脑中某种机制来实现的——这种机制会压制你的长期偏好）。
-
Then focus on making superintelligence cheap, widely available, and not too concentrated with any person, company, or country. Society is resilient, creative, and adapts quickly. If we can harness the collective will and wisdom of people, then although we’ll make plenty of mistakes and some things will go really wrong, we will learn and adapt quickly and be able to use this technology to get maximum upside and minimal downside. Giving users a lot of freedom, within broad bounds society has to decide on, seems very important. The sooner the world can start a conversation about what these broad bounds are and how we define collective alignment, the better. 继而应聚焦于如何让超级智能变得廉价、广泛可用且不过度集中于任何个人、公司或国家之手。社会是坚韧的、富有创造力的，并能快速适应变化。若能汇聚大众的意志与智慧，尽管我们会犯许多错误，某些事态可能严重失控，但我们将快速学习调整，从而运用这项技术实现收益最大化与风险最小化。在由社会共同划定的宽泛边界内，赋予用户高度自由显得至关重要。世界越早开启关于这些边界范围及如何定义集体共识的对话，结果就会越好。
We (the whole industry, not just OpenAI) are building a brain for the world. It will be extremely personalized and easy for everyone to use; we will be limited by good ideas. For a long time, technical people in the startup industry have made fun of “the idea guys”; people who had an idea and were looking for a team to build it. It now looks to me like they are about to have their day in the sun. 我们（指整个行业，而不仅限于 OpenAI）正在为世界构建一个大脑。它将高度个性化且人人皆可轻松使用；唯一限制我们的将是优秀的创意。长期以来，初创企业圈的技术人士总嘲笑那些"空想家"——那些只有想法却寻求团队来实现的人。现在看来，这些人即将迎来属于他们的高光时刻。
OpenAI is a lot of things now, but before anything else, we are a superintelligence research company. We have a lot of work in front of us, but most of the path in front of us is now lit, and the dark areas are receding fast. We feel extraordinarily grateful to get to do what we do. OpenAI 如今涉足众多领域，但归根结底，我们是一家超级智能研究公司。前路虽任务艰巨，但大部分路径已然明朗，未知领域正快速消退。能从事这份事业，我们心怀无限感激。
Intelligence too cheap to meter is well within grasp. This may sound crazy to say, but if we told you back in 2020 we were going to be where we are today, it probably sounded more crazy than our current predictions about 2030. 近乎免费的智能触手可及。这说法或许看似疯狂，但若在 2020 年告诉你我们将抵达今日之境，恐怕比如今对 2030 的预测更令人难以置信。
May we scale smoothly, exponentially and uneventfully through superintelligence. 愿我们平稳、指数级且波澜不惊地迈向超级智能之巅。