《Brave New Words 》9.4 我们的处境及未来：呼吁受过教育的勇气

Part IX: Work and What Comes Next

第九部分：工作及其未来发展

Where This Leaves Us and Where It Will Take Us: A Call for Educated Bravery

我们的处境及未来：呼吁受过教育的勇气

I have a confession. I once thought I would be an AI researcher. I viewed, and still view, intelligence and perception—which are two different things—as the biggest mysteries of the universe. I was fascinated by the idea of being able to build something as smart, or even smarter, than any of us. I had read nearly every science-fiction book on the topic. I loved thinking about how we could prove whether another being was truly sentient. After all, we can only directly perceive our own perception. It is really a leap of faith that other creatures—including other people in our life—are truly sentient versus just acting like they are. The best way to understand intelligence, I once thought, was to construct machines that are capable of it.

我要坦白一件事。我曾以为自己会成为一名人工智能研究员。我认为，并且仍然认为，智能和感知——这是两个不同的概念——是宇宙中最大的谜团。我被能够构建某种与我们一样聪明，甚至比我们更聪明的东西这个想法所吸引。我几乎读过所有关于这个话题的科幻书。我喜欢思考我们如何能够证明另一个存在是否真正有感知能力。毕竟，我们只能直接感知到自己的感知。相信其他生物——包括我们生活中的其他人——是真正有感知能力的，而不是仅仅是表现得像是有感知力，这实际上是一种信仰的飞跃。我曾经认为，理解智力的最好方法是构建能够进行智力活动的机器。

When I was a freshman at MIT in 1994, I was lucky to have direct access to several of the titans of AI at the time. I sought out Patrick Henry Winston to be my freshman adviser. He was the director of the MIT Artificial Intelligence Laboratory and author of the canonical textbook on artificial intelligence at the time. I took his class, Introduction to Artificial Intelligence. I also took Marvin Minsky’s class, Society of Mind. Minsky was Winston’s mentor and the founder of the Artificial Intelligence Laboratory. He also won computer science’s highest award—the Turing Award—for “his central role in creating, shaping, promoting, and advancing the field of Artificial Intelligence.” His ideas were considered foundational for the field of artificial neural networks. He was also the AI adviser to Stanley Kubrick when he made perhaps the most famous AI film of all time, 2001: A Space Odyssey.

1994 年，当我还是麻省理工学院的一名新生时，我很幸运能够直接接触到当时几位人工智能领域的大师。我找到了帕特里克·亨利·温斯顿，请他做我的新生导师。他是麻省理工学院人工智能实验室的主任，也是当时权威的人工智能教材的作者。我选修了他的课《人工智能导论》。我还选修了马文·明斯基的课《心智社会》。明斯基是温斯顿的导师，也是人工智能实验室的创始人。他还因“在创造、塑造、推动和发展人工智能领域的核心作用”而获得了计算机科学领域的最高奖项——图灵奖。他的思想被认为是人工神经网络领域的基础。他还是斯坦利·库布里克拍摄也许是有史以来最著名的人工智能电影《2001：太空漫游》时的人工智能顾问。

These professors were incredibly intelligent, creative, and inspiring, but I found myself disappointed in where the field was and how slowly it seemed to be developing. The most impressive AI systems that could play games like chess were just good at anticipating decisions several moves ahead. No matter how proficient these systems got, no one really believed that they would be intelligent in the same way we are. Artificial neural nets were compelling from a philosophical point of view, but they weren’t really capable of doing anything truly mind-blowing at the time. There hadn’t been any big, new ideas in twenty or thirty years. Little did I know that this was the tail end of what would later be considered an “AI winter” among researchers.

这些教授极其聪明、富有创造力和启发性，但我对该领域的发展现状以及它的发展速度感到失望。那些能够下棋的最令人印象深刻的人工智能系统只是擅长提前几步预测决策。无论这些系统变得多么熟练，没有人真的相信它们会像我们一样智能。从哲学角度来看，人工神经网络是引人入胜的，但它们当时并没有真正实现任何令人惊叹的事情。在过去的二三十年里，几乎没有任何重大、新颖的想法。我不知道的是，这实际上是后来被认为的“人工智能寒冬”的尾声。

So I decided to move on. I still loved computer science and thought that I would eventually try to start some type of tech company. But questions around intelligence and, by extension, education continued to draw me in because they seemed so fundamental to the advancing of society. The summer after my junior year, I received a fellowship to create software that allows students to learn and practice math at their own time and pace. Sound familiar?

于是我决定继续前进。我依然热爱计算机科学，并认为自己最终会尝试创办某种科技公司。但有关智能的问题以及扩展到教育的问题继续吸引着我，因为它们对社会的进步显得如此根本。在我大三后的那个夏天，我获得了一项奖学金，用于创建一种软件，允许学生按照自己的时间和节奏学习和练习数学。听起来很熟悉吗？

I started to believe that people had a lot of latent, unused potential. For every person born with the raw material to be Albert Einstein or Marie Curie, how many get the education and support to do so? What if, with broader, more accessible education, we could increase by a factor of ten or one hundred the number of people capable of making the next major scientific, artistic, or entrepreneurial leap for us all? How many more diseases might we cure? How much faster might we explore the cosmos?

我开始相信人们有很多潜在的、未被利用的潜力。对于每个拥有成为爱因斯坦或居里夫人的原材料的人来说，有多少人能获得相应的教育和支持？如果通过更广泛、更便捷的教育，我们可以将能够为我们所有人做出下一次重大科学、艺术或创业飞跃的人数增加十倍或一百倍呢？我们可能治愈多少更多的疾病？我们探索宇宙的速度可能会有多快？

My curiosity wasn’t just about fostering genius. If everyone had access to truly great education, I wondered, how many more billions of people might attain purpose and meaning in their lives?

我的好奇心不仅仅是关于培养天才。我想知道，如果每个人都能获得真正优质的教育，有多少亿人可以在他们的生活中获得目标和意义？

But practical reality was there in the background. I grew up in a single-mother household. My parents separated shortly after I was born, and I only met my father once before he died when I was fourteen. He was a pediatrician and came from a prominent family of politicians and academics in Bangladesh, but we never received any financial support because I think he was barely making it himself. When he died, my sister and I inherited a Nissan Sentra that had more debt on it than it was worth. The only narrative I can piece together is that he and my mother were wildly incompatible, as they had an arranged marriage, and he likely suffered from depression. For most of my life, my mother was a cashier at various convenience stores, making enough money to be slightly below the poverty line. MIT was generous with financial aid, but I still had about thirty thousand dollars in debt upon graduation. The tech boom was heating up, and when I found out that I could make eighty thousand dollars a year as a new computer science graduate, which was about five times what my mother was making, I could not pass the opportunity by and took a job at Oracle Corporation.

但现实情况在背景中存在。我在单亲家庭中长大。我的父母在我出生后不久就分开了，在我十四岁之前，我只见过我父亲一次。他是一名儿科医生，来自孟加拉国一个著名的政治家和学者家庭，但我们从未获得过任何经济支持，因为我想他自己也勉强维持生计。当他去世时，我和我姐姐继承了一辆日产 Sentra，但这辆车的债务比它的价值还要高。我唯一能拼凑起来的故事是，他和我母亲非常不合，因为他们是包办婚姻，他可能患有抑郁症。在我大部分生活中，我母亲在各种便利店当收银员，收入仅能略低于贫困线。麻省理工学院在经济援助方面很慷慨，但我毕业时仍然背负大约三万美元的债务。科技热潮正在升温，当我发现作为一名新的计算机科学毕业生可以每年赚八万美元时，这大约是我母亲收入的五倍，我无法放过这个机会，于是我在甲骨文公司找到了一份工作。

I later went to business school and found myself as an analyst at a hedge fund. My then fiancée and now wife would give me grief about how I wasn’t doing anything helpful for humanity with my talents and education. I found investing to be intellectually fascinating, though. It allowed me to study how the world worked, along with the animal spirits of the market. I also needed the money. I had further debt to pay from business school. I also knew that I was going to support my mother and other family members, and I was pretty determined to not perpetuate the financial insecurity that I grew up with. If I’m honest, I’m still more insecure about this than most of my friends. I would also tell folks that I was only going to do this until I was independently wealthy so that I could start a school on my own terms. I had some ideas about one day being the Dumbledore at a school that focused on putting students at the center and giving much more time and space for them to explore their passions.

后来我去了商学院，发现自己成为了一名对冲基金的分析师。我当时的未婚妻、现在的妻子，会为我没有利用我的才华和教育做对人类有帮助的事情而感到难过。不过，我发现投资在智力上非常吸引人。它让我能够研究世界如何运作，以及市场的动物精神。我也需要钱。我还有商学院的债务需要偿还。我也知道我将赡养我的母亲和其他家庭成员，我决心不再延续我成长过程中经历的经济不安全感。说实话，我在这方面比我的大多数朋友都更加没有安全感。我还会告诉人们，我只打算做这份工作，直到我我经济独立，这样我就可以按照自己的条件创办一所学校。我曾设想过有一天成为一所学校的邓布利多，这所学校专注于以学生为中心，给予他们更多的时间和空间去探索他们的激情。

It was at that time, in 2004, when I had family from New Orleans visiting me in Boston after my wedding. It came out of a conversation with my aunt that my twelve-year-old cousin Nadia was having trouble in math, and I offered to tutor her remotely. That led to the beginning of Khan Academy, which at its essence has been all about trying to scale the type of personalized learning that I did with Nadia to hundreds of millions of learners, across subjects, grades, and geographies.

就在那时，2004 年，我的家人从新奥尔良来波士顿参加我的婚礼。在与我姨妈的一次谈话中，她提到我十二岁的表妹纳迪亚在数学上遇到了困难，我主动提出远程辅导她。这导致了可汗学院的开始，其本质一直是将我与纳迪亚进行的个性化学习扩展到数亿名学习者，涵盖各个学科、年级和地域。

Through the years, many people have asked me why I set up Khan Academy as a nonprofit. After all, my previous career was very for-profit, and I live in the middle of Silicon Valley, where scalable tech-enabled solutions can be worth a lot of money. Many have been skeptical whether a nonprofit could even compete with for-profit companies. There were two notions I couldn’t get out of my head, however. First, I tend to believe in market forces, but there are a few sectors—namely, education and health care—where the outcomes of market forces don’t always align with our values. Education and health care are two areas where our shared values tell us that, ideally, family resources shouldn’t be a limiting factor in accessing the best possible opportunities. Most of us believe that every mind and life deserves to reach its full potential.

多年来，许多人问我为什么将可汗学院设立为非营利机构。毕竟，我之前的职业生涯是非常盈利的，我住在硅谷中部，在那里可扩展的技术解决方案可能价值不菲。许多人对非营利组织是否能与营利性公司竞争持怀疑态度。然而，有两个概念我无法摆脱。首先，我倾向于相信市场力量，但有几个行业——即教育和医疗——市场力量的结果并不总是与我们的价值观一致。教育和医疗是两个领域，我们的共同价值观告诉我们，理想情况下，家庭资源不应该成为获得最佳机会的限制因素。我们大多数人都相信每个头脑和生命都应该有机会发挥其全部潜力。

The second notion was more grandiose, if not outright delusional. One of my favorite sets of books is the Foundation series by Isaac Asimov. It takes place tens of thousands of years in the future, when humanity has colonized the galaxy, unified under one empire. Within that empire, an academic by the name of Hari Seldon has developed a new field called psychohistory—something of a combination of history, economics, and statistics—that can probabilistically predict large-scale historic trends. This science tells him that the Galactic Empire will enter a ten-thousand-year dark age within the next few hundred years. This will be ten thousand years of war, famine, and lost knowledge. His calculations show that nothing can prevent the coming dark ages, but it can be shortened. So he starts a foundation at the periphery of the galaxy to preserve knowledge and technology, which can then be used to shorten the coming chaos to “only” one thousand years. The book series mainly focuses on how the ensuing hundreds of years actually play out.

第二个概念更为宏大，甚至有些妄想。我最喜欢的一套书是艾萨克·阿西莫夫的《基地》系列。故事发生在几万年后的未来，那时人类已经在银河系殖民，并在一个帝国的统治下统一。在那个帝国里，一个名叫哈里·谢顿的学者开发了一个新的领域，叫做心理史学——历史、经济学和统计学的结合——可以概率性地预测大规模的历史趋势。这门科学告诉他，银河帝国将在未来几百年内进入一万年的黑暗时代。这将是万年的战争、饥荒和知识丧失。他的计算显示，没有什么可以阻止即将到来的黑暗时代，但可以缩短它。所以他在银河系的边缘建立了一个基地来保存知识和技术，然后可以用来将即将到来的混乱缩短到“只有”一千年。这套丛书主要讲述的是接下来的几百年实际上是如何发展的。

When I first read the Foundation series in middle school, I found it inspiring to think along those time scales. It was also the first moment I truly appreciated that the strength of a civilization doesn’t lie in its physical size, power, and wealth. Those are just by-products of where the real strength lies: a society’s culture, know-how, and mindset.

当我在中学第一次阅读《基地》系列时，我发现以那种时间尺度来思考是很有启发性的。这也是我第一次真正认识到，一个文明的力量不在于其物理规模、力量和财富。这些只是真正力量所在的副产品：一个社会的文化、知识和思维方式。

Jump ahead to when Khan Academy was beginning, and I realized that very few people in our society think on a scale of more than a few years or decades, much less hundreds or thousands of years. Beyond this, the internet was clearly the transformational technology of our time, but no real institutions were being built with it. I began to wonder whether Khan Academy might just be able to become one of the first of them; something that could help educate billions of people for hundreds of years to come. It would be like Hari Seldon’s foundation, except in our case, we could uplift humanity so that the present moment would feel like a dark age when looked back upon from fifty or a hundred years in the future. We only have one life—why not swing for the fences?

跳到可汗学院刚开始的时候，我意识到我们社会中很少有人以超过几年或几十年的规模思考，更不用说几百或几千年了。除此之外，互联网显然是我们这个时代的变革性技术，但没有真正的机构是用它来构建的。我开始思考，可汗学院是否可能成为其中之一；一个可以在未来几百年里帮助教育数十亿人的机构。它就像哈里·谢顿的基地，只是在我们的案例中，我们可以提升人类，这样，从五十年或一百年后回首，现在就会感觉像是一个黑暗时代。我们只有一次生命——为什么不放手一搏呢？

As Khan Academy grew and scaled from tens to hundreds of millions of people, that dream seemed to feel less and less delusional. Amazing people came out of the woodwork to help us. By the fall of 2009, I had quit my hedge fund job to work on Khan Academy full time. Ten months later, my family was quickly depleting our savings. Our first child was born, and I was having trouble sleeping because of financial stress and, to some degree, the shame of giving up a lucrative job for something that didn’t seem to have a future. At what seemed like Khan Academy’s darkest moment, Ann Doerr—who is now our chairperson—and John Doerr miraculously showed up and donated enough money for me to keep going. Since then, hundreds of thousands of people have donated to support us. Despite being a nonprofit, we have been able to build a team that rivals those of the most resource-rich tech companies. Hundreds of incredibly talented people have committed a major part of their careers to be part of the Khan Academy team, often taking considerable pay cuts to do so. Thousands of volunteers all over the world have now translated Khan Academy into over fifty languages. Inspirational leaders like Bill Gates, Reed Hastings, and Elon Musk have become some of our biggest supporters and advocates. This journey seems so serendipitous that it has become something of an inside joke among the Khan Academy team that perhaps benevolent aliens are helping us so that, through education, we can prepare humanity for first contact.

随着可汗学院的发展和扩展，从数十万到数亿人，这个梦想似乎越来越不像是妄想。许多了不起的人物自发地来帮助我们。到 2009 年秋天，我辞去了对冲基金的工作，全职投入到可汗学院的工作中。十个月后，我的家庭很快耗尽了我们的积蓄。我们的第一个孩子出生了，由于财务压力，我难以入睡，在某种程度上，还感到因放弃一份有前途的高薪工作而感到羞愧。在看似可汗学院最黑暗的时刻，安·多尔——她现在是我们的主席——和约翰·多尔奇迹般地出现了，他们捐赠了足够的钱让我继续下去。从那时起，成千上万的人捐赠支持我们。尽管是非营利组织，我们已经能够建立一个能与资源最丰富的科技公司媲美的团队。数百名才华横溢的人才将他们职业生涯的重要部分投入到可汗学院团队中，他们常常为此大幅降薪。现在，全世界数千名志愿者已经将可汗学院翻译成五十多种语言。像比尔·盖茨、里德·哈斯廷斯和伊隆·马斯克这样的鼓舞人心的领导人成为我们最大的支持者和倡导者。这段旅程看起来如此偶然，以至于在可汗学院团队中成为一种内部笑话，也许是仁慈的外星人在帮助我们，通过教育，为人类准备好第一次接触。

This narrative seemed to be reinforced when Sam Altman and Greg Brockman of OpenAI reached out to us before anyone else with a technology that seemed to tie together every thread of my journey. GPT-4 was built on years of important innovations from many people and companies, but it was the first AI technology that truly made me wonder whether I was dreaming (or perhaps living in a simulation). It surpassed anything that, back in 1994, aspiring AI researcher Sal could have ever imagined happening in his lifetime. More important, it was the potential missing piece to our goal of delivering a truly world-class education for anyone, anywhere. I realized that as thrilling as it would be to be an AI researcher now, it was even more exciting to think about how the technology could be applied to help human potential.

当 OpenAI 的萨姆·奥特曼和格雷格·布罗克曼在其他人之前联系到我们时，这种说法似乎得到了加强，他们带来了一个看似将我旅程的每一个线索联系在一起的技术。GPT-4 建立在许多人和公司多年来的重要创新之上，但它是第一个真正让我怀疑自己是否在做梦（或许生活在一个模拟中）的人工智能技术。它超越了 1994 年时有抱负的人工智能研究员萨尔在他一生中可能想象到的任何事情。更重要的是，它是我们为任何人、任何地方提供真正世界级教育目标中缺失的部分。我意识到，尽管现在成为一名人工智能研究员会令人兴奋，但思考如何这项技术如何能够应用于帮助人类潜力更加令人兴奋。

This is not something to be taken lightly; there is real urgency here. Despite making us far more productive as a whole, this technology also has the potential to displace or disrupt many industries and jobs. The traditional labor pyramid—with less-skilled manual labor forming the bottom layer, bureaucratic white-collar jobs making up the middle layer, and highly skilled knowledge work and entrepreneurship making up the top—no longer applies. Robotics, including self-driving cars and trucks, is going to dramatically reduce the need for humans in that bottom layer. Generative AI can clearly perform large aspects of the work of the middle, white-collar layer and even parts of today’s most skilled professions. A society in which all the productivity and resulting wealth accrues to only the tippy-top of the traditional labor pyramid, likely concentrated in Silicon Valley, with many others out of work, will not be a stable one. It might lead to massive wealth redistribution efforts. This scenario is dystopian because most people aren’t looking for a handout. Rather, they want to have a sense of purpose and a feeling of contributing to the world.

这不是一件可以轻视的事情；这里存在真正的紧迫感。尽管整体上使我们生产力大幅提升，这项技术也有可能取代或颠覆许多行业和工作。传统的劳动金字塔——底层是不熟练的体力劳动，中层是官僚的白领工作，顶层由高技能知识工作和创业构成——不再适用。包括自动驾驶汽车和卡车在内的机器人技术将极大减少对底层人力的需求。生成式人 AI 显然可以完成中层白领的大部分工作，甚至可以完成当今一些最熟练职业的一部分工作。在一个所有生产力和由此产生的财富都只集中在传统劳动金字塔的顶端，可能集中在硅谷，而许多人失业的社会，将不会是一个稳定的社会。这可能导致大规模的财富重新分配。这种情景是反乌托邦式的，因为大多数人不希望得到施舍。他们更希望有一种目标感和为世界做贡献的感觉。

The real solution is to invert that labor pyramid so that most people can operate at the top and use AI and other technology for their own productivity and entrepreneurship. The only way we have a hope of doing this is to use the same AI technology to lift the skills of a large chunk of humanity in the coming decades.

真正的解决方案是颠倒劳动金字塔，使大多数人能够在顶层工作，并使用人工智能和其他技术来提高他们自己的生产力和创业能力。我们唯一希望做到这一点的方法是使用相同的人工智能技术，在未来几十年内提升大部分人类的技能。

Few people may view the Star Trek universe through an economic lens, but doing so provides a window into a world that might soon be upon us. All of classical economics is based on the notion of scarcity—namely, that there isn’t usually enough of anything to give everyone everything they want or need. Because of that, we use markets and pricing to allocate those goods, services, and resources to where they might result in the highest benefit. In Star Trek, however, there isn’t much scarcity. Technology has allowed that society to replicate any food they want, transport themselves thousands of miles in the blink of an eye, communicate over light-years, and travel among the stars. All of humanity in that world has been fully educated so that they can participate in this bounty. Everyone is an explorer, researcher, engineer, artist, doctor, or counselor. Generative AI has the potential to allow many dimensions of our own society to be similarly low scarcity or highly abundant. Do we have the will to take us to the utopia of Star Trek?

很少有人会从经济角度来看《星际迷航》宇宙，但这样做提供了一个可能即将到来的世界的窗口。所有经典经济学都基于稀缺的概念——即，通常没有足够的任何东西来满足每个人所有的需求和欲望。因此，我们使用市场和定价将这些商品、服务和资源分配到可能带来最高收益的地方。然而，在《星际迷航》中，几乎没有什么是稀缺的。技术使那个社会能够复制他们想要的任何食物，在眨眼之间将自己传送数千英里，跨越光年进行通信，并在星际间旅行。那个世界的所有人类都得到了全面的教育，以便他们能够参与这场盛宴。每个人都是探险家、研究员、工程师、艺术家、医生或顾问。生成式 AI 有可能使我们自己的社会在许多方面变得同样稀缺或高度丰富。我们是否有意愿将我们带入《星际迷航》的乌托邦？

If we don’t, societies will increasingly fall prey to populism. People with time but no sense of purpose or meaning don’t tend to be good for themselves or others. They are susceptible to the ideas of demagogues. Generative AI can be used to move us in this negative direction by reinforcing “fake news” with fabricated videos and images. It can be used by governments to police their own populations much more tightly than anything George Orwell imagined in 1984. For decades it has been possible to put cameras and sensors throughout a city and to tap phone lines, but it was difficult to monitor all the information and make sense of it. AI could soon flag any recording or observation that seems like disobedience to the eyes of the state. Big Brother will not only be watching but will also have comprehension.

如果我们不这样做，社会将越来越容易成为民粹主义的猎物。那些有时间但没有目标或意义的人往往对自己或他人都不利。他们容易受到煽动者的思想影响。生成式 AI 可以通过用捏造的视频和图像强化“假新闻”来将我们引向这个负面方向。政府可以利用它来比乔治·奥威尔在《1984》中想象的任何东西都更紧密地监控自己的公民。几十年来，人们已经可以在整个城市安装摄像头和传感器并窃听电话线，但很难监控所有信息并理解它们。人工智能很快就能标记出任何在国家眼中看起来像是不服从的记录或观察。老大哥不仅会监视你，而且还会理解你。

Without proper countermeasures and AI literacy, people will also fall victim to increasingly sophisticated fraud. In the near future, expect to get phone calls, or even video chats, from something that looks like your family member, telling you that they are in an emergency and that you need to wire them money.

如果没有适当的对策和人工智能素养，人们也将成为日益复杂的欺诈行为的受害者。在不久的将来，预计会接到电话，甚至视频聊天，看起来像是你的家庭成员，告诉你他们处于紧急情况，需要你汇款给他们。

AI will increasingly play a role in national security. Foreign enemies will have the capability to orchestrate increasingly sophisticated attacks on infrastructure using this technology—attacks that potentially involve manipulating human beings. AI-generated images of people waiting in line to get their deposits could go viral on social media and start a run on banks. State and nonstate actors will use generative AI within social media to try to influence the outcomes of our elections and make us more divided as a society. The best tacticians in the battles of the future are likely to be AI, not human.

人工智能将在国家安全中扮演越来越重要的角色。外国敌人将有能力利用这项技术策划越来越复杂的基础设施攻击——这些攻击可能涉及操纵人类。生成式 AI 生成的排队取款的人们的图像可能在社交媒体上迅速传播，引发银行挤兑。国家和非国家行为者将利用生成式 AI 在社交媒体上试图影响我们的选举结果，使我们社会更加分裂。未来战争中最好的战术家可能是人工智能，而不是人类。

These very real possibilities may motivate some to advocate for slowing down innovation. Honestly, even I find the pace of its development dizzying. But the genie is out of the bottle, and the bad actors are not about to slow down because we want them to. Today, the good actors have the edge, but it really is a race. The countermeasure for every risk is not slowing down; it is ensuring that those favoring liberty and empowering humanity have better AI than those on the side of chaos and despotism.

这些非常现实的可能性可能会激励一些人主张放慢创新的步伐。坦率地说，即使是我也觉得其发展的速度令人眼花缭乱。但魔瓶已经打开，坏人不会因为我们希望他们放慢脚步而放慢。今天，好人占了上风，但这确实是一场竞赛。对每种风险的对策不是放慢速度，而是确保那些支持自由和赋权人类的人拥有比混乱和专制一方更好的人工智能。

This moment can be an existential risk or an existential opportunity for us. People have every right to be both scared and hopeful of what this leap in technology and innovation means. I do not, however, think our fate is subject to the flip of a coin. Rather, each of us is an active participant in the decision about how we will use AI moving forward. If we act with fear, the rule followers might pause, but the rule breakers, from totalitarian governments to criminal organizations, are going to accelerate their development of AI. The only way that we can ensure that we are closer to reaching a utopian Star Trek scenario is if we double down our efforts on using large language models for the good of society.

这一时刻对我们来说可能是生存风险或生存机遇。人们完全有理由对这次技术和创新的飞跃既感到害怕又充满希望。然而，我不认为我们的命运取决于投掷硬币。相反，我们每个人都是关于我们将如何使用人工智能的决定中的积极参与者。如果我们以恐惧行事，守规则的人可能会暂停，但从极权政府到犯罪组织的破坏规则的人将加速他们的人工智能开发。我们确保自己更接近实现《星际迷航》乌托邦式场景的唯一方法是加倍努力使用大型语言模型为社会造福。

This is not a drill: generative AI is here to stay. The AI tsunami has drawn back from the shore, and it is now barreling toward us. Faced with the choice between running from it or riding it, I believe in jumping in with both feet, while taking proper precautions so that we don’t get hit with the flotsam.

这不是演习：生成式 AI 将永远存在。人工智能海啸已经从岸边退去，现在正向我们扑来。面对逃离或驾驭它的选择，我相信应该双脚跳入，同时采取适当的预防措施，以免被漂浮物击中。

Each of us has an obligation to make sure that we use this technology responsibly. This means that as developers we must put the necessary guardrails on it to protect our children. When problems arise, we should apply reasonable regulations, regulations that don’t give an edge to rule breakers. All the while, we must accelerate our efforts and make sure that we are developing the technology with the right intent and the right pedagogy. This will allow us to accelerate the improvement of human purpose and potential. Let’s use AI to create a new golden age for humanity, a time that will make today look like a dark age. From my vantage point, nothing could be more inspiring and important than that.

我们每个人都有义务确保我们负责任地使用这项技术。这意味着，作为开发者，我们必须为其设置必要的护栏，以保护我们的孩子。当问题出现时，我们应该采取合理的法规，这些监管不会让那些打破规则的人占便宜。同时，我们必须加快努力，确保我们以正确的意图和正确的教学法来开发这项技术。这将使我们能够加速人类目标和潜力的提升。让我们利用人工智能为人类创造一个新的黄金时代，一个将使今天看起来像是黑暗时代的时代。从我的角度来看，没有什么比这更令人振奋和重要的了。

“点赞有美意，赞赏是鼓励”