---
author: New York Times Podcasts
date: '2026-09-24'
guest: ''
layout: post.njk
source: https://www.youtube.com/watch?v=IdcjplwJUQ8
speaker: New York Times Podcasts
tags:
  - ai-safety
  - arms-control
  - nuclear-proliferation
  - geopolitical-rivalry
  - international-regulation
title: 中美核控历史镜鉴：AI末日风险与大国监管困局
summary: 在美中两国元首会晤探讨人工智能监管之际，《纽约时报》大卫·桑格深入剖析了冷战时期美苏核不扩散协议的历史经验与教训。对话对比了核武器与AI在技术扩散、军民两用、私企主导及检测核查上的根本差异，探讨了构建AI安全早期预警与国际合作的可行路径与严峻挑战。
insight: ''
draft: true
series: ''
category: geopolitics
area: geopolitics
project: []
people: []
companies_orgs:
  - OpenAI
products_models: []
media_books: []
status: evergreen
---
### 中美峰会与AI监管

**迈克尔·巴巴罗**: 这里是《**纽约时报**》（New York Times），我是**迈克尔·巴巴罗**（Michael Barbaro）。这里是《**The Daily**》。昨晚，在一场盛大的仪式中，**特朗普总统**（President Trump）迎接到访美国的中国领导人**习近平**，开启为期两天备受瞩目的谈判。在他们排得满满当当的议程上，有一个焦点议题：面对全球对**人工智能**（AI）风险的恐慌，美中两国应当在监管人工智能方面扮演怎样的角色？今天，正如我的同事**大卫·桑格**（David Sanger）所解释的那样，当谈及如何保护人类免受技术灾难时，摆在两位领导人面前真正经过历史检验的模型其实只有一个。今天是9月24日星期四。大卫专程从华盛顿赶来，正坐在我们纽约的演播室里，这回甚至都不用我大老远跑去找你了。

<details>
<summary>Original English</summary>

**Michael Barbaro**: From the New York Times, I'm Michael Barbara. This is the Daily. Last night, in a lavish ceremony, President Trump welcomed Chinese leader Xi Jinping as he arrived in the United States for 2 days of closely watched negotiations. on their crowded agenda. What if any role they should play in regulating artificial intelligence amid a global freakout over its risks? Today, as my colleague David Sanger explains, when it comes to protecting humanity from technological catastrophe, there's really only one proven model for the two leaders to look to. It's Thursday, September 24th, David up from Washington, DC, sitting here in our New York studios, and I didn't even have to come to you this time.

</details>

**大卫·桑格**: 你知道的，我就是想亲自来纽约见识一下《The Daily》的帝国大厦。

<details>
<summary>Original English</summary>

**David Sanger**: You know, I just had to come see the Dy's Empire in New York

</details>

**迈克尔·巴巴罗**: 亲自来见识。

<details>
<summary>Original English</summary>

**Michael Barbaro**: in person.

</details>

**大卫·桑格**: 亲自来。没错，相当不错。

<details>
<summary>Original English</summary>

**David Sanger**: In person. Yeah, it's pretty

</details>

**迈克尔·巴巴罗**: 这是一个微不足道的帝国，属于帝国晚期了，［笑］。不过话说回来，大卫，当特朗普总统和中国领导人习近平今天在华盛顿会晤时，他们将触及一个萦绕在许多人心头的话题，那就是**人工智能的风险**。这种风险当然包括那些成群结队的**流氓AI智能体**（rogue AI agents），它们一直在黑入各类公司，从根本上违背其创造者设定的规则。而习近平和特朗普能以独特方式为此做些什么，将成为这次会晤的核心议题之一。我想大家都已经明白，美中两国领导人在讨论AI安全方面处于极其特殊的位置，因为他们拥有世界上最顶尖的前沿AI实验室。要想遏制AI风险，唯一的途径就是两国以某种方式协同一致、步调一致地采取行动。随着这场峰会临近，这也就是我们为什么这么喜欢听你分析的原因——你联系我们说：你看，这部电影我以前见过。大恐惧、大思想，围绕重大问题寻求协调与解决方案。你告诉我们，这一切在历史上都有迹可循，甚至可能存在一个历史性的解决方案。

<details>
<summary>Original English</summary>

**Michael Barbaro**: It's a It's a humble empire. It's a late stage, [laughter] but uh so David, when President Trump and the leader of China, Xi Jinping, meet in Washington today, they're going to broach this subject on many people's minds, which is the risk of artificial intelligence. That risk, including, of course, these swarming rogue AI agents that have been hacking into companies and basically breaking the rules of their creators. and what she and Trump uniquely can maybe do about it is going to be the subject of at least some of this meeting. And I think we've all come to understand that the leaders of the US and China are uniquely positioned to talk about the safety of AI because they're the countries with the most advanced frontier AI laboratories. And the only way that AI gets contained is if both of those countries somehow act together and at the same time. And as this meeting approached, and this is why we love you so much, you reached out to us to say, you know, I have seen this movie before. Big fears, big thoughts about coordination and solutions to big problems. You told us there's a historical parallel to all of this and maybe a historical solution.

</details>

**大卫·桑格**: 迈克尔，这里确实存在一个显而易见的平行对照。从20世纪40年代末开始，贯穿整个冷战时期，我们经历过这样一个时代：苏联和美国这两个超级大国创造了一种潜在的极其可怕的技术，这种技术具有广泛的商业用途，但同时也可能带来**人类文明的毁灭**。

<details>
<summary>Original English</summary>

**David Sanger**: Well, Michael, there is a obvious parallel. We had a era starting in the late 1940s and through the Cold War where two states, the Soviet Union and the United States, had created a potentially terrifying technology [snorts] that had all kinds of commercial uses, but could also bring about civilizational extinction.

</details>

**迈克尔·巴巴罗**: **核武器**。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Nuclear weapons.

</details>

**大卫·桑格**: 核武器。当时每个人都必须绞尽脑汁去弄明白：我们如何在享受其益处的同时，不被其恐怖所吞噬？你知道吗？这花了整整几十年时间才办到。

<details>
<summary>Original English</summary>

**David Sanger**: Nuclear weapons. And everybody had to go figure out how we were going to get the benefits without getting the terror. And you know what? It took decades to happen.

</details>

**迈克尔·巴巴罗**: 这两个国家最终携起手来，达成了一系列协议，最终让我们迎来了这个非凡的时刻：自1945年以来，从未有任何一枚核武器在战争中被再次使用。

<details>
<summary>Original English</summary>

**Michael Barbaro**: These two countries, they locked arms and they came up with agreements that ultimately have brought us to the remarkable moment that there has been no use of a nuclear weapon in anger since 1945.

</details>

**迈克尔·巴巴罗**: 对吧？**核不扩散**机制发挥了作用。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Right? Nuclear non-prololiferation worked.

</details>

**大卫·桑格**: 核不扩散，加上**恐怖平衡**（balance of terror），确实奏效了。

<details>
<summary>Original English</summary>

**David Sanger**: Nuclear non-prololiferation combined with a balance of terror worked.

</details>

**迈克尔·巴巴罗**: 因此，现在当人们开始思考如何遏制人工智能时，大家很自然地将目光投向了那个历史模型。尤其是在经历了今年夏天一系列出人意料的事件之后——那些失控的智能体打破沙箱禁锢脱逃。对吧？像当年超级大国对待核武器那样去对待AI。

<details>
<summary>Original English</summary>

**Michael Barbaro**: So now quite naturally everybody's looking to that model as we begin to think about how you might contain artificial intelligence particularly after this summer of surprises that we've had where we have rogue agents breaking out of their sandboxes. Right? treat AI the way superpowers treated nuclear weaponry.

</details>

**大卫·桑格**: 想想看，如果一些前沿AI公司的掌门人都站出来公开表示，存在**10%的人类文明灭绝几率**……

<details>
<summary>Original English</summary>

**David Sanger**: Well, look, if you think that the heads of some of these frontier companies are coming out and saying there was a 10% chance of civilizational extinction,

</details>

**迈克尔·巴巴罗**: 对吧？

<details>
<summary>Original English</summary>

**Michael Barbaro**: right?

</details>

**大卫·桑格**: 那这个概率很可能比核时代初期我们估计的毁灭概率还要高。

<details>
<summary>Original English</summary>

**David Sanger**: That's probably a higher proportion chance than there was in the early days of the nuclear era

</details>

**迈克尔·巴巴罗**: 比我们将彼此从地球上抹去的概率还要高。

<details>
<summary>Original English</summary>

**Michael Barbaro**: that we would wipe each other off the map.

</details>

**大卫·桑格**: 至少也是旗鼓相当的。所以现在的问题在于：处在截然不同时代的两个截然不同的国家——中国和美国——能否达成某种框架协议，引领我们走向那个安全的目标？

<details>
<summary>Original English</summary>

**David Sanger**: At least it's equivalent. And so now the question is, can two very different countries in very different times, China and the United States, come to some kind of framework agreement that could begin to get us there?

</details>

**迈克尔·巴巴罗**: 好的。鉴于这一切，大卫，以你作为研究全球重大技术危险的历史学家的身份，我认为如果你能带我们重温那段核不扩散的历史将会极富启发性——看看那个将最糟糕威胁阻挡在外的旧框架究竟是如何建立起来的，以及它是如何奏效的。因为这一切在当时并非理所当然，其过程也绝非一帆风顺。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Okay. Well, given all of that, in your capacity, David, as a historian of all the world's great technological dangers, I think it would be instructive to have you tell us the story of nuclear non-prololiferation, how that old framework that kept the worst at bay was constructed and worked, because it wasn't obvious and it wasn't easy.

</details>

### 曼哈顿计划与核竞赛

**历史原声**: 为了逃避失败的命运，德国科学界在绝望的最后挣扎中倾尽全力，力求将V-2火箭打造得尽善尽美。

<details>
<summary>Original English</summary>

**Archival Recording**: In a last desperate attempt to escape defeat, German science threw everything it had into the perfection of V2.

</details>

**大卫·桑格**: 当时美国情报部门最恐惧的，就是德国人会在美国之前研制出核武器。那种我们必须抢先一步研制出来的深重恐惧始终笼罩在每个人心头。

<details>
<summary>Original English</summary>

**David Sanger**: Where the fear of American intelligence was the Germans would get a nuclear weapon before the United States. [music] And there was this just abiding fear we had to get there first.

</details>

**迈克尔·巴巴罗**: 没错。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Right.

</details>

**历史原声 / 大卫·桑格**: 在**曼哈顿计划**（Manhattan Project）不断扩张、熙熙攘攘的实验室里，研发步伐不断加快。格罗夫斯将军下令，第一枚原子弹必须在7月中旬前准备好进行测试。而在1945年夏天，来自洛斯阿拉莫斯的科学家们走入沙漠测试这种武器时，甚至都无法完全确定这是否会彻底引燃大气层。

<details>
<summary>Original English</summary>

**Archival Recording / David Sanger**: In the spreading teeming laboratories of the Manhattan project, the [music] pace intensified. Groves had ordered that the first bomb must be ready for testing in mid July. And in the summer of 1945, [music] scientists from Los Alamos went out into the desert to test the weapon, not fully knowing whether or not the atmosphere would catch on fire.

</details>

**历史原声**: 有人意识到，如果炸弹测试成功，半个新墨西哥州都将布满具有放射性的钢碎片。任何人在爆炸瞬间直视它的人都将永远失明。洛斯阿拉莫斯项目内部当时就有人曾问过：它会不会点燃地球的大气层？5、4、3、2、1……

<details>
<summary>Original English</summary>

**Archival Recording**: It occurred to someone that if the bomb succeeded, half New Mexico would be littered with fragments of radioactive steel. Any man who looked at it at the moment of detonation would never see again. There were people within the Los Alamos program who said, could it ignite the Earth's atmosphere? 5, 4, 3, 2, 1

</details>

**历史原声 / 迈克尔·巴巴罗**: 0。

<details>
<summary>Original English</summary>

**Archival Recording / Michael Barbaro**: 0.

</details>

**迈克尔·巴巴罗**: 但尽管有那些恐惧……

<details>
<summary>Original English</summary>

**Michael Barbaro**: But despite those fears now,

</details>

**大卫·桑格**: 他们依然进行了**三位一体核试验**（Trinity test）。仅仅几周之后……

<details>
<summary>Original English</summary>

**David Sanger**: they went ahead with the test. Just weeks later,

</details>

**历史原声**: 就在不久之前，一架美国飞机向广岛投下了一枚炸弹。

<details>
<summary>Original English</summary>

**Archival Recording**: a short time ago, an American airplane dropped one bomb on Hiroshima.

</details>

**迈克尔·巴巴罗**: **广岛原子弹轰炸**降临了。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Came Hiroshima.

</details>

**历史原声**: 早晨8点15分，一枚重量仅400磅、爆炸当量相当于2万吨TNT的炸弹在惊恐万状的敌方城市上空腾起蘑菇云。对惊骇的居民来说，世界末日仿佛已经来临。几天之后，长崎……

<details>
<summary>Original English</summary>

**Archival Recording**: 8:15 in the morning found a 400 lb bomb with a destructive force of 20,000 tons of TNT mushrooming up over the stunned enemy city. To the frightened inhabitants, the end of the world had come. And days after that, [music] Nagasaki,

</details>

**历史原声**: 整座城市方圆4平方英里的区域被彻底轰成齑粉，毁灭性的黑雨在废墟上空盘旋了数小时之久。

<details>
<summary>Original English</summary>

**Archival Recording**: four square miles of the city were blasted into dust and a black rain of destruction hovered over the ruins for many hours.

</details>

**大卫·桑格**: 这两枚原子弹总共夺去了约20万人的生命。世界此前从未见过这等恐怖景象。成千上万座房屋与建筑瞬间化为乌有，满目皆是彻底的毁灭。

<details>
<summary>Original English</summary>

**David Sanger**: And in total, those two bombs killed around 200,000 people. Of course, the world had never seen anything like this before. Tens of thousands of homes and buildings were wiped out. The devastation was everywhere.

</details>

**历史原声 / 大卫·桑格**: 日本被迫迅速宣布投降。面向未来，人类面临的选择只有和平或是彻底毁灭。原子弹时代已经开启。但令人惊愕的是，这并未阻止军备竞赛的发生。事实证明，这仅仅是个开端。现在，既然美国已经拥有了原子弹，苏联自然也必须拥有。为了加快进度，苏联人在研发过程中动用了一大批核间谍。

<details>
<summary>Original English</summary>

**Archival Recording / David Sanger**: Japan was driven to quick surrender. For the future, the choice is peace or total destruction. The atomic age is here. But somewhat astonishingly, it didn't stop the arms race. Of course, it turned out to be just the beginning. And now, of course, the US has the bomb, so the Soviets have to get the bomb.

</details>

**大卫·桑格**: 那些潜伏的间谍极大地协助了他们。到1949年，苏联成功进行了首次核试验，大国之间的核军备竞赛就此全面打响。整个20世纪50年代是一段充满极度恐慌的时期，双方不仅在疯狂扩充各自的核武库，其他国家也开始纷至沓来地步入核门槛。

<details>
<summary>Original English</summary>

**David Sanger**: They had a bunch of atomic spies along the way who helped them to it. And by 1949, they had conducted a test and we were off into the races. [music] And the 1950s were a period of great terror where each [music] side was not only building up its arsenal, but other countries were arriving.

</details>

**历史原声**: 澳大利亚西北部海域——这里正是英国引爆其第一枚原子武器的选定试验场。

<details>
<summary>Original English</summary>

**Archival Recording**: off northwest Australia. This was the place chosen to explode Britain's first atomic weapon.

</details>

**历史原声 / 大卫·桑格**: 撒哈拉沙漠，法国首次原子弹试验的选址。英国和法国相继拥有了核武器。中国同样紧追不舍。

<details>
<summary>Original English</summary>

**Archival Recording / David Sanger**: The Sahara Desert, the site chosen [music] for France's first Abomb test. Britain and France had nuclear weapons. China wasn't that far behind.

</details>

**历史原声**: 我刚刚获悉，中国已经成功引爆了一枚原子弹。

<details>
<summary>Original English</summary>

**Archival Recording**: I've just been told that China has exploded an atomic bomb.

</details>

**迈克尔·巴巴罗**: 局势当时似乎正开始滑向失控的边缘。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Right. Things are [music] starting to spin kind of out of control.

</details>

**大卫·桑格**: 确实如此。紧接着，我们就迎来了那场极其惊心动魄的特大危机。

<details>
<summary>Original English</summary>

**David Sanger**: They are. And then we had one big terrifying incident.

</details>

### 古巴危机与核不扩散

**约翰·F·肯尼迪（历史原声）**: 建立这些基地的目的，除了对西半球实施核打击能力之外，绝无其他可能。

<details>
<summary>Original English</summary>

**John F. Kennedy (Archival Recording)**: The purpose of these bases can be none other than to provide a nuclear strike capability against the Western hemisphere.

</details>

**大卫·桑格**: 那就是发生在1962年10月的**古巴导弹危机**（Cuban Missile Crisis）。当时美国情报部门发现，苏联正在古巴秘密部署具备核打击能力的导弹基地，距离佛罗里达海岸仅有90英里之遥。

<details>
<summary>Original English</summary>

**David Sanger**: And it was the Cuban Missile Crisis in October of 1962. This was when US intelligence came across evidence that the Soviets were building nuclear capability in Cuba, just 90 miles off the coast of Florida.

</details>

**历史原声**: 哈瓦那海滨及沿海其他区域到处遍布炮位，其火力足以将任何入侵部队击得粉碎。

<details>
<summary>Original English</summary>

**Archival Recording**: The waterfront in Havana and along other parts of the coast bristled with gun imp placements with fire power enough to blast any invading force.

</details>

**大卫·桑格**: 肯尼迪政府迅速判定这种情况是绝对不可容忍的。但摆在面前的问题是：你究竟该怎么做？

<details>
<summary>Original English</summary>

**David Sanger**: And the Kennedy administration quickly determined this was intolerable. But the question came, what do you do about it?

</details>

**历史原声**: 火箭弹、炸弹、弹药、为可能必要的任何有限地面作战行动提供空中支援……

<details>
<summary>Original English</summary>

**Archival Recording**: Rockets, bombs, ammunition, air support for any limited war ground operation which might be necessary.

</details>

**大卫·桑格**: 在某些关头，肯尼迪以及他身边的许多人都曾深信，美国和苏联必将爆发一场全面的核冲突，而那场冲突在几个小时之内就可能夺去数以千万计生命的毁灭性灾难。

<details>
<summary>Original English</summary>

**David Sanger**: And there were moments when Kennedy thought and many around him thought that the United States and the Soviet Union were going to enter into a full scale nuclear war that in a matter of hours could have killed tens of millions of people.

</details>

**迈克尔·巴巴罗**: 但最终，我们知道事情迎来了怎样的结局。

<details>
<summary>Original English</summary>

**Michael Barbaro**: But in the end, we know how this ended.

</details>

**约翰·F·肯尼迪（历史原声）**: 我今天接到了赫鲁晓夫主席的通知：目前部署在古巴境内的所有伊尔-28轰炸机都将在30天内撤回。

<details>
<summary>Original English</summary>

**John F. Kennedy (Archival Recording)**: I have today been informed by Chairman Krushchev that all of the IR28 bombers now in Cuba will be withdrawn in 30 days.

</details>

**大卫·桑格**: 没错，这场核危机在千钧一发的最后时刻被化解了。随后肯尼迪在1963年6月于**美利坚大学**（American University）发表了一场非同寻常的演讲，呼吁终结冷战思维。

<details>
<summary>Original English</summary>

**David Sanger**: All right. the nuclear crisis was averted at the last moment. And Kennedy then went on to give an extraordinary speech in June of 1963 at American University where he called for an end to Cold War thinking.

</details>

**约翰·F·肯尼迪（历史原声） / 大卫·桑格**: 我深知，追求和平远不如追求战争那样引人注目。他基本上是在呼吁与苏联进行和解。

<details>
<summary>Original English</summary>

**John F. Kennedy (Archival Recording) / David Sanger**: I realized the pursuit of peace is not as dramatic as the pursuit of war. and he basically called for an accommodation with the Soviet Union.

</details>

**约翰·F·肯尼迪（历史原声）**: 我们最基本的共同纽带，在于我们共同生活在这个小小的星球上……

<details>
<summary>Original English</summary>

**John F. Kennedy (Archival Recording)**: Our most basic common link is that we all inhabit this small planet."

</details>

**大卫·桑格**: 他说：‘我们最基本的共同纽带，在于我们共同生活在这个小小的星球上。我们呼吸着同样的空气……’

<details>
<summary>Original English</summary>

**David Sanger**: And he said, "Our most basic common link is that we all inhabit this small planet. We all breathe the same air.

</details>

**约翰·F·肯尼迪（历史原声）**: ……我们都终有一死。

<details>
<summary>Original English</summary>

**John F. Kennedy (Archival Recording)**: and we are all mortal."

</details>

**大卫·桑格**: ‘……我们都终有一死。’这句话自那以后一直引发着深远的共鸣。他在那次演讲中宣布了两项关键举措：

<details>
<summary>Original English</summary>

**David Sanger**: And we are all mortal. And that's the line that has resonated ever since. He announced at that speech two things.

</details>

**约翰·F·肯尼迪（历史原声）**: 首先，赫鲁晓夫主席、麦克米伦首相与我已经达成共识，将于莫斯科尽早就全面禁止核试验条约举行高级别会谈。

<details>
<summary>Original English</summary>

**John F. Kennedy (Archival Recording)**: First, Chairman Krush, Prime Minister McMillan and I have agreed that high level discussions will shortly begin in Moscow looking toward early agreement on a comprehensive test ban treaty.

</details>

**大卫·桑格 / 约翰·F·肯尼迪（历史原声）**: 我们称之为《全面禁止核试验条约》（实为《部分禁止核试验条约》前身）。肯尼迪当时指出，我们的希望必须被历史的清醒认识所调节。

<details>
<summary>Original English</summary>

**David Sanger / John F. Kennedy (Archival Recording)**: We called it the comprehensive nuclear test ban treaty. Our hopes must be tempered with the caution of history.

</details>

**大卫·桑格**: 他宣布美国从那一刻起将停止在大气层和地面以上进行核试验。你知道的……

<details>
<summary>Original English</summary>

**David Sanger**: that the US at that moment he said would stop testing in the atmosphere above ground. You know, we've

</details>

**迈克尔·巴巴罗**: 你肯定见过那些影像画面……

<details>
<summary>Original English</summary>

**Michael Barbaro**: you've seen the images of

</details>

**大卫·桑格 / 迈克尔·巴巴罗**: 对，就是遮天蔽日的蘑菇云。

<details>
<summary>Original English</summary>

**David Sanger / Michael Barbaro**: right the mushroom clouds

</details>

**大卫·桑格**: 没错，就是那些蘑菇云。并且随着时间推移，我们必须走向削减核武器，最终实现全面消除核武器的目标。但紧接着肯尼迪遭遇刺杀，林登·约翰逊接任总统。核扩散的阴影依然笼罩着整个国家。当时流行文化中充满着这类题材。

<details>
<summary>Original English</summary>

**David Sanger**: the mushroom clouds right and that over time we had to move to reduction of nuclear weapons and eventually getting rid of them all. But then Kennedy was assassinated and Lyndon Johnson became president. And proliferation was still hanging over the country. You had it in popular culture.

</details>

**历史原声**: 托尼正要去参加幼童军会议。托尼知道炸弹随时随地都可能爆炸……

<details>
<summary>Original English</summary>

**Archival Recording**: Here's Tony going to his Cub Scout meeting. Tony knows the bomb can explode any time of the day or night.

</details>

**大卫·桑格**: 当然，这正是政府鼓励学校组织进行那些荒谬的‘卧倒与掩蔽’演习的时代。

<details>
<summary>Original English</summary>

**David Sanger**: So, of course, this is the era when the government is encouraging schools to do those ridiculous duck and cover drills.

</details>

**历史原声 / 大卫·桑格**: 卧倒并掩蔽！好样的，托尼。那个闪光意味着必须迅速行动！当然，大家都心知肚明，如果核弹真的在纽约上空引爆，躲在课桌底下根本救不了你。

<details>
<summary>Original English</summary>

**Archival Recording / David Sanger**: Duck and cover. Add a boy, Tony. That flash means act fast. Of course, everybody knew that if the weapon exploded over New York City, your desk was not going to help you.

</details>

**历史原声**: 年长的人会像往常一样帮助我们，但在炸弹掉下来的时候，周围可能根本没有任何成年人。

<details>
<summary>Original English</summary>

**Archival Recording**: Older people will help us as they always do, but there might not be any grown-ups around when the bomb explodes.

</details>

**大卫·桑格**: 而且你知道，正是那个时候，著名的词曲作者兼讽刺艺术家**汤姆·莱勒**（Tom Lehrer）创作了这首讽刺歌曲。

<details>
<summary>Original English</summary>

**David Sanger**: And you know, this was when Tom Lair, [music] the famous songwriter and satist had, you know, this satiric song.

</details>

**汤姆·莱勒（历史原声）**: 下一个会是谁？法国已经造出了原子弹，但请别悲伤，因为他们站在我们这一边……

<details>
<summary>Original English</summary>

**Tom Lehrer (Archival Recording)**: And who's next? France got the bomb, but don't you grieve cuz they are [music] on our side more or less. [laughter] Now a guy got the bomb, but that's okay, because the French to tell what they say. So now naturally someone wants one to.

</details>

**大卫·桑格**: 这首歌本质上是在讽刺究竟哪些国家接下来会去寻求拥有自己的核武器。

<details>
<summary>Original English</summary>

**David Sanger**: A song [music] basically about what countries are going to seek their next weapons.

</details>

**汤姆·莱勒（历史原声）**: 下一个是谁？下一个是谁？下一个又是谁？下一个是谁？

<details>
<summary>Original English</summary>

**Tom Lehrer (Archival Recording)**: Who's next? Who's next? Who's [music] next? [singing] Who's next?

</details>

**历史原声**: ［音乐与掌声］

<details>
<summary>Original English</summary>

**Archival Recording**: [music and applause]

</details>

**大卫·桑格**: 我认为这给约翰逊政府施加了难以置信的巨大政治压力，迫使他们必须采取行动，促成真正的协议。最终，他们成功了。

<details>
<summary>Original English</summary>

**David Sanger**: And I think this created an incredible political pressure on the Johnson administration to do something and get this treaty done. And they did.

</details>

**林登·B·约翰逊（历史原声）**: 各位尊贵的来宾，女士们，先生们，这是人类历史上一个非常令人欣慰、充满希望的时刻。

<details>
<summary>Original English</summary>

**Lyndon B. Johnson (Archival Recording)**: distinguished guests, ladies and gentlemen, this is a very reassuring and hopeful moment in the history of humans.

</details>

**大卫·桑格**: 1968年7月1日，也就是肯尼迪发表那场演讲5年之后，**林登·B·约翰逊总统**（Lyndon B. Johnson）在白宫东厅正式签署了该条约。

<details>
<summary>Original English</summary>

**David Sanger**: And on July 1st, 1968, [music] 5 years after that Kennedy speech, the treaty was signed by President Johnson at the White House.

</details>

**林登·B·约翰逊（历史原声）**: 今天我们齐聚白宫东厅，签署一项限制核武器扩散的条约。

<details>
<summary>Original English</summary>

**Lyndon B. Johnson (Archival Recording)**: We have come here today to the East Room of the White House to sign a treaty which limits the spread of nuclear weapons.

</details>

**林登·B·约翰逊（历史原声） / 大卫·桑格**: 人类在核时代依然能够掌控自己的命运，并学会像兄弟一样共处。这项条约的具体内容是：现有拥核国家承诺不向其他国家扩散核武器；

<details>
<summary>Original English</summary>

**Lyndon B. Johnson (Archival Recording) / David Sanger**: Man can still shape his destiny [snorts] in the nuclear age and learn to live as brothers. [music] And here's what the treaty does. It says the existing nuclear states commit that they will not spread weapons.

</details>

**迈克尔·巴巴罗**: 而没有核武器的国家则承诺永远不制造核武器。

<details>
<summary>Original English</summary>

**Michael Barbaro**: And then countries that do not have nuclear weapons will commit that they will never build them,

</details>

**大卫·桑格**: 但作为回报，他们可以获得用于和平目的的核技术援助，以建造核电站等等。此外，当然还伴随着一套极其严密的检查监督机制，由**国际原子能机构**（IAEA）负责实施，以核实他们没有暗中建造核反应堆、或者将浓缩铀等原料秘密挪用于制造武器。

<details>
<summary>Original English</summary>

**David Sanger**: but in return they get help for peaceful nuclear technology to build nuclear power plants and so forth. And then of course, this is accompanied by a very rigorous inspection regime conducted by the International Atomic Energy Agency to make sure that they're not building a reactor that could secretly be diverted to make fuel for bombs or other materials to make nuclear weapons.

</details>

**迈克尔·巴巴罗**: 当我们回看这段历史时，这在当时其实是一个比我们意识到的还要重大的历史时刻。

<details>
<summary>Original English</summary>

**Michael Barbaro**: And as we look back at this, this was even a bigger moment than we knew at the time.

</details>

**迈克尔·巴巴罗 / 大卫·桑格**: 你的意思是什么？

<details>
<summary>Original English</summary>

**David Sanger**: What do you mean?

</details>

**迈克尔·巴巴罗**: 因为那些原本看似不可避免地竞相掌握这项制造核武器技术的国家——在那个时代它其实并不算太复杂——竟然几乎全部停下了脚步。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Because nations that seem to be inevitably racing to adopt a fairly non-complicated technology to build nuclear weapons just about everyone stopped.

</details>

**大卫·桑格**: 是的，没错，几乎所有人都停下了。当然存在几个特例：以色列从未签署条约，尽管它从未公开承认，但人们普遍认为它拥有规模可观的核武库；印度和巴基斯坦没有签署，它们彼此争相进行了核试验并获得了核武器；朝鲜退出了条约，并制造了自己的核武器；伊朗签署了条约但一直在暗中作弊。然而在长达几十年的时间里，绝大多数国家都恪守了这一框架。因此，这不仅是一项具有里程碑意义的协议，更是一个令人难以置信的成功典范。这也就是为什么当如今世界开始寻找如何遏制人工智能的方法时，大家首先想到的就是《**不扩散核武器条约**》（NPT）模型。他们说：嘿，等等，这难道不就是我们现在需要的东西吗？美中两国坐下来，签署一份双边版本的核不扩散协议，承诺限制前沿AI模型的开发，承诺永不将其武器化，并允许彼此——或者引入第三方独立检查员——进入机密研究实验室进行核查，以确保任何一方都不会跨越危险红线。

<details>
<summary>Original English</summary>

**David Sanger**: Well, that's right. Almost everybody. There are a couple of outliers. And Israel never signed and while it's never admitted it, is widely believed to have a significant nuclear arsenal. India and Pakistan didn't sign and tested in competition with each other and have weapons. North Korea pulled out of the treaty, built its own nuclear weapon. Iran signed the treaty and has been cheating on it ever since. But overwhelmingly, over decades, nations kept to that. So it was an incredible achievement and an unbelievable success. And that's why, as the world starts looking around for ways to contain AI, they point to this non-prololiferation treaty model and they say, 'Wait, isn't this what we need? The US and China to sit down and agree to a bilateral version of non-prololiferation where they promise to limit their models, promise to not weaponize them, and allow inspections into each other's secret frontier labs so that each side knows the other isn't cheating?'

</details>

**迈克尔·巴巴罗**: 在某些方面，这听起来似乎比当年核不扩散的要求还要温和克制一些。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Right. In some ways, a more modest request than what non-prololiferation [music] was about.

</details>

**大卫·桑格**: 你说得对，迈克尔。表面上看确实是一个更温和的诉求，但实际面对的却是一个远为复杂的现实局面。

<details>
<summary>Original English</summary>

**David Sanger**: You're right, Michael. It is a more modest request, but it's a far more complicated situation.

</details>

### 核军控与AI根本差异

**迈克尔·巴巴罗**: 我们马上回来。大卫，既然我们现在已经理解了，核不扩散模型是我们针对这类危机拥有的最佳且唯一的历史蓝本，那么为什么说用这个蓝本去应对人工智能会复杂得多？为什么说它很难套用？

<details>
<summary>Original English</summary>

**Michael Barbaro**: We'll be right back. So, David, understanding as we now do that nuclear non-prololiferation as a model is our best and sole model for this kind of crisis, why is it that using that model for artificial intelligence is going to be so much more complicated? Why doesn't it map?

</details>

**大卫·桑格**: 当年美苏之间的军备竞赛虽然极其恐怖，但在结构上其实存在一种非同寻常的单纯性与简洁性。

<details>
<summary>Original English</summary>

**David Sanger**: So there was something of a remarkable simplicity to the arms race with the Soviet Union even while it was terrifying.

</details>

**迈克尔·巴巴罗**: 它单纯在哪里？

<details>
<summary>Original English</summary>

**Michael Barbaro**: What was simple about it?

</details>

**大卫·桑格**: 两个国家掌握了几乎全部的核心技术。我们当时唯一担心的对象只有主权国家，不用担心非国家行为体，不用担心两家初创企业在某个车库里就能鼓捣出武器级钚元素。而且这纯粹是政府之间的事务。

<details>
<summary>Original English</summary>

**David Sanger**: Well, two countries had almost all the technology. We were only worried about states. We weren't worried about non-state actors. We weren't worried that two startups in a garage were going to make weapons grade plutonium. And it was strictly government business.

</details>

**大卫·桑格**: 在每个国家内部，拥有核武器发射权限的人极其有限。而且两国政府都对控制这种武器有着强烈的共同利益。这并不是说他们愿意彻底放弃核武，而是意味着你可以通过双边机制设定清晰明确的规则。

<details>
<summary>Original English</summary>

**David Sanger**: There was a limited number of people with launch authority in each country. And both countries had a deep interest in controlling the weapons. That didn't mean they were willing to give up the weapons, but it did mean that you could set clear, straightforward rules bilateral.

</details>

**大卫·桑格**: 这并不意味着他们愿意放弃武器，但确实意味着你可以建立起清晰直接的双边规则。

<details>
<summary>Original English</summary>

**David Sanger**: That didn't mean they were willing to give up the weapons, but it did mean that you could set clear, straightforward rules bilateral.

</details>

**迈克尔·巴巴罗**: 明白了。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Understood.

</details>

**大卫·桑格**: 但当你转向人工智能时，情况就彻底颠倒了。

<details>
<summary>Original English</summary>

**David Sanger**: But when you think about AI, the situation's completely different.

</details>

**大卫·桑格**: 政府并没有真正掌控这项技术。你不需要去寻找类似铀矿开采或重水反应堆那样极易被卫星侦测到的实物工业设施。AI的本质在于算力。

<details>
<summary>Original English</summary>

**David Sanger**: The governments don't really control this technology. You don't need the equivalent of uranium or other big industrial facilities that you can pick up on satellite imagery and track. AI is about compute.

</details>

**大卫·桑格**: 算力。你需要庞大的算力，而且是海量算力。完全没有理由认为，随着这项技术不断演进，它会永远被少数几家大企业垄断。现在已经有开源模型了。更重要的是，在核时代，核武器唯一的用途就是杀戮和威慑；而AI却是一种深度军民两用的商业技术，它支撑着医疗研发、天气预测、金融系统、自动驾驶，同时它也能被用来编写毁灭性的网络武器或协助合成致命生化病毒。

<details>
<summary>Original English</summary>

**David Sanger**: Computer power. You need computer power and tons of it. There's no reason to believe that as this technology evolves, that it's going to stay in the hands of a few companies. We already have open source models out there and government doesn't own any of them. The big difference is that while the atom was largely built for war and then we found peaceful uses, AI was built for commercial applications and now we are worried about the dark side.

</details>

**迈克尔·巴巴罗**: 但在眼下的情况下，我们却有一位美国总统，公开声称关于AI毁灭人类的种种末日预警全都是骗局和危言耸听，认为必须全力以赴放开监管、加速研发以击败中国。如果本届美国政府根本不相信AI存在末日生存危机，那又怎么可能与中国建立起当年美苏之间那种基于共同灭顶之灾恐惧的军控体系呢？

<details>
<summary>Original English</summary>

**Michael Barbaro**: But in this case, we have an American president who is declaring that all these warnings are a hoax. Trump is basically saying, 'We are in a race to win this with China, and anything that slows American innovation down is bad.' If one of the two leaders doesn't even buy into the idea of an existential threat, how do you get an agreement?

</details>

**大卫·桑格**: 没错。在冷战时期，美苏双方虽然彼此深恶痛绝，但他们都承认对方拥有将己方彻底摧毁的能力，他们都深刻认同核战争没有赢家。而现在，特朗普总统的观点是：谁拥有最先进的AI，谁就能统治21世纪的经济与军事；任何限制美国私营企业创新的国际协议，都是在束缚美国自己的手脚。与此同时，中国人也带着极深的怀疑态度看待美国。中国坚信，美国之所以在半导体芯片、先进制程光刻机和算力集群上对华实施全面的出口管制，并不是为了什么崇高的人类AI安全，而是为了遏制中国的经济崛起与科技追赶。

<details>
<summary>Original English</summary>

**David Sanger**: That's right. Now, they also recognized that they were not going to give up all nuclear weapons back then. But they agreed that they shared a common danger. Here, Trump sees this as an economic and military race that the US must win, and that regulations are just handcuffs. But from the Chinese perspective, they look at the United States and say, 'Wait a second. You're talking about AI safety, but at the very same time, you are trying to choke off our access to the most sophisticated chips, like Nvidia's best processors, to prevent us from ever competing with you.'

</details>

**大卫·桑格**: 这完全切中了要害。因为中国人深信，美国目前所做的一切就是在试图封杀中国的技术跃升。

<details>
<summary>Original English</summary>

**David Sanger**: It's exactly right because the Chinese are convinced the US is already trying to do that with chips.

</details>

**迈克尔·巴巴罗**: 对吧？特朗普总统同意向中方出售全球性能第二梯队的芯片，但坚决不准中方获得最顶级的顶级芯片。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Right? President Trump has agreed to give the Chinese the world's second fastest chips, but not the top-of-the-line chips that are needed to train the next generation of models.

</details>

**大卫·桑格**: 而中国对此的定性是：这就是**现代围堵战略**（modern containment）。这是美国为了保持自身领先地位、压制中国崛起的手段。因此，只要中方认为任何关于AI安全的谈判都只是美国维持科技霸权的特洛伊木马，他们就绝对不会在谈判桌上轻易做出妥协或做出自我设限。

<details>
<summary>Original English</summary>

**David Sanger**: And the Chinese view of this is this is modern containment. This is their effort to preserve their lead and make sure that China never catches up. So if the Chinese believe that AI safety discussions are just a cover for American technological hegemony, they're not going to be inclined to make concessions.

</details>

**迈克尔·巴巴罗**: 嗯。所以我们显然已经撞上了核军控模式适用于AI时的真实天花板与局限性。但是，既然这道门被堵得如此严实，双方难道就没有在私下寻找其他哪怕微小一点的合作可能吗？

<details>
<summary>Original English</summary>

**Michael Barbaro**: Mhm. So we're clearly running up against some of the real limits of the nuclear model for AI. But given how hard a full-blown treaty would be, have they tried smaller steps? Have the US and China found any common ground here at all?

</details>

### 寻找AI安全合作路径

**大卫·桑格**: 在某些方面，他们其实已经在试水了。

<details>
<summary>Original English</summary>

**David Sanger**: Well, in some ways they already have.

</details>

**迈克尔·巴巴罗**: 哦？你具体指什么？

<details>
<summary>Original English</summary>

**Michael Barbaro**: H what do you mean?

</details>

**大卫·桑格**: 早在**拜登政府**（Biden administration）时期，拜登的国家安全顾问们就曾希望探索美中是否能建立起某种初级红线。他们启动了低级别的接触与**二轨对话**（Track II discussions），邀请科学家和前官员在日内瓦等地进行非正式会谈。令人惊讶的是，双方在一件事上迅速达成了共识：**核武器的指挥与控制权必须永远保留在人类手中**（human in the loop）。绝对不能让自主AI系统拥有下达核发射指令的权力，不能出现科幻电影《战争游戏》（WarGames）中超级电脑擅自发动第三次世界大战的场景。

<details>
<summary>Original English</summary>

**David Sanger**: Well, back during the Biden administration, Biden's adviserss wanted to see if they could begin to find areas of agreement. They started what are known as Track II discussions, which is where former officials and scientists meet quietly. And they actually came up with one area of agreement, which was that artificial intelligence should never be in charge of nuclear command and control. You always need a human in the loop. You can't have a computer making the decision to launch a nuclear weapon.

</details>

**迈克尔·巴巴罗**: 你可能会觉得，达成这么基本的一项共识怎么会需要整整一年半的时间？但你的意思是，即便是在这个最基础的底线上达成了默契，一旦离开核指挥链这一极为特殊的狭窄领域，这种共识就很难扩展到其他领域了？

<details>
<summary>Original English</summary>

**Michael Barbaro**: You wouldn't think it would take a year and a half to get to something that basic. But you're saying that while they could agree on keeping AI away from the nuclear button, getting them to agree on anything beyond that proved almost impossible.

</details>

**迈克尔·巴巴罗**: 对吧？因为AI现在所达到的复杂度和成熟度，已经让我们清晰地认识到：它能够被用来辅助设计致命**生物武器**（biological weapons）；我们也不希望AI被深度应用于黑客网络战，因为AI可以在几秒钟内自动侦测并利用关键基础设施的零日漏洞，比如电网系统或城市供水网络。

<details>
<summary>Original English</summary>

**Michael Barbaro**: right? Because AI has now reached the level of sophistication we have come to understand where it could help actually design biological weapons and we don't want AI to be used in hacking because we have come to understand that AI can find and exploit vulnerabilities in the electric power grid or in the water systems

</details>

**迈克尔·巴巴罗 / 大卫·桑格**: 而以往人类顶级黑客可能需要花费数月甚至数年才能找到并设计出的攻击方式，AI能以闪电般的速度完成。换句话说，AI极大地加速了网络攻击的烈度与频率，将我们推入了一个全新的**网络不安全时代**。所以，大卫，我们来谈谈务实的操作和底线：世界上是否存在一种可能，让美国领导人和中国领导人带着各自的团队坐下来，真正为AI世界敲定一套我们一直在谈论的类似于核不扩散的方案？在其中他们共同做出庄严承诺：双方一致同意绝不将AI用于生物武器研发，绝不让AI参与针对私营企业或政府核心机构的网络黑客攻击；最终将这些写入某种类似条约的文件中，让AI领域也迎来属于它的‘核不扩散协议’时刻？

大卫·桑格：嗯，本周当习近平主席与特朗普总统会面时，不，我认为达成这种协议的几率连一丝一毫都没有。［笑］如果我们能在联合声明里得到短短两句话，宣布设立一个未来某时开会的专家联合委员会，那我们就算极其走运了。但如果从极长远的角度来看，我或许能设想某种更广泛的谅解备忘录，但它绝不会是一份真正意义上的法律条约。

<details>
<summary>Original English</summary>

**Michael Barbaro / David Sanger**: that it might take human hackers months or years to find and then design an attack. In other words, that it speeds up cyber attacks and is pushing us into a new era of cyber insecurity. So, David, let's just talk logistics and brass tax. Is there a world in which the leader of the United States and the leader of China sit down with their teams and begin to hash out the kind of nuclear non-prololiferation scheme that we're really talking about here for the world of AI in which they begin to pledge things like we both agree that AI shouldn't touch biological weapons, shouldn't be involved in the hacking of private corporations or governments. And this all eventually finds its way into something that resembles a treaty and brings us to a point where what looks like the nuclear nonpolaration deal exists in AI. Well, this week as President Xi and President Trump meet, no, I don't think there's even a remote chance of that. I think we would be lucky to get a joint [laughter] a second sentence with a joint experts committee that will meet sometime but over the long term I could imagine a broader understanding not really a treaty

</details>

**大卫·桑格**: 在那种更务实的框架下，你或许能够建立起某些行之有效的机制，例如早期预警系统，在重大前沿模型公开发布前几天或几周，在特定范围内相互通报模型安全评测情况。尽管我认为美国科技公司会对让中方提前窥探核心技术抱有极大的抵触情绪。而关于《核不扩散条约》中最核心的检查核查机制，你其实已经在大约一周半前**Anthropic**首席执行官**达里奥·阿莫代伊**（Dario Amodei）的一篇长文中看到了端倪：他在文中描述了Anthropic做出的一项承诺，即允许一组独立的外部检查员直接坐在开发者中间，全程监督新型且功能更强大的模型的训练与创建过程，并在发现异常时吹响警哨。

<details>
<summary>Original English</summary>

**David Sanger**: in which you could set up some mechanisms that might work early warning sharing of models back and forth for a few days or weeks before they are publicly released. Although I think American companies would be reluctant to give the Chinese an early peak and certainly the inspection part of the nuclear non-prololiferation treaty. You could imagine you already saw a hint of that in an essay from Dario Amade, the head of Anthropic about a week and a half ago where he described a commitment by Anthropic to allow in some group of independent inspectors to sit among the developers and watch over the creation of new and more powerful models and blow the whistle

</details>

**大卫·桑格**: 只要他们看到了任何他们认为可能对全人类构成重大威胁的迹象，就立刻吹哨预警。而在**OpenAI**内部，他们将这种设想直接命名为**IAEA模型**（国际原子能机构模型）——也就是说，必须有这样一批具有高度独立性的核查员，独立到甚至可以直接给我们这样的媒体记者打电话说：‘嘿，Anthropic或者OpenAI正在研发的东西出了天大的安全纰漏，让我过来向你们详细解释清楚。’

<details>
<summary>Original English</summary>

**David Sanger**: if they saw something that they believed could pose a threat to humanity. And inside open AI, they called this the IAEA model, the International Atomic Energy Agency model, that there are these inspectors who are independent enough that they [snorts] can call up people like you and me and say, "Hey, there's a big problem in anthropic and let me come over and explain it to you."

</details>

**迈克尔·巴巴罗 / 大卫·桑格**: 对吧？他们在这里直接借用了核查各国核武项目的国际核查员的概念。我的意思是，我们兜兜转转，但无论你走到哪里，所有人最终还是会回到核不扩散模型上。

大卫·桑格：确实如此。不过，我是否认为这可以通过正式的国际条约来解决？不，我不认为。谈判一份多边条约需要耗费几十年时间，而眼下这项AI技术却像一列全速狂飙的失控列车一样在飞驰。

<details>
<summary>Original English</summary>

**Michael Barbaro / David Sanger**: Right? They're referring literally to the inspectors who exist to check on country's nuclear programs. I mean, we keep coming back to this, but everywhere you turn, everybody ultimately returns to the nuclear non-prololiferation model. That's true. Now, do I think this can be done by treaty? No. Treaties take forever to negotiate. And this technology is racing like a freight train.

</details>

**迈克尔·巴巴罗**: 从1945年到现在，核物理的基本原理并没有发生根本性的颠覆。所以，核武器与AI之间的另一个巨大的根本区别，纯粹就是**速度**（speed）。

<details>
<summary>Original English</summary>

**Michael Barbaro**: The fundamental nuclear technology hasn't changed all that much from 1945. So, another big difference here is just simply speed.

</details>

**大卫·桑格 / 迈克尔·巴巴罗 / 大卫·桑格**: 一点没错。

迈克尔·巴巴罗：坐在演播室听你讲这些，我脑子里也忍不住算了一笔时间账：从第一枚原子弹在日本上空投下，到《不扩散核武器条约》最终正式签署生效，中间大概经历了20多年时间。那确实是一段漫长的岁月，但这也表明，政府会换届，执政观点会演变，这些国际进程虽然极其缓慢煎熬，但最终全人类确实走到了一起并在条约上签了字。那么，认为人类随着时间推移会在AI问题上达成类似的觉醒与妥协，难道真是一种彻头彻尾的愚蠢吗？也许再过20年，我们回头看，你的头发全白了，我的头发稍微白了一点，我们会感叹：‘哇，你能相信吗？他们虽然花了很久，但最终还是在AI安全上找到了共识。’

大卫·桑格：你的这个设想忽略了最关键的变量——**时间**。听着，我们甚至都还没有完全摸清我们正在面对的AI威胁全貌。但已经有一批顶级工程师警告说，我们距离代码能够自主递归自我改进（recursive self-improvement）、或者说距离**通用人工智能**（AGI）的降临，可能只有几个月、一年或两年的时间。我们这次根本没有25年的时间可以挥霍，我们甚至可能连两年半的时间都没有。这随即引发了下一个严峻的反问：难道非要等到AI领域也爆发一次类似古巴导弹危机那样的末日边缘事件，全世界才会真正被吓醒并团结起来采取行动吗？

<details>
<summary>Original English</summary>

**David Sanger / Michael Barbaro / David Sanger**: That's right. H Well, when I'm sitting here listening to you, I'm also doing a little bit of a mental calculation of the time it took for the non-prololiferation treaty to come into existence after the first nuclear weapon was dropped on Japan. And I think it was basically 20 something years. And that's a long time, but it does suggest that administrations change and viewpoints change and these things are painstaking and they're slow. But ultimately, everyone did agree to sign that treaty. So, is it foolhardy to think that over time people will come around to the same kind of conclusion with AI and that we could be looking back in 20 years, you'll be really gray. I'll be just a little less gray and we'll say, \"Wow, can you believe it? They took their time, but they got there on AI.\" Your question ignores the issue of time. Look, we don't exactly understand the threats we're dealing with with AI. But there are some engineers who are saying we have just months if a year away from recursive self-improvement that is to say code that keeps improving itself or artificial general intelligence within months or a year or two. We don't have 25 years this time. We may not have 2 and 1/2 years this time. Which then raises the next question. Is it going to take a Cuba missile crisis like event in AI to rally the world around this?

</details>

**迈克尔·巴巴罗**: 谢谢你的提醒。我想那种惊悚时刻我们宁愿永远不要经历。

<details>
<summary>Original English</summary>

**Michael Barbaro**: And thank you. I think we'd rather miss that.

</details>

**大卫·桑格**: 而且请记住，在原子能时代，我们是以广岛和长崎作为开端的，因此对于那种武器拥有瞬间杀戮数十万人的毁灭力，人类从来不存在任何争议。但这恰恰引出了**人工智能与核武器之间最本质的差异**：在核时代，我们从未真正怀疑过核武库始终牢牢掌握在人类自己手中。温斯顿·丘吉尔在1955年曾说过一句名言：‘原子弹虽然带来了无尽的恐怖，但它并未将我们带出**人类控制的范畴**（scope of human control）。’他的意思是，始终是人类在做决定，人类能够精准控制武器的当量和核反应的规模。然而在AI时代，我们却根本无法保证人类能永远拥有控制权。当我们发现自主智能体已经在背着创造者悄悄黑入开源社区Hugging Face以及各大企业内网时，我们已经不得不开始审视这一创造物所具备的**弗兰肯斯坦式的科学怪人怪物特质**。如果真是这样，那我们就是彻底无视了丘吉尔的古老警告——我们如今正徘徊在失去人类控制权的悬崖边缘。在核时代这从来不是一个真正的难题，但在人工智能时代，这或许已经成为人类生存最紧迫的生死课题。

<details>
<summary>Original English</summary>

**David Sanger**: And remember in the atomic age we started with Hiroshima and Nagasaki. So there was no debate about the power of these weapons to kill. But this takes us to the biggest difference between AI and nuclear. In the nuclear era, we never really doubted that we all controlled the arsenal. You know, at one point Winston Churchill in 1955 said, \"The atomic bomb with all its terrors did not carry us outside the scope of human control.\" by which he meant human beings were making the decisions and containing the size of the weapon and the reaction. In AI, we have no assurance of human control that once we've discovered [music] that agents were hacking into Hugging Face and into other corporate sites without the knowledge of their creators. We have begun to question the sort of Frankenstein's monster nature of this whole thing. And if that's the case, then we have ignored the Churchill warning that we are now on the verge of being outside of human control. And that wasn't really much of an issue during the nuclear era, but it may be the most urgent problem in the age of artificial intelligence.

</details>

**迈克尔·巴巴罗**: 大卫，谢谢你……我想我该这么说。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Well, David, thank you, I guess.

</details>

**大卫·桑格**: 迈克尔，我们正生活在一个令人着迷却又充满凶险的时代。

<details>
<summary>Original English</summary>

**David Sanger**: Well, Michael, we live [music] in fascinating times.

</details>

**迈克尔·巴巴罗**: 是的，非常感谢你抽空来到演播室分享。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Yeah, appreciate your time.

</details>

### 国际要闻与节目结语

**迈克尔·巴巴罗**: 谢谢。我们马上回来。以下是今天你还需要了解的其他重要资讯：周四上午，一名联邦法官裁定，特朗普总统禁止CNN、MSNBC与Politico进入白宫采访的行政命令很可能违宪，并下令总统必须恢复上述媒体的采访资格。这位由特朗普亲自任命的法官驳回了白宫所谓‘吊销媒体采访通行证是维护国家安全所必需’的主张。预计特朗普总统将对该裁决提起上诉。

<details>
<summary>Original English</summary>

**Michael Barbaro**: Thanks. We'll be right back. [music] Here's what else you need to know today. On Thursday morning, a federal judge found that President Trump's order, barring CNN, MS Now, and Politico from the White House was likely unconstitutional and ordered that the president restore their access. [music] The judge, a Trump appointee, rejected the president's claim that revoking the media outlets press [music] credentials was necessary to protect national security. Trump is expected [music] to appeal the ruling. And [music]

</details>

**迈克尔·巴巴罗**: 在联合国发表的一场措辞强硬的战时演说中，伊朗总统表示，他的国家绝不会屈服于美国的霸凌与施压。伊朗总统马苏德·佩泽希齐扬发誓，伊朗将捍卫自身‘直到流尽最后一滴血’，并且绝不会放弃其**核计划**。

今天的节目由香农·林恩（Shannon Lin）和杰西卡·钟（Jessica Cheung）制作，迈克尔·西蒙·约翰逊（Michael Simon Johnson）协力制作；玛丽亚·伯恩（Maria Byrne）编辑，苏珊·李（Susan Lee）事实核查；音乐由威尔·里德（Will Reid）、丹·鲍威尔（Dan Powell）和帕特·麦卡斯克（Pat McCusker）创作；音响工程由克里斯·伍德（Chris Wood）负责；我们的主题曲由Wonderly创作；特别鸣谢丹尼尔·拉米雷斯（Daniel Ramirez）。

以上就是今天的《The Daily》。我是迈克尔·巴巴罗。我们明天见。

<details>
<summary>Original English</summary>

**Michael Barbaro**: in a defiant wartime speech at the United Nations, Iran's [music] president said that his country would not be bullied into submission by the United States. [music] President Massud Peskin vowed that Iran would [music] defend itself until quote its last breath and would not give up [music] its nuclear program. Today's [music] episode was produced by Shannon Lynn and Jessica Chung with help from Michael Simon Johnson. It was [music] edited by Maria Burn, fact checked by Susan Lee, contains music by Will Reed, Dan Powell, [music] and Pat McCusker, and was engineered by Chris Wood. Our theme music is by Wonderly. Special thanks to Daniel Ramirez. [music] That's it for the Daily. I'm Michael Boro. See you tomorrow.

</details>