马斯克给 OpenAI 算过一次命:成功率 0%。
那是 2018 年前后的事。他在 2015 年参与创立了 OpenAI,出了钱,也出了名声。三年后他判断这家机构走不通,决定撤出去做自己的事。临走前留下的判断是,你们这条路成功的概率是零。
七年过去,OpenAI 成了全球估值最高的 AI 公司之一,ChatGPT 的周活跃用户到了几亿的量级。
2025 年,马斯克把 OpenAI 告上法庭。塔克·卡尔森(Tucker Carlson)把萨姆·奥特曼(Sam Altman)请到镜头前,问他这场官司到底是怎么回事。奥特曼的回答构成了这份文字稿的前半段;后半段,他谈的是更远的事:AI 会取代谁,会不会取代人本身。
整段访谈里没有火药味。奥特曼说话的节奏很慢,用词克制,回答问题时会先给自己留退路。这种语气本身就有信息量。
他给 OpenAI 判过死刑:成功率 0%
卡尔森的开场问得很直接:马斯克攻击你,这场争端的核心是什么?
奥特曼先给了个前提:马斯克帮我们创办了 OpenAI,我对此非常感激。
然后他把故事讲成了一条时间线。马斯克后来认定 OpenAI 走不通,告诉他们成功率是零,自己要去做竞争性的事。再后来 OpenAI 做得还不错,马斯克"可以理解地感到不快"。
"换我在那个位置,我也会不好受。"(萨姆·奥特曼)
再往后,用奥特曼的话说,马斯克搞了一个竞争性的"克隆",然后开始想办法拖慢 OpenAI,起诉,以及各种别的动作。
这套叙述有个特点:他把马斯克的情绪解释成失望,而不是恶意。一个人押注了很久的东西,被自己判了死刑,结果活了下来还活得很好,这种别扭是人之常情。
奥特曼也没回避自己这一方的立场。他说"这是我版本的经过,他肯定有另一个版本"。
这里值得停一下。同一段历史,两个人讲出来会是两个故事。马斯克那边的版本大概率是这样的:我出了钱,我提了建议,我没拿到控制权,我走了,然后这家标榜"开放"的机构变成了微软的附庸,还改了章程。这两种叙述未必有一个是假的,它们只是各自挑了对自己有利的那一段。
读者能拿到的确定信息很有限。能确定的只有:2015 年那笔钱是真的,2018 年的离开是真的,2023 年 xAI 的成立是真的,2024 年到 2025 年的诉讼是真的。剩下的都是各自的口述。
从"人类的宝石"到"我不欣赏的特质"
卡尔森追问了一句:"你现在对他是什么感觉?不再是人类的宝石了?"
"人类的宝石"(jewel of humanity)这个词是奥特曼自己用的。他说自己很长一段时间把马斯克当作英雄,一个了不起的人,人类的一颗宝石。
然后他补了一句:现在感觉不一样了。
被追问到底是什么感觉,他的回答很小心:
"他身上有一些了不起的地方,我感激他做过的很多事。也有很多地方,我认为是我并不欣赏的特质。"(萨姆·奥特曼)
注意这个句子的结构。他否认的不是马斯克的成就,是"特质"。而且他说的是"我不欣赏",没有说"坏"。
这种措辞在公开场合是精心打磨过的。既表达了疏远,又不给对方留下可以截屏攻击的把柄。
卡尔森问他们现在还联系吗。奥特曼说,很少。
从"人类的宝石"到"很少联系",中间隔了七年,隔了一家公司,隔了一场官司。
AI 比人聪明之后,人还算不算世界的中心
话题从人事转到正事。卡尔森抛出的问题其实是个哲学问题:如果 AI 已经比任何一个人都聪明,如果它将来比人更明智,能做出比人更好的决定,那它按定义就把人从世界的中心挪开了。
奥特曼不同意这个推论。他的回答分两层。
第一层是体验层面:它不会让人感觉像一个有自主意志的存在。它更像一台极其聪明的电脑,给我们建议。我们有时候听,有时候不听。它不会削弱我们对自身能动性的感受。
第二层是现实层面:这件事已经部分发生了。很多人用 ChatGPT 的时候会说,它在几乎所有事情上都比我聪明。但他们仍然在做决定,决定问什么,决定听哪句,决定哪句不听。
他把这个叫"技术的形状"(the shape of technology)。
这个回答有道理,也有一个没被追问的缝隙。
有道理的地方在于,"聪明"和"有权"是两回事。一个顾问可以比老板聪明一百倍,最后签字的还是老板。
缝隙在于:如果这个顾问每次都对,老板每次都听,那"老板决定听"这个动作还剩多少实质内容?一个人连续十年完全听从另一个存在的建议,我们通常不会说这个人保持了主体性。
奥特曼的答案是"我们有时候听有时候不听"。这个"有时候"是关键。它成立的前提是,人还有能力判断什么时候该不听。这个能力靠什么维持?训练?经验?还是说,用进废退,越依赖越退化?
访谈里他没回答这个。这是整段对话里最值得往下追的一处。
客服会消失,护士不会
卡尔森问得更具体了:谁会因为这个技术丢掉工作?
奥特曼先设了一道防线:没人能预测未来。他说如果自己试图精确回答,会说出一堆蠢话。然后他划了两块地盘,一块有把握,一块没把握。
有把握的一块:现在通过电话或网页完成的大量客服工作,会被 AI 接管。这些岗位的人会失业,而且 AI 会做得更好。
他补了一句限定:也有些客服场景,你确实希望确认对面是个真人。
有把握的另一块,方向正好相反:护士不会受太大影响。理由是,人在那个时刻需要和一个真实的人建立深度的连接,AI 或机器人的建议再好,也替代不了这一点。
这两个例子放在一起,划出了一条线。
线的左边是信息传递:查订单、改地址、解释政策、回答常见问题。这类工作的核心是把正确的信息从 A 挪到 B,AI 在这件事上比人快、比人便宜、比人有耐心。
线的右边是关系在场:一个人躺在病床上,需要的不是正确答案,是有人握住他的手。
这条线不是新的。几十年来,自动化一直在沿着这条线往右推。ATM 吃掉了取款,没吃掉柜员的咨询;电商吃掉了门店,没吃掉理发。每一次技术跃迁,这条线都会往右挪一截。
值得留意的是挪动的速度。这一轮快得多。
程序员这个答案,奥特曼自己也不确定
第三个例子,奥特曼明显犹豫了。
他说自己对程序员的未来远没有那么确定。理由是一条正在发生的事实:今天"程序员"这个职业的含义,和两年前已经完全不一样了。
两年前,写代码主要是人写。现在,人用 AI 工具,生产力被放大好几倍。但位置上还坐着一个人,他能产出比过去多得多的代码,也因此赚到比过去更多的钱。
为什么没出现大规模的替代?奥特曼给了一个经济学解释:世界对软件的需求,一直远远超过世界的产能。这个缺口存在了几十年。AI 补上的是供给端,而需求端还有个巨大的坑等着填。他把这个叫"需求积压"(demand overhang)。
这个观察很扎实。软件行业几十年的瓶颈从来不在需求侧。企业想数字化的东西排着队,只是没人写。AI 松开的是供给端的阀门,于是短期内所有增量都能被吞咽掉,甚至还不够。
但奥特曼马上补了一个转折:快进五年到十年,情况会怎样?工作变多还是变少?他不知道。
他确定的是中间会发生什么:大规模的职业迁移(massive displacement)。
这段话的分量在于它出自谁的口。亲手做这件事的人说"我不知道十年后会怎样",比任何外部评论都更值得当真。
每 75 年,一半工作换一次
卡尔森抛出一个数据:有人告诉他,历史上平均每 75 年,大约 50% 的工作会发生显著变化。不一定彻底消失,但内容会变得面目全非。他把这个叫"半衰期"。
奥特曼接住了这个词,然后给了一个他自称"有争议"的判断。
这个判断有两半。
前半:这会是一个"间断平衡"(punctuated equilibrium)时刻。原本摊在 75 年里的变化,会压缩到很短一段时间里集中发生。
后半:但把镜头拉远,总量上未必和历史速率差太多。短期内剧烈,长期看总的工作更替比我们想象的要少。
"间断平衡"是演化生物学里的概念。物种的形态长期稳定,然后在很短的地质时间里突然变化,再进入新的稳定期。奥特曼把它借到就业市场上,意思很清楚:曲线不是平滑上升的,是长时间的平,加上短时间的陡。
陡的那一段有多陡,决定了这件事的难度。如果 75 年的量压缩到 10 年,等于每年 5% 的岗位被重塑。看起来不多,但换算成人,是每年上千万人要重新学一门手艺。
奥特曼接着给了个历史类比。他说 200 年前不存在"经营一家科技公司"这种工作,今天有了。也有很多工作和 200 年前方向类似,只是工具换了。反过来,200 年前很常见的工作,今天已经没了。
然后他做了个假设性的推演:如果确实是每 75 年 50% 的更替,那 75 年后,一半人在做今天完全想象不到的事,另一半人在做和今天差不多的事。这个画面他觉得完全可信。
注意他用词的谨慎。整段话里他反复说"我不知道这是不是真的""只是拿这个数字打个比方"。一个掌管着最前沿模型的人,在谈预测时把不确定性标得这么清楚,这件事本身说明了一些东西。
上一次工业革命,带来了革命和世界大战
卡尔森的问题很尖锐:上一次工业革命的时候,伴随的是革命和世界大战。这次会不会也有?
奥特曼先说没把握。然后给了他的直觉:今天的世界比工业革命时期富太多,能吸收变化的速度也快得多。
他紧接着补了一句更重要的:很多东西和钱、和工作无关。意义、价值、社群,这些东西撑不住的时候,钱多也没用。
然后是这段访谈里最沉的一句:
"不幸的是,社会在这件事上已经处在相当糟糕的位置了。我不确定还能坏到哪去。当然,它可能更坏。"(萨姆·奥特曼)
这句话值得单独拎出来。一个掌管着全球最热门 AI 公司的人,在谈社会韧性时,给出的判断是"我们已经很糟了"。
这不是修辞。过去十几年,发达国家的孤独感、药物滥用、自杀率、政治极化,都在往上走。这些指标测的恰恰是他说的那三样东西:意义、价值、社群。
如果这三样本来就在塌,那 AI 带来的冲击落在什么地基上,答案不太乐观。
新冠给他的信心
不过奥特曼在这件事上是乐观的。他提到一个让他改观的经历:新冠。
2020 年 3 月,世界在一周之内变得完全不同。他当时非常担心社会撑不住,结果显然不完美,但整体上算是过关了。人们很快找到了新的生活方式。
他把这算作"社会韧性"的一个加分点。而 AI 带来的变化,不会有那么突然。
这个类比有一定道理,也有一个不太对等的地方。
新冠的特点是外部冲击。街道封了,公司关了,但社会结构本身还在,家庭还在,社区还在,人们知道敌人是谁,也知道它终会过去。
AI 不一样。它不封街道,它悄悄改掉每一份工作的内容。没有明确的开始时间,也没有明确的结束时间。它给人的不是休克,是持续的低烧。
休克能激发应急反应,低烧不会。这是新冠经验未必能直接搬到 AI 上的原因。人在急诊室里能爆发出来的适应力,和在一间慢慢变空的办公室里能爆发出来的,不是一个量级。
他最怕的是"未知的未知"
卡尔森问 downside。
奥特曼的回答从一句自我描述开始:他天生最担心"未知的未知"。
然后他分了两类。
第一类是已知的已知。这些模型在生物领域的水平正在快速提升,它们可能被用来设计生物武器,或者造出另一场新冠级别的大流行。他担心这个。
但他马上补了一句:正因为担心,他和业界很多人一直在想怎么防。
这句话的逻辑值得注意。一个风险被明确命名、被公开讨论、被行业当成议题列上日程,它造成灾难的概率就下降了。已知的已知,是可以被管理的。真正难的是另一种。
一个关于长横线的例子
第二类是未知的未知。奥特曼举的例子,是这段访谈里最有意思的一段。
他说,这类风险里有一种是"社会规模的效应":非常多的人,在同一时间,和同一个模型说话。
然后他讲了个具体的事。
大语言模型有自己的一套说话风格。某种节奏,某种略微不寻常的用词,还有,可能过度使用了长横线。他最近注意到,真人也开始这么说话了。
"我当时想过 ChatGPT 会让现实里的人用更多长横线吗?当然没有。这不是什么大事,但它是'未知的未知'的一个例子。这是个美丽新世界。"(萨姆·奥特曼)
这个例子小得几乎可笑,但它的指向一点都不小。
语言是思维的载体。你用什么句式组织一句话,反过来会影响你怎么切分这个世界。一个群体共用一套句式,共用久了,思考的路径也会趋同。
现在趋同的只是标点符号和几个连接词。但机制已经跑通。几亿人同时和同一个"语言中枢"对话,这件事在人类历史上没有先例。
过去的语言演化是分散的。不同地区、不同阶层、不同圈子,各自长出各自的说法,然后缓慢交融。现在多了一个中心化的节点,它同时向所有人输出。
今天输出的是标点习惯。明天可能是判断框架,是情绪基调,是什么值得愤怒、什么不值一提。
奥特曼没说这些。他只说了长横线。但从长横线到判断框架,中间没有断点,只有量的差别。
这也解释了为什么他把它归到"未知的未知"。已知的风险可以列清单,可以立法,可以设红线。这种风险没法列,因为它还没发生,甚至还没被想出来。等到它显形的时候,它已经在几亿人的语言习惯里扎根了。
还有一层更麻烦的。这类效应的因果链太长,长到没法归责。没人能证明某一年社会情绪的转向,源于某个模型在某个版本里的措辞偏好。没有证据,就没有约束。没有约束,就只会继续。
他给的答案里,没有一个是确定的
把整段访谈拉通看一遍,会发现一个明显的模式。
问谁会失业,他说客服会、护士不会、程序员不知道。问工作总量会怎样,他说短期剧变、长期差不多,但"我不确定"。问会不会重演工业革命那场动荡,他说"我没把握,而且社会本来就已经很糟了"。问最大的风险是什么,他说是那些我们还没想到的风险。
一个站在技术最前沿的人,每个关键问题给出的都是"我不知道"。
这种诚实不多见。绝大多数行业领袖在这种场合会给出一个乐观的、可引用的、听起来很有把握的答案。奥特曼没有。他反复给自己划出"有把握"和"没把握"两块区域,然后老老实实地把大多数问题放进后一块。
但它也说明一件事:没有人手里有地图,包括造地图的人。
这不必然是坏消息。没有人能预测,意味着也没有人被锁死。历史上每一次技术跃迁,最终落在哪,都是无数个体在具体处境里一次次选择的结果,而不是某个人提前规划好的。
马斯克那场官司会有结果,法院会给个说法,那部分不用你操心。真正和你有关的是后半段那些没有结论的部分。它们没有结论,是因为还没有人去做那些选择。
- 先给自己的岗位做一次分类。它是"把信息从 A 挪到 B",还是"在关系里提供在场"?前者在未来五年会持续被压价,后者会持续溢价。这不意味着前者就该辞职,意味着前者必须往前者的上游走:从执行挪到判断,从判断挪到定义问题。
- 别把"会用 AI 工具"当护城河。奥特曼说得很清楚,今天程序员的生产力提升来自工具,但位置上还坐着一个人。工具是默认配置了。真正稀缺的是决定做什么,以及决定什么不做。
- 关注需求侧,别只盯供给侧。AI 让供给变便宜,值钱的东西会往"发现还没被满足的需求"那边移动。奥特曼说的"需求积压",在软件行业存在了几十年,在其他行业同样存在。找到那个坑,比提高自己的产出速度更重要。
- 留意自己的语言。翻一翻你最近写的邮件、发的帖子、做的笔记,看有多少句子是模型帮你组织的。这个习惯本身无害,但它是"社会规模效应"在你身上的一个切片。偶尔关掉它写点东西,保持自己的表达肌肉不萎缩。
- 给"未知的未知"留制度性冗余。具体做法很朴素:手上有一笔能撑一年的现金,身上有一门不依赖当前雇主的技能,身边有几个不在同一行业的朋友。这三样东西不会让你预测未来,会让你在预测失败时还站得住。
- 如果你是管理者,把"哪个岗位会消失"这个问题换成"哪个岗位的内容会先变"。前者没法回答,后者可以拆解到具体任务。拆解完了,你会发现真正要重新安置的是任务,不是人。
- 护士那个例子值得抄作业。奥特曼说这个岗位不会被替代,不是因为技术做不到,是因为人不要。任何时候,只要你的工作里有一部分是"对方明确希望是个真人",这部分就有价格。找到它,放大它。
0:00I have to ask your version of Elon Musk has like attacked you and all this. Is what is the core of that dispute from your perspective? Look, I know he's a friend of yours and I know what side you'll I actually don't have a position on this because I don't understand it well enough to understand.
0:18helped us start OpenAI. I'm very grateful for that. I really for a long time looked up to him as just an incredible hero and you know, great jewel of humanity. I have different feelings now. What are your feelings now? No longer jewel of humanity? There are things about him that are incredible and I'm grateful for a lot of things he's done. There's a lot of things about him that I think are traits I don't admire.
0:44Anyway, he helped us start OpenAI and he later decided that we weren't on a trajectory to be successful and he didn't want to you know, he kind of told us we had a 0% chance of success and he was going to go do his competitive thing and then we did okay and I think he got understandably upset. Like I'd feel bad in that situation and since then has just sort of been trying to he had run a competitive kind of clone and has been trying to sort of slow us down and sue us and do this and that and That's kind of my version of it. I'm sure he'd have a different one.
1:19You don't talk to him anymore? Very little. If AI becomes smarter, I think it already probably is smarter than any person and if it becomes wiser, if we can agree that it reaches better decisions than people, then it by definition kind of displaces people at the center of the world, right?
1:43I don't think it'll feel like that at all. I think it'll feel like a you know, really smart computer that may advise us and we listen to it sometimes, we ignore it sometimes. It won't I don't think it'll feel like agency. I don't think it'll diminish our sense of agency. people are already using ChatGPT in a way where many of them would say it's much smarter than me at almost everything. But they're still making the decisions. They're still deciding what to ask, what to listen to, what not. And I think this is sort of just the shape of technology.
2:12Who loses their jobs because of this technology? I'll caveat this with the obvious but important the statement that no one can predict the future and I will in trying to if I try to answer that precisely, I will make a lot of I will say like a lot of dumb things, but I'll try to pick an area that I'm confident about and then areas that I'm much less confident about. I'm confident that a lot of current customer support that happens over a phone or computer, those people will lose their jobs and that'll be better done by an AI.
2:49Now, there may be other kinds of customer support where you really want to know it's the right person. A job that I'm confident will not be that impacted is like nurses. I think people really want a deep human connection with a person in that time and no matter how good the advice of AI is or the robot or whatever, like you'll really want that.
3:08A job that I feel like way less certain about what the future looks for looks like for is computer programmers. What it means to be a computer programmer today is very different than what it meant 2 years ago. You're able to use these AI tools to just be hugely more productive. But it's still a person there and they're like able to generate way more code, make way more money than ever before and it turns out that the world wanted so much more software than the world previously had capacity to create that there's just incredible demand overhang.
3:37But if we fast forward another 5 or 10 years, what does that look like? Is it more jobs or less? That one I'm uncertain on. But there's going to be massive displacement and maybe those people will find something new and interesting and you know, lucrative to do, but what how big is that displacement do you think?
3:53Someone told me recently that the historical average is about 50% of jobs significantly change. Maybe they don't totally go away, but significantly change every 75 years on average. That's the kind of that's the half-life of stuff. And my controversial take would be that this is going to be like a punctuated equal moment where a lot of that will happen in a short period of time, but if we zoom out, it's not going to be dramatically different than the historical rate. Like we'll do we'll have a lot in this short period of time and then it'll somehow be less total job turnover than we think. There will still be a job that is There there will be some totally new
4:34categories like my job, like, you know, running a tech company and would have been hard to think about 200 years ago. But there's a lot of other jobs that are directionally similar to jobs that did exist 200 years ago and there's jobs that were common 200 years ago that now aren't. And if we again, I've no idea if this is true or not, but I'll use the number for the sake of argument. If we assume it's 50% turnover every 75 years, then I could totally believe a world where 75 years from now half the people are doing something totally new and half the people are doing something that looks kind of like some jobs of today. I mean, last time we had an industrial
5:10revolution, there was like revolution and world wars. Do you think we'll see that this time? We've got a new partner. It's a company called Cowboy Colostrum. It's a brand that is serious about actual health and the product is designed to work with your body, not against your body.
5:28It is a pure and simple product, all natural. Unlike other brands, Cowboy Colostrum is never diluted. It always comes directly from American grass-fed cows. There's no filler, there's no junk. It's all good. It tastes good, believe it or not. So before you reach for more pills for every problem that pills can't solve, we recommend you give this product, Cowboy Colostrum, a try. It's got everything your body needs to heal and thrive. It's like the original superfood loaded with nutrients, antibodies, proteins. They'll build a strong immune system, stronger hair, skin, and nails. I threw my wig away. I'm right back to my natural hair after using this product. You just take
6:08a scoop of it every morning in your beverage, coffee, or a smoothie and you will feel the difference every time. For a limited time, people listen to our show get 25% off the entire order. So go to cowboycolostrum.com, use the code Tucker at checkout, 25% off when you use that code, Tucker at cowboycolostrum.com.
6:26Remember you mentioned you heard it here first. I Again, no one knows for sure. I'm not confident on this answer, but my instinct is the world is so much richer now than it was at the time of the industrial revolution that we can actually absorb more change faster than we could before. There's a lot that's not about money of job. There's meaning, there's value, there's community.
6:49I think we're already unfortunately in society in a pretty bad place there. I'm not sure how much worse it can get. I'm sure it can, but I I I have been pleasantly surprised on the ability of people to pretty quickly adapt to big changes. like COVID was an interesting example to me of this where the world kind of stopped all at once and the world was like very different from one week to the next and I and I was very worried about how society was going to be able to adapt to that world. And it obviously didn't go perfectly, but on the whole I was like, all right, this is one point in favor of societal resilience and people find, you know, new kind of ways to live their
7:28lives very quickly. I don't think AI will be that nearly that abrupt. So what will be the downside I mean, I can see the upsides for sure. Efficiency, medical diagnosis seems like it's going to be much more accurate. Fewer lawyers. Thank you very much for that. But what will What are the downsides that you worry about?
7:49I I think this is just like kind of how I'm wired. I always worry the most about the unknown unknowns. If it if it's a if it's a downside that we can really like be confident about and think about, you know, we talked about one earlier, which is these models are getting very good at bio and they could help us design biological weapons, you know, engineer like another COVID-style pandemic. I worry about that, but because we worry about it, I think we and many other people in the industry are thinking hard about how to mitigate that.
8:14The the unknown unknowns where okay, there's like a there's a societal scale effect from a lot of people talking to the same model at the same time. This is like a silly example, but it's one that struck me recently. LLMs like ours and our language model and others have a kind of certain style to them. You know, they talk in a certain rhythm and they have a little bit unusual diction and maybe they overuse em dashes and whatever.
8:40And I noticed recently that real people have like picked that up. And it was an example for me of like, man, you have enough people talking to the same language model and it actually does cause a change in societal scale behavior. Yes. And you know, did I think that ChatGPT was going to make people use way more em dashes in real life? Certainly not. It's not a big deal, but it's an example of where there can be this unknown unknowns of this is just like this is a brave new world.
9:14Thanks for watching our YouTube channel. We hope you'll subscribe to it. And by the way, you can hit the little bell on there and get notifications every time we produce a video. We hope you'll do that, also.
资源与出处
- 视频在 YouTube 收看
- 档案silicon-valley.tech 档案页本页视频与文稿数据来自 Elon Musk Archive
- 精读本篇精读 Markdown 版