股东会·中文精读

马斯克在 2026 年股东大会的精彩演讲

Elon Musk’s impressive speech at the 2026 shareholders' meeting.

中文精读由 AI 基于原始视频与文稿改写,非官方翻译,可能存在偏差或错误;表述以原视频为准。说明

股东大会的舞台上,几个机器人正在跳舞,身上没有一根电线。台下的股东在鼓掌,有人笑出声。

站在台上的马斯克开口没多久就把话说满了:接下来要做的这个产品,会是人类历史上最大的产品,比手机大得多,总量可能到几百亿台。

底下笑得更欢了。过去一年特斯拉的股价让股东有笑的资本,台上这个人也确实在讲一些别处听不到的东西。整场演讲信息密度很高,Optimus 的产能规划、FSD 的监管困境、AI5 芯片的具体参数、储能业务被低估的逻辑、公司使命的改写,每一样都值得单独拎出来讲。下面按重要性往下捋。

Optimus 的总量,他给到了数百亿台

马斯克判断需求侧的依据很朴素:地球上每个人都会想要一个自己的 R2-D2 或者 C-3PO。他补了一句,Optimus 会比那两个机器人更好用,因为 R2-D2 只会对着人哔哔叫,谁也听不懂它在说什么,还得靠 C-3PO 翻译。

更关键的数字在后面。他估计工业场景和家庭场景的机器人比例大概是三到五比一,每有一个家庭个人机器人,工业里就有三到五个在干活。按这个比例外推,人形机器人的总量会落在数百亿台这个量级。

说到安全,他用了一个很马斯克的比喻。要的是《星球大战》那种未来,别搞成詹姆斯·卡梅隆拍的那种。他特意补了一句自己很喜欢卡梅隆的电影,台下都听懂了,指的是《终结者》。

玩笑背后压着真问题。数百亿台人形机器人在街上走,安全标准怎么定,出事了谁负责,这些他用一句“我们必须非常注意安全”带过。整场演讲里,留白的地方不多,这里算一处。

从 100 万台到 10 亿台,这条曲线没人走过

给 Optimus 排的产能爬坡计划,斜率在人类制造业史上没出现过。

第一步,在弗里蒙特(Fremont)建一条年产 100 万台的产线。这是第一条线。

第二步,在股东大会所在的这家工厂,建一条年产 1000 万台的产线。现场在这一刻响起了掌声。

第三步,年产 1 亿台的产线放哪儿,他自己也没想好。他说也许放火星。台下笑了,他说这话的时候表情没在开玩笑。

再往后,他认为年产量会真的摸到 1 亿台,甚至可能到 10 亿台。他强调,这会是任何大型复杂制成品有史以来最快的产能爬坡。

这句话值得停下来想一想。汽车工业花了一百年才把全球年产量做到 9000 万辆左右,手机行业在巅峰期一年出货十几亿部。给一个包含几万个零件的双足机器人排一条比手机还陡的曲线,工程难度和供应链难度都不在同一个量级。

消除贫困,他给出了唯一解

演讲里有一段容易被当成口号,逻辑链其实是完整的。

人们经常谈论消除贫困、让每个人都享受到最好的医疗。马斯克的说法是,实现这两件事只有一条路,就是人形机器人。

医疗这块讲得最具体。Optimus 的精度最终会超过最好的人类外科医生,达到人类做不到的水平。他用的词是超出人类能力范围。全世界顶尖外科医生只有那么多,培养一个要二十年,机器人可以复制。

然后是那段他自己都承认会被断章取义的话。

有人犯了罪怎么办?他的方案是给这个人配一台免费的 Optimus,二十四小时跟着,阻止他犯罪。除了不能犯罪,其他事随便做。这样就不必把人关进监狱了。

我知道我说的有些话会被剪成片段到处传,但无所谓,我还是要说。(马斯克)

这个想法在法理上显然站不住脚。自由的定义、隐私的边界、谁来判定“即将犯罪”,全是坑。他把它抛出来的方式倒是很有意思:清楚这会被骂,还是说了。先扔一个极端版本,再看公众反应,这是他惯常的做法。

他推荐的书单,藏着对未来的想象

有人问马斯克,心里的未来到底长什么样?他给了书单。

首推伊恩·班克斯(Iain Banks)的《文化》系列(Culture novels)。按他的说法,那是最接近乌托邦的科幻未来。想快速理解他对 AI 和机器人时代的想象,去看班克斯,他的原话是“有点像班克斯式的”(Banksian)。

他还提到了阿西莫夫和海因莱因,重心明显落在班克斯身上。

这个细节值得记一下。《文化》系列描绘的是一个由超级 AI(书里叫“思想”)管理的星际文明,物质极度丰裕,人类不需要为生计工作,可以自由选择生活方式,AI 在背后处理一切复杂决策。马斯克把这本书当作自己未来的参照系,说明他心里的终局既没有“人类被机器统治”的恐惧,也没有“人类和机器开战”的预设,指向的是物质极大丰富、AI 承担管理职能的社会。

至于那个社会里人的意义是什么,他没展开。这个问题比产能规划难回答得多。

经济规模能翻 100 倍,钱可能用瓦特计量

接着讲了一段经济学。

有了 AI 和机器人,全球经济规模可以扩大 10 倍,甚至 100 倍。他的原话是没有明显的上限。

然后他说 Optimus 就像游戏里的“无限金钱漏洞”(infinite money glitch)。玩过游戏的人都懂这个词意味着什么:找到这个漏洞,钱就不再是约束,游戏的其他规则也跟着失效。

再往下就是那句容易被截图的话:未来可能根本没有钱。即使还有钱,衡量的单位可能是瓦数,也就是能调动多少电功率。

用能量做价值尺度,在物理学上比用黄金或美元更硬。卡尔达肖夫指数就是用能量消耗来衡量文明等级的。假如机器人能生产几乎一切商品,稀缺的只剩能源和原材料,货币退化成能源计量单位,逻辑上是通的。

他给股东的建议很短:拿住特斯拉的股票。

使命改写:从可持续能源到可持续富足

这一段是整场演讲的骨架,很多人会忽略。

特斯拉原来的使命是加速可持续能源的到来。马斯克回顾了成果:特斯拉在电动车、电池包、太阳能上带了个头,其他公司跟上了,曾经不存在的电动车现在是主流。他举了一个硬指标,Model Y 是全球所有车型里的销量冠军,不分动力形式。

然后他说,有了 AI 和机器人,使命需要更新。

新使命是实现可持续的富足(sustainable abundance)。

他解释了这个词。他经常问别人一个问题:能想象的最好的未来是什么样?答案通常落在两头,一头是物质和医疗上想要什么有什么,另一头是别破坏自然,雨林还在,国家公园还在,美景还在。

两头同时做到,就是他说的可持续富足。

如果有人能想出更好的未来,我洗耳恭听。(马斯克)

这次改写意味着特斯拉的自我定位从能源转型的推动者,变成了丰裕社会的建设者。前者做减法,减掉碳排放;后者做加法,加上机器人、算力、储能。公司叙事变了,资本市场的估值逻辑也会跟着变。

路上跑着几百万只穿靴子的猫

讲到 FSD,马斯克用了全场最好的一个比喻。

家里有只猫,趴在沙发上。跟别人说这猫其实是《穿靴子的猫》,会穿靴子戴帽子,会耍剑会唱歌会跳舞,别人一定觉得这人脑子有问题,那就是只猫。直到猫真的穿上靴子跳起来,人才会愣住。

特斯拉眼下的处境一模一样。路上跑着几百万辆车,每一辆都设计成可以自动驾驶,但它们看起来就是普通的车。他特意说,除了 Cybertruck,特斯拉的车长得都挺正常。

所以他判断,当下最重要的一件事是教育潜在客户。

有个数据很说明问题:部分客户已经花钱买了 FSD,却从来没打开过。马斯克说他听到这事的第一反应是“什么?至少该试一次”。

现在特斯拉的销售和售后团队会坐下来陪客户,手把手演示怎么用。他的观察是,只要用过几天,就再也回不去了。这种转化率比任何广告投放都值钱。

允许边发短信边开车,这才是真正的杀手级应用

FSD 的版本进展,他讲得很细。

V14 已经到了一个临界点,团队开始觉得可以允许驾驶员发短信。马斯克说这才是人们真正想要的东西,英文里他用了“killer app”这个词。

目前的版本对视线还很严格,摄像头盯着驾驶员的眼睛,看路面看得不够就警告。他说接下来一两个月会仔细看安全统计数据,然后放开这个限制。

理由很实在。现在的真实情况是,很多人为了发短信,先把 FSD 关掉,发完再打开。在两种模式之间来回切换,比开着 FSD 发短信更危险。与其假装禁令有效,不如承认现实。

版本节奏上,14.1 已经有部分人在用,即使在点版本上提升也很明显。真正的重大变化在 14.2 和 14.3。他的判断是,到 14.3 的时候,人可以基本睡着了,醒来就到目的地。

这句话的分量比听起来重。从辅助驾驶到可以睡觉,中间隔的是责任归属的彻底转移。真到了那一步,保险公司、交通法规、驾照制度全要重写。

欧洲像卡夫卡小说,中国 2 月或 3 月

讲到监管,马斯克的语气变了。

欧洲目前连普通的有监督 FSD 都不允许上路。他说自己跟监管机构有过一些像卡夫卡小说一样的对话。

他的说法是:手上几十亿公里的数据,证明 FSD 能提升安全性。对方的回应是:得开一堆委员会会议。他的反问是:人命关天啊。

他特意请求欧洲的特斯拉车主去推动监管机构批准。把监管博弈的一部分压力转给用户群体,这个操作在汽车行业里很少见。

中国的进展更明确。目前是部分批准,他拿到的消息是 2 月或 3 月左右会全面批准。

Cybercab:没有方向盘,5 秒一台

全场唯一一款全新车型是 Cybercab,专门为无人监督的自动驾驶设计。

没有踏板,没有方向盘,没有后视镜。马斯克说这辆车在自动驾驶模式下把每英里成本压到了极致。

产地就在股东大会所在的这家工厂,明年 4 月开始生产。

反常识的地方在制造系统。Cybercab 的产线更接近高产量消费电子的生产线,而非传统汽车产线。

目标是节拍时间(cycle time)压到 10 秒以内,也就是每 10 秒下线一台。几年内可能做到,理论上甚至能到 5 秒。

做个换算就明白意味着什么。一条普通的车用产线,节拍 1 分钟,一年大概能产 50 万辆,Model Y 就是这个量级。节拍压到 5 秒,同样的产线一年能产 500 万辆。

他补了一句,这些车未来会到处都是,外观刻意做成未来感,会改变路面的样子。

特斯拉已经是全球最大的机器人制造商

讲到 Optimus 的制造基础,马斯克抛出一个重新定义:特斯拉造的车本质上就是机器人,只不过多了四个轮子。按这个口径算,特斯拉已经是全球最大的机器人制造商。

他把人形机器人拆成零件清单:电池、功率电子、电机、变速箱、连接模块、基于视觉的 AI、AI 芯片、AI 软件、大规模车队管理能力。这些特斯拉手上全有。

Optimus 就是有胳膊有腿的机器人,底层的技术栈和轮式机器人高度重合。

做机器人真正难的地方,他数出三处。

  • 前臂和手的工程。人手是极精密的器官,灵巧度远超直觉,把一只手做好非常难。
  • 真实世界 AI。实验室里的演示和真实环境里的稳定工作,两码事。
  • 大规模量产。从一台样机到一年一百万台,中间隔着整个制造业。

其他公司通常缺其中一到两样,三样全有的只有特斯拉。他给出的依据就落在这三条上,也是他对那些拿着演示视频的竞争对手的回应。演示机器人不稀缺,稀缺的是能一年造一百万台的那个体系。

两万美元一台,一年一代

成本数字给得很干脆。年产量达到 100 万台并稳定之后,单台生产成本大约 2 万美元,按当前美元计。这个价格放在家用车市场里,属于能大规模铺开的区间。

版本节奏排得清清楚楚。

  • 现场展示的是 Optimus 2.5。
  • 明年开始量产 Optimus 3。他形容这款设计极其优秀,看到实物的感觉像看到一个真人穿着机器人服装站在那儿。
  • Optimus 4 计划 2027 年投产。
  • Optimus 5 排到 2028 年。

一年一代,每代都有明显提升,产能同步放大。这套节奏借的是消费电子的迭代方式,而非工业装备的更新周期。工业装备通常十年一换代,消费电子一年一换代,两者的供应链压力完全不同。

他说完 Optimus 3 那句“看起来像人穿着机器人服装”之后,又补了一句“我们最开始做 Optimus 的时候,里面真就是个人”。惯常的自嘲。

AI5 芯片:三分之一功耗,不到十分之一成本

这段是全场技术含量最高的部分,也是马斯克自称满脑子都是芯片的地方。

AI5 的对标对象是英伟达的 Blackwell。他的说法是:功耗大约是 Blackwell 的三分之一,性能大致相当,成本不到它的 10%。

凭什么做到?他给了两个原因。

第一个是专用化。英伟达要服务的是过去和未来所有客户的需求,所有历史软件都得跑得动,极难兼顾。特斯拉只需要让自己的 AI 软件跑得最好,芯片可以大幅简化。

第二个更技术,也更有意思:特斯拉走整数推理路线。

整数运算在能效和芯片面积上都优于浮点运算,逻辑门层面就简单得多。代价是必须专门为整数推理去训练模型。全行业默认用浮点训练,特斯拉反着来。

马斯克说这是个小众的技术细节,但非常重要。这话说得轻,含金量不低。芯片优势和模型训练方式绑在一起,别人想抄没那么容易,要改的是整个软件栈。

产能布局上,AI5 会在四个地方生产:台积电台湾、三星韩国、台积电亚利桑那、台积电德州。

AI6 已经在规划中。目标是 AI5 投产不到一年,就在同一座晶圆厂切换成 AI6,所有性能指标翻倍。

他对台积电和三星这些合作伙伴评价很高,也提到也许可以跟英特尔谈谈,目前没签任何协议,但值得聊一聊。

然后是这个:即使把所有供应商的最好情况外推,芯片还是不够。所以特斯拉可能得自己建一座 Terafab。

Terafab 这个词照着 Gigafactory 造,Tera 比 Giga 高三个数量级。他说自己看不到别的办法能拿到需要的芯片量。一家车企自建晶圆厂,资本开支的量级和风险,跟建一座整车厂完全不在一个层面。

太阳占了太阳系 99.8% 的质量

讲能源的那段,马斯克用一个物理事实反驳了另一件事。

经常有人说,我们在地球上建核聚变反应堆吧。他的回应是:天上那个巨型聚变反应堆基本不可能被超越。

数据是这样的:太阳占太阳系总质量的 99.8%,木星占 0.1%,剩下所有东西加起来占 0.1%,地球就在那 0.1% 里面。

他接着往下推。就算能把木星扔进一个热核反应堆里烧掉,太阳输出的能量在小数点后四舍五入还是 100%。

所以太阳能成为未来,是算术决定的,跟立场无关。

他还提了一句,未来会有大量太阳能驱动的 AI 卫星,特斯拉会在这件事里扮演角色。太空里没有昼夜,没有云层,日照强度比地面高不少,给算力供电的条件反而更好。

不建新电厂,光靠电池让美国电力翻倍

整场演讲里最被低估的一段在后面。

马斯克给了两个数字。美国的总发电能力大约 1 太瓦,平均用电功率不到 0.5 太瓦。

中间那一半去哪了?昼夜差异和季节差异。白天用电高峰,电厂必须开足;深夜用电低谷,大量机组闲置。没有储能,这部分产能就是白建的。

电池真正的作用在于让已经建好的电厂跑满,存电只是顺带。他的结论很硬:即使一座新电厂都不建,光靠电池,美国的能源输出就能翻倍。

他说这才是美国新增能源供给的主要来源,比大多数人以为的重要得多。这个说法跟大众印象里的“新能源等于风光发电”有落差,落差的部分就是储能的估值空间。

产品层面,Megablock 把公用事业级储能的部署大幅简化,更多部件做进了电池内部,到场卸下来就能用。

下一步的目标更激进:让 Megapack 直接输出 35 kV,连变电站都不需要。他拿 Powerwall 打比方,家用储能一头接电网线,一头接家里主电箱,装完就完事。他想让公用事业储能也变成这个体验。

锂矿、阴极、4680:供应链自己做备份

原材料那段讲的是抗风险能力。

在南德州科珀斯克里斯蒂(Corpus Christi),特斯拉建了中国以外最大的锂精炼厂,起步产能 50 GWh,之后继续扩。理由是,最坏情况下,造一块电池需要的原料必须拿得到。

同一片厂区还有阴极材料工厂,就在半英里外那栋巨大的建筑里。这是为了在地缘政治出问题时,供应链不至于断。

4680 电池也在这家工厂生产,而且越做越好。它已经在 Cybertruck 上用,接下来会用在 Cybercab 和 Optimus 上。

但他明确说了,电池仍然从大量供应商采购。供应商能给多少,特斯拉就要多少;供应商给不出的增量,特斯拉只能自己造,否则就被卡住。芯片是同一个逻辑,所以才有 Terafab 那个念头。

顺便提的几件事

Cybertruck:史上最硬的卡车,字面上防弹,比保时捷 911 快,拖拽能力超过福特 F-350。

Tesla Semi:明年在内华达北部工厂开始量产。百事可乐等公司已经用原型车跑了一段时间。

他点名明年有三个大产品进入量产节奏:Optimus、Tesla Semi、Cybertruck。三件事同时压在同一年,执行风险不小。

超充网络:全球规模最大,已经成了北美的事实标准。用他的话说,其他车厂最后都说那就用特斯拉的超充网吧,特斯拉的回答是行啊,我们没意见。

工厂安全:他讲了一个判断公司好坏的方法,走进工厂或办公室,感受一下氛围。他说特斯拉工厂里的氛围是好的,人是开心的。这个方法很主观,但比看 PPT 靠谱。

产品线:S3XY 全线刷新,加上 Cybertruck。他建议没试过的人至少去试驾一次,重点体验 FSD,他相信试完会被震住。

整场演讲的三段式叙事

把所有信息拼起来,能看到一条完整的三段式故事线。

第一段已经完成:用电动车和储能,把世界从化石能源推向太阳能加电池的经济结构。Model Y 全球销量第一,是这段故事的成绩单。

第二段正在发生:把汽车变成能自己跑的机器人,把 AI 芯片做成垂直整合的核心能力,把 FSD 从一项功能变成一套基础设施。

第三段还没开始:Optimus 大规模量产,劳动力成本趋近于零,物质丰裕成为默认状态,经济规模的约束从人力换成能源。

三段之间层层递进。第一段提供了制造能力和供应链,第二段提供了 AI 和芯片,第三段是前两段合起来的结果。少了任何一段,后面的都推不动,这也是他花那么多篇幅讲产线、讲晶圆厂、讲锂精炼的原因。

至于这些能不能兑现,历史记录是混合的。马斯克说过的量产时间表,兑现的少,推迟的多。但他也确实把电动车、可回收火箭、卫星互联网这些被认定不可能的事做成了。

判断这场演讲的可信度,重点不在他讲得多激动,在于几个能验证的节点:明年 4 月 Cybercab 是否真的投产,Optimus 3 是否真的量产,AI5 是否真的流片。这些都排在未来 18 个月内,藏不住,也拖不起。

接下来盯这几件事

  • 等 Optimus 3 的实机演示视频。马斯克说它看起来像真人穿着机器人服装,这句话能被验证,也能被打脸,先看实物再判断。
  • 关注弗里蒙特 100 万台产线的动工信号。这是“史上最快产能爬坡”说法的第一个物理证据,产线没动,后面的 1000 万台和火星都是空话。
  • 查 FSD 14.3 的接管率数据。他说到 14.3 可以睡着到达目的地,找第三方统计的每千英里接管次数,看曲线有没有真的拐弯。
  • 跟踪欧洲监管的时间表。想观察“用数据说服监管”这条路走不走得通,这里是最好的样本,卡了很久,动一下就知道阻力在哪。
  • 把中国 2 月或 3 月全面批准当作兑现率的压力测试。这是他本人给出的具体时间点,能不能对上,很能说明问题。
  • 留意 AI5 流片消息和 Terafab 的落地进展。特斯拉真开始自建晶圆厂,那是资本开支级别的豪赌,值得单独建一个观察清单。
  • 看 Megapack 直出 35 kV 的工程进度。真做到了,公用事业储能的安装成本会掉一个台阶,变电站这个环节直接被跳过,商业模式跟着变。
  • 读一本伊恩·班克斯的《文化》系列。想理解马斯克脑子里那个未来长什么样,这是最直接的入口,比任何分析文章都准。
YouTube 平台字幕(自动或人工),可能有识别错误,人名与数字请以视频为准。已按语句合并成段落,并去掉 um、uh 等口头填充词。 文稿来源:YouTube(自动/人工字幕)。

0:05[cheering] [applause] [music] Welcome. [music] So, What we're [cheering]

0:38What we're What we're going to [laughter] [music]

1:04And those bots are just dancing. They have no wires. Those are actual robots. [cheering] [applause]

1:18Thanks, guys. first of all, I'd like to just give a heartfelt thanks to everyone who supported the the shareholder votes. I I super appreciate it. Thank you, everyone. [applause]

1:32I'd like to take to thank the the the Tesla board for their immense support. We have a fantastic board, fantastic group of shareholders. Thank you, all. [applause]

1:42And And and what we're about to embark upon is not merely a new chapter of the future of Tesla, but a whole new book. And I'm going to talk about that. [applause]

1:56So, this this really is This really is going be quite the story. and Optimus is a fundamental part of that. The this the sheer scale of Optimus I mean I'm going to say a bunch of things that probably I shouldn't say, you know, but But but that's what keeps it interesting. I mean Have you Have you watched any other annual shareholder meeting?

2:29I mean, honestly, it was like I mean, if you need to go to sleep, sure. I mean that other shareholder meetings are like snooze fests. I mean, but ours ours are bangers. I mean, look at this. This is sick. [applause]

2:46And we're going to like the the cyberpunk nightclub here with real robots just standing there and milling around and dancing. And in around at our engineering headquarters in Palo Alto, the robots are just walking around the office 24/7 with no one minding them. They just And then they go charge themselves.

3:07And Yeah, so the the the scale of of of Optimus, I like I said, that's really going to be something else. I think it's going to be the biggest product of all time by far. Yeah. So Like bigger than cell phones, bigger than anything. I guess the way to think about it is that every human on Earth is going to want to have their own personal R2-D2 C-3PO.

3:43So, who wouldn't? But actually Optimus will be even better than them. You know, like you know, R2-D2 it's kind of would beep at you and it's kind of hard to figure out what he's got talking about. you know, you C-3PO to translate, but Optimus is going to be I I everyone's going to want one.

4:03I think with in terms of industry providing products and services, I think it's probably I don't know, three to five robots in industry for every, you know, one that's a personal robot. I I I think there could be tens of billions of of Optimus robots out there. Now, it's obviously it's very important we pay close attention to safety here cuz we we we do we do want the the the Star Wars movie, not the Jim Cameron movie.

4:36I I love Jim Cameron's movies, but you know, [laughter]

4:42you know what I mean. So, Yeah, yeah, so so we're we're we're going to launch on the the fastest production ramp of of any product of any large complex manufactured product ever. And starting with building a million unit production line in in Fremont. And that's that's that's line one.

5:19and then a 10 million unit per year production line here on the [cheering] [applause]

5:32I don't know where we're going to put the 100 million unit production line. Maybe on Mars, I don't know. but but I think it's it's going to literally get to 100 million a year, maybe even a billion a year. and you know, people often talk about like eliminating poverty, giving everyone amazing medical care.

5:58Well, there's actually only one way to do that, and that's with Op- Optimus robot. With with humanoid robots, you can actually give everyone amazing medical care. so it's in terms of Optimus will be more precise. Optimus will ultimately be better than the best human surgeon with a level of precision that that isn't possible that that that is beyond human.

6:24So I think that's that's a pretty wild concept to say, "Okay, if you you know, there's always people always talk about eliminating poverty but actually Optimus will actually eliminate poverty. Optimus will actually give people incredible medical care. [applause]

6:49So I mean and and so you you start getting like sort of some pretty wild sci-fi sort of scenarios where Yeah, and some of these things I say, "Well, those will be taken out of context and used in snippets and you know, sent around but whatever." I'm still going to say them. You know, like I think we may we might may be able to give people a more if somebody's committed crime, a more humane form of containment of future crime. Which is if if you if you you say like, "You [laughter and clears throat] now get a You get now get a free Optimus and it's just going to follow you around and stop you from doing crime." [laughter]

7:31But other than that, you get to do anything. Just it's just going to stop you from going to commit a crime. That's that's really it. you know, then you don't have to you don't have to put people in like prisons and stuff, I think. it's pretty wild to think of the the various of all the possibilities, but I think it's it's it's clearly it's clearly the future.

7:50and you know, my book recommendation for the maybe the best mostly utopian sci-fi future are the Iain Banks books, the Culture books. So if you're curious like, "What What do I think the future's probably like?" I think it's probably a bit like that. or, you know, Asimov to some degree, but but but I think it's and Heinlein, but Iain Banks. If If you're like saying what what what what does Elon think the future probably probably will be like for AI and robots, it's kind of Banksian.

8:25So, Now now and and and things do get kind of wild from an economic standpoint. because at a certain point with AI and robotics, you can actually increase the global economy by a factor of 10 or maybe 100. There's There's not like an obvious limit. so, like Optimus is kind of like an infinite money glitch.

9:02You know, and I'm maybe they won't even be money in the future. Or or or money but if you might be measured in terms of wattage, like how much you know, power can you bring to bear from an electrical standpoint. So, I guess what I'm saying is hang on to your Tesla stock. Oh, yeah. Yeah. [applause]

9:35So, yeah, man, it's pretty wild. Thank you.

9:41You're welcome. let's see. I think we got some slides to go through. I'm going to ad-lib a lot of this stuff, but [laughter]

9:52So, the the You know, when [snorts] we started Tesla, the the the goal was to accelerate the advent of sustainable energy. And that is that's what we've done. I mean, Tesla has really led the way in with with electric vehicles, with battery packs, with a lot of solar, and other companies have then followed our lead and done that. And electric cars, which used to be non-existent, are now prevalent.

10:22And the the Model Y, for example, is the number one selling car of any kind on Earth. Yeah. [applause]

10:29So. but obviously now with with AI and robotics, we need to update our mission. And our mission, I think it's good it's it's it's it's a great mission, which is to achieve sustainable abundance. Which is it's like cuz I often ask people like what is the future that you want? What's the best future you can imagine?

11:02Yeah, cuz we want to try to make that future. Like make the best future you can imagine. And and I guess probably the best future is if people you know, can have whatever they want from a goods and services standpoint or medical standpoint. And but at the same time, we we don't destroy nature, and we we keep the rainforests and the beautiful national parks and and all the all that stuff.

11:28And so that's that's what I mean by sustainable abundance. is that people, you know, can have whatever they want, have all their needs met, but we still keep all of the natural beauty that we want. Okay. [applause]

11:48I mean, if somebody can think of a better future, I I'm all ears. [laughter]

11:52But I think that's that's probably the best the best way to go. so, yeah. See. So. Yeah. as you know, every Tesla is designed to be autonomous. So, the It's so nice difficult to explain to people if they've not In fact, I'm sure you've all encountered this where you try to tell people that the Tesla can drive itself and they they think you're crazy or something.

12:22I mean especially like apart from the Cybertruck, our cars look pretty normal. I mean they're good-looking cars, but they don't look super They look They look normal. but I guess it it's kind of like having a cat or like and and the cat's just sitting but let's say you've got a cat and it's like just sitting there on the couch and you try to tell people that the cat can actually it's actually Puss in Boots and and it and it can actually put on boots and a hat and and swashbuckle and sing and dance.

12:55And people are like, "No way, man. That's a cat." Until the cat does all those things. You're like, "Damn, what the So we've got, you know, millions of Tesla cars out there that are the They're kind of like Puss in Boots. They're They're intelligent, but people don't know that they're intelligent.

13:19They look like normal cars, but actually they're super smart and can drive themselves. So as I I think that's probably the single biggest thing we need to do is to educate potential customers that you you can either have a car You can either have a cat that's like normal cat or you can have Puss in Boots. And Puss in Boots is very cool.

13:40So that's So we've these days when people come to our stores and or even people that have the car haven't turned it on we find. And sometimes people have paid for FSD and haven't turned it on. We're like, "What? You should at least try it once." And so so now we're like we'll the sales team and service team will actually sit with customers and say, "Look, let let us show you how it works and how easy it is." And then once they've tried it for even just a few days they they can't live without it.

14:15And and and now with the version 14, we're actually getting to the point where we we almost feel comfortable allowing people to text and drive, which is kind of the killer app. Cuz that that that's really what people want to do. And and do do. and actually right now the you know the versions you know, the car's a little strict about keeping your eyes on the road.

14:46and but I'm confident that in the next month or two, we should we we we're going to look closely at the safety statistics, but we will allow you to text and drive essentially. Yeah. So. [applause] Yeah. [laughter]

15:16That's That's It's It's certainly better than the current situation, which often people will actually turn off FSD to text, then turn it back on, which is less safe. So, So, yeah, that's probably the the single biggest thing is is just kept people aware of of FSD. and then obviously we need to get it approved in in Europe. So, we certainly appreciate the support of our customers in Europe pushing the regulators to approve FSD.

15:43Cuz you can't even get super even just normal supervised FSD is not allowed in Europe currently, which doesn't make any sense. And I've had these like crazy conversations with the regulators that seem like a Franz Kafka novel, where where I'm like, "Well, look, we have, you know, billions of kilometers of data that shows that FSD increases safety." and they're like, "Well, we have to have all these committee meetings." I'm like, "Yeah, but people's lives are at stake here." So, definitely pressure from our customers in Europe to to push the regulators to approve would be appreciated. and then we have partial approval in China, and we hopeful

16:21hopefully we'll have full approval in China around February or March or so. That's what That's what they've told us. [applause]

16:36yeah, but Yeah, the the fact The fact that every Tesla car is capable of full self-driving, every car we build and have built for the last several years is capable of full self-driving is is pretty wild, and most people don't don't know that. So, and then we've got the the first car that is specifically built for unsupervised full self-driving to be to be a robotaxi, it's called a cybercab. It doesn't even have pedals or steering wheel. yeah. [applause]

17:15So, no no There's no side view mirrors, there's no Yeah, so it's it's and it's it's very much optimized to for for the lowest cost per mile in a in an autonomous mode. and that production is happening right here in this factory, and we'll be starting production in April next year. [applause]

17:42So, the the way that cybercab is designed is is is is designed obviously for a purely autonomous world, but also the manufacturing system is unlike any other car. it The manufacturing system of the cybercab It's it's sort of It's closer to a high-volume consumer electronics device than it is a a car manufacturing line.

18:11So, the net result is that I think we should be able to achieve I think ultimately less than a a 10-second cycle time, basically unit every 10 seconds. maybe ultimately take a few years to get there, but it's theoretically possible to get to a 5-second production time. And so so what that would mean is you could get on a on a line that would normally produce, say 500,000 cars a year at a 1-minute cycle time, like Model Y, the this this would be maybe

18:57as much as you know, two or three million, maybe ultimately you know, it's theoretically possible to achieve a 5 million unit production line if you if you can get to the 5-second cycle time. So, [applause]

19:19it's a lot of cars. So, the these will be everywhere in the future. And we wanted to we wanted it to look futuristic, so like it changes the look of the roads. Now, the yeah. The ingredients, when you look at at at what Optimus is what's required to make Optimus and the various ingredients, what do you need to do to make to do high volume humanoid robot production, I think it's worth considering that really the cars we make are already robots, but they're four-wheeled robots.

20:09So, Tesla is already the biggest robot manufacturer in the world because every car we make is a robot. and when you break it down to the fundamental elements, you've got you've got batteries, power electronics, motors, gearboxes, you've got you know, connectivity, you've got a vision-based AI. Hi, Optimus.

20:35and you know, all the various pieces that you need for a humanoid robot. You need the AI chip, you need the AI software. You need to be able to manage a large fleet. And and so really the Optimus is a robot with arms and legs as opposed to as opposed to a robot with wheels. So, you know, Tesla is ideally suited, I think, to to make to succeed in this arena.

21:09You you you will see certainly many companies showing demonstration robots. But there's really three things that are super difficult about robots. One is the engineering of the forearm and hand because the human hand is an an incredible incredible thing, actually. It's super dextrous.

21:30So, engineering the hand really well, the real-world AI, and then volume manufacturing. Those are generally the things that are missing one or more of those things are missing from other companies. so, Tesla is the only one that has all three of those. yeah. So, yeah. [cheering] [applause]

21:59So, this is the Optimus kind of initial it's a the kind of a prototype production line. The the high-volume production line will be very automated, obviously, but this is this is really the production line that we use to make the prototypes. So, you can get a sort of rough sense for what it takes to build the robot.

22:37You ever pull the finger? and then as I've said before, I think once we reach about a million units per year of sustained production or in excess of that, I think probably the cost of production is around $20,000 in current year current year dollars. So, this will be certainly very affordable.

23:15and yeah, like I said, I think Optimus will ultimately increase the size of the economy probably by a factor of 10 or more. you know, next year we start production with Optimus version 3. this what you'll see here is Optimus version 2.5. Optimus 3 is is an incredibly good good design.

23:42The Tesla engineering team is amazing. when you see Optimus 3, yeah. [applause]

23:56you it it will seem as though that there's someone like a like a person in a robot outfit. which is how we started with Optimus. it really is going to be something special. and then Optimus Optimus 4, you know, that that hopefully starts production in '27. And then Optimus 5 in '28, so it's kind of like an annual release cycle with significant improvements.

24:25with each one and and gigantic increases in the scale of production. So, yeah, sustainable abundance via AI and robotics. That's the future we're headed for. [cheering] [applause]

24:50And as you as you I think most people here know the safety statistics show that miles driven on FSD are much safer than miles driven with without it. So, what this will translate to ultimately is saving the lives of millions of people and preventing hundreds of millions of accidents.

25:11So, a massive increase in the in you know, lives saved and and tragedies avoided. it's going to be amazing. [applause]

25:30And how many people here have tried 14.1? Okay. All right, cool. yeah, you can see the even with the point releases it's getting quite a bit better. it should be pretty smooth at this point. but really 14.2 is the major changes to 14.2 and then 14.3. and I think I think by 14.3 is when we're we're really be at the point where you can just pretty much fall asleep and wake up at your destination. So, [applause]

26:12and then I've been putting a lot of time into the new Tesla chip design because in order to have a functional robot, you have to have a great AI chip. and it needs to be an inexpensive chip and it needs to be very power efficient. So, we we we believe the AI 5 chip will be probably about a third of the power of say something like a Blackwell. an Nvidia Blackwell, which is a great a great chip. for roughly a comparable performance.

26:46and about and and much less than 10% of the cost. So [applause]

26:58This is a chip that is that is very much optimized for the Tesla AI software stack. So it's it's not meant to be a general purpose chip. It's meant to be an amazing chip for the Tesla AI software. and I mean a couple things that I think make like how is Tesla able to achieve such an improvement? I think it is because because we are specialized. We're not trying to You know, Nvidia has to serve the super set of all past and future customers. So all of their requirements, all of the software that they've written has to work, which is a very difficult problem.

27:35whereas we just need to make it work for our software. and and so we we're able to simplify the chip dramatically. and then we we also I think I think we're unique in this, but like we we have an integer based system. and integer operations are fundamentally more efficient than floating point operations.

28:00So we can do floating point, but we but the vast majority of of our inference is done in integer. which is if you're familiar with sort of logic gates, like the simplicity of of integer and it's integer is much more power efficient, much more silicon efficient. but you have to you actually have to train for integer inference, which everyone else is training for floating point.

28:29those that's kind of like a niche technical detail, but it's actually very important. So, Yeah, this this is this is going to be a great chip. So, this this chip will be made in basically in in four places. TSMC Taiwan, Samsung Korea, TSMC Arizona, and TSMC Texas. [applause] [clears throat]

29:02And and we already know what improvements to make for AI 6. So, I I I'm hopeful that we can within less than a year of AI 5 starting production, we can actually transition in the same fab to AI 6 and double all of the performance metrics. So, [applause]

29:26I I'm super hardcore on chips right now as you may be able to tell. I have chips on the brain. I dream about chips. Literally. I can I can I can draw the the the the you know, the the at least the broad brush stroke physical design of the AI 5 chip by off by heart at this point.

29:52It's it's it's a good chip. It's a good chip, sir. So, this this is this is really key. Now, one of the things I'm trying to figure out is how do we make enough chips? So, You know, I I have a lot of respect for our the the the Tesla partners, TSMC and Samsung. Yeah, you know, maybe we'll we'll we'll we'll do something with Intel. we haven't we haven't signed any deal, but it's probably worth having discussions with Intel.

30:31but even when we extrapolate the best case scenario for chip production from our suppliers, it's still not enough. So, I think we may have to do a Tesla Terafab. Oh. [applause]

30:55It's like Giga, but way bigger. I can't see any other way to get to the volume of chips that that we're looking for. So, I think we're probably going to have to build a gigantic chip fab. Yeah. [applause] [applause]

31:23Got to be done. So, anyway, some of the stuff I've already talked about. Yep. We've we've done a tremendous amount for sustainable energy, and that that's only going to grow over time. The world is moving towards a a solar battery economy, which is ultimately the that's it's where that's where it was going to go anyway, but what Tesla has done is accelerate accelerate that outcome.

31:58Sometimes people don't understand quite how much energy comes from the sun. So, the sun has 99.8% of all mass in the solar system, Jupiter being 0.1% and then 0.1% is miscellaneous, Earth being in the miscellaneous category. So, the the total amount of energy like so sometimes people what say like, "Well, we'll we'll build the fusion reactors on Earth." It's like, "Well, actually the giant fusion reactor in the sky is basically impossible to beat." to such a degree that even if you could burn Jupiter in a thermonuclear reactor, the amount of energy produced by the sun would still round up to 100%.

32:37That's how much energy the sun produces. So solar power is necessarily the future. and I think there's going to be a lot of solar-powered AI satellites. And I think Tesla's going to play a role in that. [applause]

33:06we've obviously refreshed the product line. So S3XY, if if you will haven't tried the Model S 3 X or Y or the Cybertruck, I recommend at least getting a test drive or a test ride, as the case may be. Try out the the full self-driving. and I think you'll be blown away. So those who do not If you may be listening and don't have a Tesla, you should try one.

33:31So [applause] And of course we got the Cybertruck, which is the toughest truck of all time. it's literally bulletproof, faster than a Porsche 911, and can out-tow a a Ford F-350. So it's a it's a great car, great truck. [applause]

33:57and then starting next year we we manufacture the Tesla Semi. So This this is already we're we're already have a lot of prototype Tesla Semis in operation. PepsiCo and other companies have been using the Tesla Semi for quite some time. but we will start volume production at our Northern Nevada factory in 2026. So we've got we've got two big products or three three massive products starting production next year. We've got Optimus, we've got Tesla Semi, and we've got the Cybertruck. [applause]

34:49And then battery packs. So the if you if you look at total US power generation capability, it's roughly a terawatt. but the average power usage is less than half a a terawatt. And that's because there there big differences in power usage between day and night. So the the the the daily and seasonal variations in power consumption mean that the United States and and really every country is only using about half is only producing about half as much electrical energy as it could.

35:27Because without batteries, there's no effective way to buffer the energy. so what batteries actually enable is even if you don't build any incremental power plants, you could double the energy output of the United States just with batteries. Now this is a super big deal. and in fact, I think that's really where most of the incremental energy production in the United States is going to come from is literally batteries.

35:54So a bigger deal than it may seem. [applause]

36:04And then we've got we we keep improving the battery design. So the the Megablock which makes it really easy to deploy battery utility-scale batteries. so we've we've just simplified and and brought more of the components to be internal to the batteries, so you can just show up and drop off a battery and it works.

36:26and then hopefully with I don't know how to hopefully over time we will actually add more and more of the power electronics so that Megapack will actually be able able to output up to 35 kV directly. so you you won't need a substation, that's what I'm saying. You can just literally drop it off, kind of like the way that a power wall, you just connect it to the house, the utility wires go in one side, and the other side goes to the house mains, and that's it.

36:55so, we want to get Megapack to the point where you just literally take the the utility wires, and you plug them in, and it just works. So. [applause]

37:11And we've got the We've also built the world's largest Supercharger network. So, we do a lot of things here at Tesla. that's the biggest Supercharging network in the world by far. and ultimately, you'll be able to go anywhere on Earth using a Tesla Supercharger. And it's on a pretty close to anywhere on Earth, but it's you know, it's going to be ultimately just anywhere. It'll just work anywhere.

37:38So, the the Supercharger team has done done great work expanding that and improving the efficiency of the Supercharger network. and in in North America, they did such a good job that the the other car companies basically said, "Well, they'll just use the Tesla Supercharger network." Like, like, "Okay, sounds good to us." [applause]

38:06you know, it's always important to have a a slide on safety in the factory, so we continue to improve safety for our factory workers. We're We care a lot about their well-being. and you know, like one way you can just tell like if a company's a good company or not, if you if you just walk through the factory or walk through the office, and catch the vibe.

38:27And the the vibe in the Tesla factory is is is good. People are happy. You know, that's how you know it's a good company. [applause]

38:40we've also put a lot of investment into raw materials. So, the We've built in in South Texas in Corpus Christi the biggest lithium refinery outside of China, I believe. so, it's it's going to it's starting off with a with 50 gigawatt hours of of of lithium and we'll expand from there. So, this is a very very important to have you know in a worst-case scenario that we have the ingredients necessary to make a battery.

39:18Very important. and and then we've got here on this site the the cathode factory, which is just the sort of giant building about a half mile that way. and and we're just making sure that we from a supply chain standpoint are resilient against any potential geopolitical challenges. [applause]

39:47And then also at this factory, we also make the 4680 cell, which is getting better and better. and that a 4680 cell will be used in the Cybertruck Cyber It is being used in the Cybertruck and will be used in the in the Cybercab and also in Optimus. So, that that's going well. but we continue obviously to get cells from many suppliers. It's kind of like the the chip bad thing. We'll We'll take as many chips as our suppliers can provide us, but then beyond that, if they can't provide us with any incremental cells or chips, we we kind of have to make them ourselves or we get stuck.

40:28That's the Tesla Semi factory. So, it's going to it's going to be pretty cool to see these going down the road in scale. Yeah, so So that's the that's the basic plan. Sustainable abundance for all. [applause] [applause]

资源与出处

相关视频

同频道、同主题或相邻年份