2023 年第一季度,欧洲最畅销的车型是 Model Y。不分燃油还是电动,所有车里排第一。美国市场卖得最好的非皮卡车型,也是它。马斯克在 2023 年股东大会上说,Model Y 大概率会成为今年全球销量第一的车型。
一台纯电动车拿下全球销冠,这在五年前还是不可能的事。
同一场大会上,马斯克还说了另外几件事:未来 12 个月全球经济会很难,会有不少公司破产;特斯拉的长期价值,大部分将来自 Optimus 机器人;FSD 一旦跑通,几百万辆车的价值会在一夜之间翻五倍。
这几句话放在一起,基本就是这场股东大会的全部信息密度。下面按顺序拆开讲。
得州工厂一年产量翻了三倍
董事会主席 Robyn Denholm 开场讲了一组对比,很能说明问题。
2021 年在同一个地方开股东大会,得州工厂还在安装设备、调试产线。2022 年开会时,周产量大约是 1600 辆。到 2023 年 5 月,周产量已经超过 5000 辆。一年时间,翻了三倍多。
2022 年全年交付量同比增长 40%,2023 年第一季度延续了同样的趋势。
减排数据也在同步放大。2021 年,特斯拉车主累计减少了 840 万吨温室气体排放,相当于燃油车行驶 200 亿英里的排放量。2022 年这个数字变成 1340 万吨,相当于 330 亿英里,同比增长接近 60%。
2022 年全球超级充电网络 100% 使用可再生能源。这个数字很多人不知道,因为它不像销量那样每个月被拿出来说。
还有一个人事变动值得单独提一句:JB Straubel 加入董事会。他是特斯拉早期的 CTO,电池和动力系统的技术底子很大程度上是他打下的,离开后创办了电池回收公司 Redwood Materials。他回来进董事会,对特斯拉的电池战略是个明确信号。
第三篇章讲了三件和直觉相反的事
马斯克花了不少篇幅回顾 Master Plan Part 3(秘密宏图第三篇章)。他说这份规划写得很技术,故意写得技术,为的是让里面的结论能被物理和数据验证,而不只是口号。
核心结论有三个,每一个都和大众直觉相反。
第一,转向可持续能源经济需要消耗的总能量更少,不是更多。很多人本能地觉得,搞风电光伏电池这些新东西,肯定要额外消耗大量能源。可算总账,从开采、运输、燃烧化石燃料的整条链路,切换到太阳能加电池的链路,能源效率是提升的。
第二,需要的采矿量也更少。这一点更反直觉。大家觉得电动车电池要用锂、镍,矿肯定挖得更多。可化石燃料是持续消耗的,每年都要挖、都要运;电池材料是一次投入、长期循环。按全生命周期算,可持续能源经济的矿物开采总量低于现在的化石能源体系。
第三,也是最关键的:锂离子电池产能的增速,直接决定全球能源转型的速度。
马斯克的原话是,电池是所有交通工具和储能的基础,而风电光伏是间歇性的,必须配电池。所以电池产能这条曲线有多陡,能源转型就有多快。这是整个系统的根本限制因素。
具体数字:全球需要约 240 TWh 的电池产能,也就是 24 万 GWh。太阳能和风电装机要达到现在的三倍,电池产量要增长约 29 倍。听着吓人,可太阳能和风电占地只需要全球陆地面积的 0.2%,不用把地球铺满。总投资约 10 万亿美元,大致相当于全球 GDP 的 10%。
马斯克的评价很直接:用全球经济的 10% 换一个可持续能源的未来,这笔账很划算。
他还主动邀请外界来挑错。任何一份分析都会有偏差,所以第三篇章公开出来就是为了收集批评意见。到目前为止,收到的反馈大多认为这份分析相当准确,也没遇到能推翻核心假设的反驳。
钴矿问题:大部分钴其实在你的手机里
会上有个股东提案,要求特斯拉出具第三方报告,说明如何消除供应链里的童工和强迫劳动。提案人 Karen Roberts Doutre 代表 SumOfUs 发言,讲得挺客气,先肯定了马斯克把整个行业拖着往前走,然后说了一句很重的话:特斯拉这家公司,在公众眼里已经成了 CEO 的同义词,讨论全跑偏了。
她还提到一个数据:特斯拉的品牌好感度去年掉了 15 个点,这件事直接反映在了毛利上。
另一个提案来自 Investor Advocates for Social Justice 的 Courtney Wix,聚焦刚果(金)的钴矿童工和新疆地区的强迫劳动风险。她引用的数据是:全球 70% 的电动车电池用钴来自刚果(金),超过四万名儿童在手工采矿,很多人在没有防护设备的情况下作业,矿坑塌方经常造成伤亡。
马斯克的回应分两层。
第一层是态度:特斯拉会做第三方审计。他甚至还加了一句玩笑,说要在矿上装个摄像头,谁看见小孩就报告。台下笑成一片,可承诺是实打实的。
第二层是事实纠正,这一层更有信息量。
特斯拉大部分电池包是铁基的,也就是磷酸铁锂,根本不用钴。剩下的镍基电池里,钴只作为粘合剂用一点点。真正大量用钴的是消费电子。马斯克的原话大意是,你们的手机是 100% 的钴,建议去跟手机厂商投诉。
这个区分很重要。舆论场上关于电动车钴矿的批评,很多时候把电池和消费电子混为一谈了。数据上,一台电动车电池包的钴用量,和一部手机比,比例完全不在一个量级。
他也表了态:哪怕只用一点点钴,特斯拉也会用各种办法确保没有童工参与,六重核查都要做足。
全球最大电动车厂,也是全球利润率最高的车厂
这两个头衔放在一起才是重点。
马斯克说得很直白:造电动车还能赚钱,这件事非常难。很多竞争对手每卖一台电动车就亏一笔钱。特斯拉不亏,而且是行业里几乎唯一不亏的。
他重复了那句老话,原型车容易,量产难。量产还能保持正现金流,是“极度痛苦”的事,痛苦程度外人很难想象。这句话他在很多场合说过,可在股东大会上说,面对的是一群真正关心财报的人,分量不一样。
自由现金流逐年上升,资产负债表在改善。2022 年特斯拉拿到了投资级评级,这是从 2018 年“产能地狱”里爬出来后最重要的一次身份转变。
还有一组容易被忽略的数据:特斯拉目前直接雇佣的员工数量已经相当可观,还没算承包商。而每创造一个制造业岗位,沿着供应链和配套服务算下来,能带出至少五个、有时多达十个就业机会。工厂落地带动的是教师、律师、木匠、电工、餐馆这一整条生态。这也是各州各国争抢超级工厂的真正原因。
特斯拉去年收到 360 万份求职申请。工程类人才最想去的两家公司,是 SpaceX 和特斯拉,排在全球前两位。
马斯克对这件事的解读很干脆:一家公司的竞争力,取决于最有才华的人愿意去哪里上班。哪家公司能吸引到最聪明、最有驱动力的人,哪家公司就会赢。这话适用范围不止特斯拉。
下一代动力单元:不用稀土,工厂面积减半
这部分是整场大会技术含量最高的段落之一。
新的下一代驱动单元有三个关键指标:
- 碳化硅用量大幅减少
- 工厂占用空间减半
- 完全不使用稀土元素
稀土这条最值得说。永磁电机长期依赖钕、镝这类稀土材料,开采和提炼高度集中在中国,地缘风险一直被行业讨论。特斯拉下一代驱动单元直接绕开了这个依赖,对整个供应链的议价能力和抗风险能力都是实质提升。
另一个变化是整车低压架构,从 12V 切到 48V。
这件事听起来小,其实是汽车工业一百年来没动过的基础设定。从福特 T 型车开始,车上的低压电器一直跑在 12V。切到 48V 之后,同样功率下电流只需要四分之一,整车用铜量也降到大约四分之一。
铜的供应一直是电动车扩产被质疑的点之一。48V 架构直接把这个担忧的量级压下去了。马斯克说,很多人问铜够不够用,答案是够。
FSD:一次软件更新,让几百万辆车的价值翻五倍
这是整场大会马斯克讲得最兴奋的一段,也是最需要拆开算账的一段。
先说进展。FSD Beta 的测试规模在指数级增长,那条曲线在图表上几乎是一堵墙。马斯克现场问台下有多少人用过 FSD Beta,举手的人一大片。他自己在奥斯汀随机打点开了好几天,零干预。旧金山那种复杂路况,也很长时间没有安全干预了。
技术上还差最后一块:规划和控制模块要变成神经网络。现在感知是神经网络,规控还是传统代码。这块换掉之后,整个系统就是端到端,视频输入,控制输出,中间全是神经网络。这是自动驾驶架构的最后一公里。
然后是那段最关键的算账。
一台私家车,正常使用强度大概是每周 10 到 12 小时,一天一个半小时左右。一周有 168 小时,也就是说,一辆车超过 90% 的时间是停在停车场里的。
一旦实现自动驾驶,同一辆车每周可以跑 50 小时甚至更多。车还是那台车,成本没变,使用率翻了五倍。
马斯克的说法是,这将是地球历史上最大的一次资产价值跃升。几百万辆已经在路上的车,不需要换硬件,一次软件更新,从手动驾驶变成自动驾驶。
他还补了一句:让我惊讶的是,这么明显的事,居然没多少人意识到,或者意识到了但不相信。
这个逻辑对特斯拉的商业模式影响很深远。它意味着卖车只是第一步,真正的价值在软件激活之后释放。也意味着特斯拉的估值逻辑不能按传统车企的市盈率去套。
未来 12 个月:别指望好消息
整场大会里,这一段最不像“报喜”。
马斯克说得很清楚:未来至少 12 个月,宏观经济层面对所有人都不好过。特斯拉不可能独善其身。
逻辑链条是这样的。绝大多数人买车看的是月供,不看总价。利率一涨,月供就涨,购买力直接被压缩。同时信贷在收紧,而今年几家银行接连出事,用马斯克的话说,一家银行如果自己都在担心能不能活下去,扩大车贷组合根本不是它眼下最关心的事。他把那种状态形容得很好笑:一队人正往墓地走,你让他顺路多放几笔车贷,这显然不在他的日程上。
所以他建议:未来 12 个月别盯着股价看。如果跌了,那就是买入的机会。他的判断是,全球经济大约一年后回暖,而特斯拉届时会处在非常有利的位置。
他还举了个佐证。伯克希尔哈撒韦开股东大会时,巴菲特和芒格都说今年旗下公司会少赚钱。那是一家管理极其优秀的机构,它对宏观的判断有参考价值。
这段话放在一场股东大会上说,其实需要一点勇气。可他讲了,而且讲得很具体,包括点名说会有不少公司在这一轮里破产。
他也没把话说死。黑暗不会永久,好日子会跟着难日子来。这是经济周期,特斯拉会熬过去,而且会以更强的姿态出来。
储能业务的增速,已经超过汽车业务
这一段容易被忽略,可数据很硬。
Megapack 现在的成本已经比天然气调峰电厂更有竞争力。调峰电厂是电网里专门应付用电高峰的那类机组,效率低、成本高,长期是电网最贵的一块。Megapack 能在这个场景下打赢,意味着储能不再只是环保选项,同时是纯经济账上的优选项。
马斯克几年前预测过,特斯拉的储能业务会增长得比汽车业务还快。现在应验了。那条增长曲线的斜率,已经超过车辆销售。
产能布局上,加州和上海的 Megafactory 各自瞄准每年 40 GWh。马斯克说,长期看这个数字可能会超过每年 500 GWh,因为需求几乎是无限的。电网要稳定,AI 数据中心要供电,新能源要削峰填谷,每一块都在拉动储能。
现场有股东问到加州的轮流停电和户用储能。马斯克的回答里有个具体障碍:Powerwall 要当作家里的不间断电源,得装一个切换开关,把房子从电网断开。否则停电时电能照旧往电网里送,家里还是黑的。这个开关的审批在各州推进得很慢。他专门感谢了 PG&E,说最近帮特斯拉拿到了加州装这个开关的许可。团队正在跟每一个辖区谈。
锂矿不缺,缺的是精炼产能
现场有股东问,未来十年电池级锂的供应够不够。他的担心是,锂矿从勘探到投产动辄十年以上,通用汽车这类对手已经在抢长协锁供应。
马斯克的回答直接推翻了前提。
锂不稀缺。锂是地球上最常见的元素之一,几乎每个国家都有。它不像石油那样集中在少数地区。
真正的瓶颈是把矿石提纯到电池级。电池对纯度的要求极高,一点杂质就会显著缩短电芯寿命。所以限制因素落在精炼产能上,挖矿速度并非关键。矿再多,没有对应的精炼厂消化,挖出来也没用。矿石的开采节奏会自动跟着精炼厂的处理能力走。
特斯拉在得州 Corpus Christi 破土动工的锂精炼厂,主要处理锂辉石,原料大头来自澳大利亚,约占四分之三。马斯克说,这个厂的锂精炼产能,大概会超过中国以外所有精炼厂的总和。
他在财报电话会上喊过好几次“谁能来做锂精炼”,没人接。最后只能自己上。特斯拉也不想多揽事,可没人干,不干就卡住。
Cybertruck 和两款还没露面的新车
Cybertruck 是全场被欢呼声打断最多的部分。
马斯克承认这车难造,难在它用了外骨骼结构,并非传统车身结构。普通汽车是内骨骼,底盘承受结构强度,外面覆盖车身面板。Cybertruck 是不锈钢车身本身充当结构件。这意味着传统的冲压、焊装、涂装整套流程都用不上,制造工艺得从头发明。
进度是今年晚些时候开始交付量产版。马斯克自己说,他日常会开这台车。产能爬坡按 S 曲线走,先慢后快。年产量他给了一个大概区间,25 万辆左右,如果需求旺盛可能到 50 万辆。他同时提醒,这是一台全新车型配全新制造工艺,前期成本下降会很慢,相对特斯拉整体产量盘子,占比不会大。
他还提了一个容易被忽略的开放思路:Cybertruck 会预留大量外挂接口,让第三方公司自己去做露营套件、房车改装之类的附件。特斯拉不打算自己做房车,可别人可以做。
更有想象空间的是他后面那句话:还有两款新产品在设计中,设计和制造工艺都远超行业现有水平。他给了一个非常马斯克式的预测,这两款车合计年销量可能超过 500 万辆,然后补了一句“别告我”。
他没透露是什么车,只说需要专门的发布会。但他强调了一点,这两款产品是真的在设计、真的在推进,团队没有闲着。
Optimus:马斯克说,特斯拉未来的大部分价值在这里
这是全场信息量最大、也最容易被当成玩笑的一段。
放出来的视频是前一天晚上拍的,Optimus 团队通宵赶工。从上次亮相到现在,进步幅度看得见。走路稳了,动作复杂了。
关键信息在硬件。Optimus 身上你能看到的电机、控制器、电子系统,全是特斯拉自己设计的。团队一开始想找现成的执行器,找了一圈发现市面上根本没有。人形机器人对执行器的要求,和工业机器人、和汽车,都不一样。所以只能从零设计电机、减速器、电控。
这件事能做成,靠的是特斯拉做电动车攒下来的电机和电力电子团队。这是十几年积累的横向复用。
软件层面,Optimus 用和车一样的 FSD 计算机。FSD 在解决的是通用真实世界 AI,也就是在物理世界里理解、判断、行动。这套能力一旦成立,从开车迁移到走路、抓取、搬运,是同一条技术路径。人类的同一个大脑,既能开车也能走路、骑车、划船,逻辑是一样的。
然后是那个数字。
马斯克说,如果人形机器人真的成立,需求量会远超汽车。他给了一个粗略推算:如果机器人和人的比例是二比一,那就是 100 亿到 200 亿台的量级。
接着他说了一句让全场安静一秒的话:我的预测是,特斯拉的长期价值,大部分将来自 Optimus。我对这个预测非常有信心。
他也提了风险。要确保不出现终结者那种场景,要保证机器人可以被本地关闭,安全问题是重中之重。他自己都说,在终结者出现之前,一切都只是好玩。
现阶段 Optimus 的智能水平还很低,连煎鸡蛋都做不到,更不会思考火星这种问题。先要解决的是基础动作的敏捷性,然后去做那些枯燥、重复、危险的工作,也就是人不想干的活。
现场问答:广告、个人状态和定价策略
问答环节有几个回答值得单拎出来。
关于广告。一位股东当场算账,说特斯拉每台车让利 525 美元,一年下来就等于 Netflix 的全部广告预算,而 Netflix 的广告铺天盖地。马斯克一开始还在解释自己的推特账号已经有大量关注,属于“对唱诗班布道”。说到最后他松口了:我们会试试看投放一点广告。这句话引爆了全场,欢呼声大到他自己都说没预料到。对一个十几年坚持零广告预算的创始人来说,这是个不小的转向。他后来还补了一句自嘲,说推特高度依赖广告,现在自己也得说一句广告很棒,大家都该投。
关于他本人的状态。有人直接问“你还好吗”。马斯克承认这段时间像过山车,工作上的痛苦程度很难受,还要面对媒体的持续攻击。他提到收购推特相当于给这家公司做了一次开胸手术,为的是保住这家公司不死。现在推特已经进入稳定期,Linda Yaccarino 会接手日常运营,他自己未来在推特上投入的时间会大幅减少,主要精力放在技术和产品指导上。他说过去半年推特推出的新功能,比过去六年加起来还多。
关于车贷和金融方案。有人问能不能把车贷期限拉到 10 年。马斯克解释,特斯拉不直接放贷,绝大部分融资来自银行。今年销售额大概 1000 亿美元级别,特斯拉自己拿不出这么多现金来垫。银行现在的处境决定了放贷意愿,这件事特斯拉说了不算。所谓自营金融公司也不是万能药,贷款要证券化,而证券化市场现在很弱。他坦言希望能有一条简单路径,可惜没有。
关于定价。有股东问 Model Y 年初降价、后来又涨价,策略是什么。马斯克的回答很朴素:看需求,需求变了就调价。他补了一个常被忽略的事实,传统车企其实一直在调价,只是通过经销商完成,终端价格不透明。去年普遍加价提车,今年普遍低于指导价,只是没人把这些当成“官降”。特斯拉直营,价格变动立刻可见,所以显得动作大。需求波动本来就大,总得有机制把它和供给对上,这是所有车企都在做的动作。
关于 Roadster。第二代 Roadster 今年完成工程和设计,明年有望开始生产,不打包票。会带 SpaceX 选装包。马斯克自己说,这车对营收贡献不大,对利润贡献有限,但它会很酷。他形容这车是蛋糕上的樱桃,属于锦上添花。
关于 4680 电池。有人问现在的 4680 离电池日公布的指标还有多远。马斯克承认这条路极难,一家从零开始做电芯的公司,要做得比全球所有只做电芯的公司还好,本身就是不寻常的事。但他认为路径是清楚的,方向是更高能量密度、更低成本。
关于会不会卸任 CEO。有人问传闻中的交班。马斯克回答很干脆:不会。理由是特斯拉在 AI 和 AGI 里会扮演重要角色,他要盯着这件事往好的方向走。他还说,外界严重低估了特斯拉在 AI 上的能力,特斯拉的真实世界 AI 是当下最先进的,没有接近的对手。现实世界的自由度最高,这件事必须做好。
关于那个“老婆测试”。最后一位提问者说,他太太只愿意让孩子坐 FSD 开的车。马斯克说这个测试非常好。前几年身边人上车第一句话是“快关掉,太吓人了”,现在已经反过来了。他还给了一个观察方法:高速上如果一台车稳稳走在车道正中间,那大概率是 FSD 在开;如果一直在车道里小幅左右修正,那是人在开。
关于车机娱乐和安全。有人问会不会开放第三方给车机做主题和应用。马斯克说,等车能自己开了,娱乐和生产力的重要性会大幅上升,那时候这件事才有意义。眼下软件团队的重心还是安全和 FSD。他也顺带提了个很多人不知道的功能,Model S 和 X 已经做了完整的 Steam 集成,登录账号就能玩库里所有游戏,键鼠和手柄都支持。另外有人问到网络安全,他说特斯拉常年办第三方黑客大赛,邀请全球最好的白帽子来攻,目前还没有人能真正控制车辆,最多只能按响喇叭或者干扰信息娱乐系统。
为什么马斯克坚持说 FSD 的分量等同于电气化
大会快结束时,马斯克讲了一段话,值得单独拿出来。
他说 FSD 的研发过程并非一条平滑上升的线,而是一连串对数曲线。每次看着快成了,其实撞上了一个局部最优。要跳出去,只能重构软件架构,然后爬上下一个更高的局部最优。这就是为什么过去几年每次预测都显得过于乐观,因为线性外推在对数曲线上不成立。
他这次的判断是,今年会爬到一个比人类驾驶更安全的局部最优。目前有驾驶员监督的 FSD,安全性已经是手动驾驶的好几倍,他自己给的说法是四倍左右甚至更高。
然后他讲了那个更大的判断。全球每年约 120 万人死于交通事故,约 1000 万人严重受伤。如果自动驾驶能把事故率降一个数量级,每年就是 90 万条生命、900 万起重伤的区别。他说从道德上讲,这必须是首要目标。
这也是他反复强调 FSD 分量重的原因。他的类比是,这件事的深刻程度等同于电气化,属于文明级别的基础设施变化,用“功能”两个字去理解它远远不够。而特斯拉的特殊之处在于,几百万辆已经在路上的车,靠一次软件更新就能完成切换。这在汽车工业史上没有先例。
现场有人希望特斯拉给出一个对外的时间表,什么时候到哪个版本,什么时候能落地。马斯克的回答很坦诚:给不了。这条路之所以难,就在于你永远不知道自己是不是又撞上了一个局部最优。他过去给过很多时间点,大多没兑现,所以这回他宁愿只说方向,不说日期。
这场股东大会到底说了什么
把两个多小时压缩一下,其实是三条线。
第一条是硬的:制造和成本。得州工厂一年三倍产能,Model Y 拿下全球销冠,下一代驱动单元去稀土,48V 架构省铜,锂精炼自己建,Megapack 打赢天然气调峰。这些是能算账的东西,也是特斯拉当下利润率的来源。
第二条是软的:FSD 和 AI。这条路走了很多年,撞了很多次局部最优,现在到了最后一个模块要换成神经网络的节点。如果成了,几百万辆车的使用率翻五倍,整个估值逻辑要重写。
第三条是远的:Optimus。马斯克把特斯拉长期价值的大部分押在这里,理由是 FSD 积累的通用真实世界 AI 可以迁移到人形机器人,而人形机器人的潜在需求量级远超汽车。
他还给了一个不太好听但很实在的提醒:未来 12 个月宏观环境会很难,会有公司倒下,特斯拉也躲不开。可他同时说,熬过这一年,特斯拉的位置会非常好。
对看这场大会的人来说,真正值得带走的是几条可以落地的判断,欢呼声可以留在现场。下面这些建议,按可跟踪的优先级排好了:
- 盯住 FSD 规控模块切换成神经网络的时间点。这是整个自动驾驶叙事的最后一块拼图,也是估值重估的触发条件。看到端到端版本开始大规模推送,就是信号。
- 跟踪 48V 低压架构和去稀土驱动单元的量产落地。这两件事直接决定下一代车型的物料成本和供应链风险,比任何销量数字都更能说明成本曲线往哪走。
- 关注锂精炼厂的投产进度,别看锂矿新闻。真正的瓶颈在精炼产能,不在矿藏储量。Corpus Christi 这个厂的爬产节奏,比任何锂矿并购消息都有信息量。
- 把储能业务单独拎出来看。它的增速已经超过汽车业务,Megapack 在调峰场景打赢天然气,这条线被市场低估了很久。财报里储能那块收入的增速值得单独记录。
- 对 Optimus 保持观察,暂时别急着定价。硬件全自研、复用 FSD 计算机这两点是真的,可离量产和商业化还有很长距离。马斯克自己说它现在连煎蛋都不会,这句话就是最好的估值锚。
- 如果你持有特斯拉,把未来 12 个月当成宏观逆风的窗口,不要拿它当公司基本面的判断依据。马斯克自己的建议是别盯盘,跌了再买。
- 对定价动作别过度解读。特斯拉直营,价格变动全透明,看起来动作大。传统车企的调价藏在经销商终端,同样频繁,只是没人统计。看需求曲线,别看单次调价。
一场股东大会能同时讲清楚一家公司现在靠什么赚钱、三年后靠什么赚钱、十年后靠什么赚钱,本身就不常见。信不信是另一回事,逻辑链是完整的。
0:01[music] [cheering] [applause] Good afternoon, everyone. [cheering] Good afternoon, everyone, and welcome to Tesla's 2023 annual shareholder meeting in Austin, Texas. [cheering] [applause] My name is Martin Viecha. I'm Vice President of Investor Relations, and I will be the chair of today's meeting.
0:37first of all, I really wanted to thank everyone who continues to do this trip on annual basis and shows up every time. It's really quite something that we have a following as dedicated as this. So, thank you very much for that. [cheering] I also wanted to thank our first-time visitors. I know that many of you won a lottery to come for the first time to a Tesla shareholder meeting.
1:06You You You'll probably notice that, unlike many other shareholder meetings of other companies, this one tends to be a bit more fun. So, hope you enjoy that. But, the last thing I really wanted to say is that just thank you so much for all your support over the years. We really feel like you're part of our mission, you know, part of our you know, story, and it's something we truly, truly value. So, thank you so much for that.
1:35There will be two parts of today's meeting. First, the formal part of the meeting, where we will cover items that stockholders have been asked to vote on, as well as any other matters that are properly presented. And after the voting, I will introduce Tesla's co-founder and CEO and technoking, Elon Musk, [cheering] who will who will give a presentation about Tesla's year in review.
2:00But at this time I'd like to thank the members of the Tesla team and the board who are here with us here today. A representative from PricewaterhouseCoopers, Tesla's independent auditor, present with us as well. But before we begin, I'd like you to introduce you to Tesla's chair, Robyn Denholm, who would like to say a few words.
2:20[applause] I'm a little bit shorter, so. Hello everyone. On behalf of the board, I am really honored to welcome you all here today to the 2023 annual shareholder meeting. It's wonderful for me to see everyone in the room today and also the many investors that have joined us today via the virtual livestream.
2:52It's our third straight year of having the annual meeting here in Austin. At the first meeting, the factory was still in the early stages of development with equipment being installed and commissioned. During the 2022 meeting, production had started to ramp as we produced about 1,600 vehicles that week.
3:15And now, less than a year later, we have more than tripled the amount, having achieved production of 5,000 cars in a single week. [cheering] A huge congratulations to the Texas team. And on top of that, we are getting ready to roll out our first Cybertrucks. [cheering] This exponential trajectory of the Gigafactory here in Texas is a reflection of Tesla as a whole.
3:54In 2022 was another record-breaking year for us with deliveries growing by 40% year year over year. And that trend has continued in the first quarter of 2023. We would not be able to achieve this growth without the dedication of our amazing employees. I've been fortunate enough over the years to actually travel to many of our locations, all of the Gigafactories around the world, and I can tell you that I've witnessed firsthand how our exceptional employees in the in our global talent across the world has together achieved our mission or furthered our mission at the accelerating the world's transition to sustainable energy.
4:43A huge round of applause, please. This year, here in Texas, we also had our very first Investor Day where we introduced our company leaders from design to engineering to supply chain, manufacturing, energy, and charging. And they each spoke about the role of their teams in making our mission come to life.
5:09And I was thrilled that investors were able to hear directly from this outstanding leadership team that Elon has been able to put together over many, many years. It is on the strength of their leadership and the hard work of our dedicated employees around the world that I believe we are not only weather the macroeconomic environment that we've seen this year, but also a continued to be in the strongest position ever to advance Tesla's mission.
5:42As a startup, Tesla proved that electric vehicles could be fun with the Roadster. They could even be better than gas-powered vehicles in every single way with the S and X, and could be all of these things as well as affordable with the Model 3 and Model Y. Which last quarter became the best-selling vehicle of the any kind in Europe and the best-selling non-pickup vehicle in the United States.
6:14[applause and cheering] And today, as part as one of the world's largest global companies, we plan to change the paradigm again by forging the path to achieving a global sustainable energy economy as set out at our investor day in our master plan three. As Tesla grows, so do our customers' positive impact on carbon emissions.
6:43When I stood here last year, I proudly announced that in 2021, our customers avoided emitting 8.4 million metric tons of greenhouse gas emissions, the equivalent of over 20 billion miles driven by internal combustion engine vehicles. Today, I can tell you that in 2022, our customers avoided releasing over 13.4 million metric tons of greenhouse gases into the atmosphere, the equivalent of over 33 billion miles of driving of internal combustion vehicles.
7:27That's That's an almost 60% increase in total emissions savings year over year. Putting more of our products in customers' hands helps grow that number, and our goals of making our products greener over time and further decarbonizing the electric grid will help multiply that effect.
7:51As part of that goal, I am delighted to say that as described in our recent impact report, our global supercharging network was again 100% renewable in 2022. We've been able to achieve all of this while maintaining Sorry, maintaining industry margins industry-leading margins and generating strong cash flows over the last several years.
8:19Going forward, we will continue our focus on capital investment plans to support our future growth with investments in batteries, in vehicle production, including the next generation platform, in the building of our lithium refinery that we recently broke ground on, in in the energy storage factories, in our sales and service footprint, and in our charging network, among other things.
8:48Before I hand it back to Martin, I would also like to thank Hiro Mizuno for his dedicated service to the Tesla shareholders and the board these past 3 years. I would also like to express our enthusiasm on behalf of my fellow directors at the prospect of JB Straubel joining our board.
9:10[cheering] [applause] I don't I don't have to tell you that his passion for Tesla and green tech will make him an important asset in our mission to accelerate the world's transition to sustainable energy. So again, on behalf of the our full board, thank you for your tremendous support and welcome to your 2023 shareholder meeting. And with that, I'll hand it back to Martin.
9:42[applause] Thank you, Robin. I will call I will now call the meeting to order. Please refer to the meeting agenda that has been provided to you and posted on the virtual meeting site. The time is now 3:10 p.m. Central Time, and I declare the polls are now open. We have already received over the last few weeks voting proxies from stockholders, meaning that almost all the votes that will be counted were already submitted before today.
10:19However, if you wish to vote your shares or change your prior vote, you may do so through the virtual meeting site. For those who are in person here today, election ballot boxes were available to you at the check-in. Tesla's Board of Directors has appointed Computershare Trust Company to serve to serve as Inspector of Elections.
10:40Computershare has taken and signed an oath as Inspector and has certified that starting on April 6th, 2023, the proxy materials or a notice of internet availability of proxy materials were mailed were mailed or provided to all Tesla stockholders of record March 20th, 2023. We have majority of outstanding shares represented at the meeting, so I declare that there's a quorum present at the meeting, and we may proceed with the meeting.
11:13The items on the agenda are as follows. Number one, the election of Class 1 directors, Elon Musk, Robyn Denholm, and JB Straubel, to serve for a term of 3 years or until the respec- the respective successors are duly elected and qualified. Number two, to approve executive compensation on a non-binding advisory basis.
11:38Number three, to approve the frequency of future votes on executive compensation on a non-binding advisory basis. Number four, to ratify the appointment of PricewaterhouseCoopers LLP as Tesla's independent public registered accounting firm for the 2023 fiscal year. Number five, to vote on a stockholder proposal included in our proxy statement which relates to the reporting of key person risk.
12:07Number six, to vote on a stockholder proposal which the stockholders did not seek to have included in our proxy statement. Tesla's board has recommended that our stock stockholders vote for each of the director nominees, for the approval by non-binding advisory vote of executive compensation, and the ratification of of appointment of PWC as an independent registered public accounting firm for fiscal year 2023.
12:38And every 3 years for non-binding advisory vote for the frequency of future votes on executive compensation. With respect to stockholder proposal included in our proxy statement which relates to the reporting on the key man risk, Karen Roberts Doutre on behalf of SumOfUs is here to present this proposal.
13:01Ms. Roberts Doutre, can you please identify yourself? I would like to invite you to speak. You will have 3 minutes. Hello. thank you for having me here today. Under discussion is a shareholder proposal concerning key person risk at Tesla requiring Tesla's board to draft a report covering the steps taken to ameliorate the potential impacts of loss of key persons in the company.
13:34however, before I began, I would first like to say thank you to Mr. Musk. For so many of us, you stepped into fields where we had so desperately wanted to see change and through bringing in a talented team willing to put in long hours, you dragged reluctant industries kicking and screaming into a better future and for that, thank you. Building up this juggernaut around us which as you largely put it runs itself these days.
14:01thank you for that. And yet now here we are today at a time when Tesla's technological leadership should be on display, the investment community largely sees us at a drift with management focused on all other matters not Tesla. Watching as Tesla's brand favorability dropped by 15 points last year, something that cost us margins.
14:22We have such a spectacular product pipeline, a team that knows how to execute at scale with low cost of goods sold, something that we are not being properly recognized for. It should go without saying, but apparently bears repeating that if you cannot produce with low COGS, you have no future. When I look around the auto industry these days, I see a lot of companies that have no future. and that's a success on behalf of an incredible team here at Tesla that deserves recognition.
14:50When people look at this company, that's not what they see. They see the company as a synonym for its CEO. And then the discussion turns to everything except for where it should be focused. It's my sincere hope that by introducing this proposal, we can not only accelerate the development of management at Tesla, but also increase its visibility. The fact that this company is so much more than its key persons is something that the public and the investment community should see.
15:17A founder often finds their greatest success in ensuring the future of the company and the development of the management talent who can come after them. Think Tim Cook at Apple, Steve Ballmer at Microsoft, and so forth. People who are ready from day one to execute without controversy, without distraction, and build a future for the company.
15:39I encourage the board should embrace this opportunity in the most public manner possible to showcase the future being built right here, so that the discussion among the public and the investment community revolves revolves around that, and not around every public statement on every other topic made by the leadership.
15:59I strongly encourage the board they shareholders to support this proposal, and for the board to embrace it. Thank you very much. Thank you, Mr. Oberstar. The board has recommended that our stockholders vote against this proposal for the reasons as set forth in our statement of opposition in our proxy statement.
16:19Finally, as you saw, and Investor Advocates for Social Justice would like to raise a proposal from the floor. The proponents did not seek to have their proposal included in the proxy statement. The board recommends that our stockholders who are entitled to vote on this proposal vote against it.
16:38As a reminder, as is disclosed in our proxy statement, I've been advised by the proxy holders that they intend to vote all all shares of stock over which they have discretionary authority against this proposal. Only shareholders who are shareholders of record or have obtained a legal proxy may vote for this proposal after it is presented.
17:02Courtney Wix, representing Investor Advocates for Social Justice, is here to present the proposal. Ms. Wix, I would like you to speak. You will have 3 minutes. Thank you. Good afternoon, Tesla board members, management, employees, and shareholders. My name is Courtney Wex, and I am the executive director for Investor Advocates for Social Justice, and I'm here on behalf of As You Sow and the Sisters of the Good Shepherd, who have filed shareholder proposals in the past on child labor and human rights abuses in Tesla's supply chain. I stand here today to call on shareholders to vote yes on requesting Tesla to conduct a third-party report detailing the
17:43company's efforts to eradicate child and forced labor in its supply chain. The human rights risks that permeate throughout Tesla's value chain, when not adequately addressed, harm shareholder value and undermine Tesla's ability to lead the just transition. Growing attention and concern have been placed on child labor in cobalt mining in the Democratic Republic of the Congo, as well as on forced labor in China's Xinjiang region. 70% of the world's cobalt needed for EV batteries are sourced in the DRC. Over 40,000 children
18:19work in artisanal and small-scale mining, often without protective equipment and in dangerous spaces that frequently collapse. Many children are seriously injured or killed in the mining process. In Tesla's latest sustainability report, it claims to have audited 80% of the miners and refiners in their cobalt supply chain against responsible production standards.
18:44However, there are many industry back standards, some of which are weak, and it isn't clear what percentage of audits were conducted by which initiatives. It is also unclear how the company's due diligence ensures cobalt sourced from child labor is not intermingled in the industrial cobalt supply chain.
19:00Subsequently, an estimated 1.8 million people have been subject to state-imposed genocide, detention and internment camps, and forced labor in the Xinjiang Uyghur region. A recent study concluded that automakers cannot conduct meaningful audits in this region. The recent Uyghur Forced Labor Prevention Act in the United States Senate inquiry into Tesla's supply chain highlights the regulatory risk Tesla faces.
19:31We believe a third-party audit would provide greater transparency into how Tesla plans to eradicate child and forced labor from its supply chain. One of Tesla's competitive advantages is its sustainable leadership. Tesla, its board of directors and management team, have a moral obligation to pursue ethical sourcing practices that do not rely on the lives of children or enslaved people. Tesla can continue to make a true leadership difference that forges a sustainable economic future that is anchored by innovation, empathy, human development, and shareholder value. Thank you.
20:08Thank you, Ms. Wicks. We will now accept votes for, against, or to abstain for this proposal, but again, only for record holders or legal proxy holders. Please note, this is the final opportunity to submit proxies in order for them to be counted. I declare the polls are now closed.
20:40Based on proxies we have previously received, I'd like to announce on preliminary basis that our stockholders have approved the recommendations of the Tesla board on all agenda items, except that shareholders have recommended 1 year for non-binding advisory vote on the frequency of future votes on executive compensation.
21:02This means that I would like to officially announce our new addition to the Tesla board, JB Straubel. Yay! [cheering] [applause] After the final tabulation is completed, we will announce specific vote tallies filing in the filing form 8K within four business days of this meeting.
21:34That concludes the official business of today's shareholder meeting, which is now adjourned. During the course of the following session, we may discuss our business outlook and make forward-looking statements. Such statements are predictions based on current expectations. Actual events could result could differ materially due to a number of reasons and certainties, including those disclosed in our most recent form 10K and 10Q filings with the SECs.
22:11Such forward-looking statements represent our views as of today, should not be relied on thereafter, and we disclaim from any obligation to update them after today. And with that, please welcome Elon Musk. [cheering]
22:43[applause and cheering] I just want to say I love you guys. So, well, we've got a lot of great news to talk about today. The the Tesla team has done an incredible job of executing over the past year. and we have many exciting years ahead of us. So, yeah. [applause] So, with the Tesla Master Plan Part 3, which admittedly it was you know, quite technical. We wanted to go into a lot of technical depth.
23:28So, you know, it it was clear that they were what what we were saying were was not just were not merely assertions, but that they were backed up with physics and with real data, that it's realistic. And and and with the Master Plan Part 3, we want to basically the goal is to give people hope, re- realistic hope.
23:51and and maybe hope's even the wrong word, but simply to say that that there is a path to a fully sustainable global economy, that we are on that path, and that that we're accelerating that path, and that's and that so long as we don't get complacent about it, it will happen. So, [applause]
24:17So, I I'll I'll I'll just I'll I'll kind of rush through a re- a recap of the Master Plan Part 3. But, the you know, it's some of the salient points are that it will actually take less energy to go to to to go sustainable, not more energy. it's it's actually more And some of these things may sound very obvious, but it's it's more energy efficient to go sustainable, not less.
24:44There's a and and and there's less mining that is required for a sustainable energy energy economy, not more. Also very important. I I I and I heard a question raised about cobalt mining and you know what? We will do a third-party audit. So [cheering] [applause] so [applause] In in fact, we'll we'll put a webcam on the mine.
25:19[laughter] And if anybody sees any children, please let us know. so [applause] I I mean I I I don't want to go go off on a tangent too much on that front, but if it's very important to appreciate that most of our battery packs are iron-based. a majority of our battery packs are iron-based, not cobalt.
25:55and the And And then our other battery packs are nickel-based, and not cobalt. The nickel batteries use a little bit of cobalt as a binder, but only a tiny amount. In contrast, your phones all use 100% cobalt. [laughter] I recommend complaining to the phone manufacturers. [cheering] [applause]
26:25But even for the small amount of cobalt that we do use, we will make sure, six ways to Sunday, that we're that no child labor is being exploited. Obviously, we're a company that cares a lot about doing the right thing, and we don't want to dilute ourselves or dilute anyone else. So, yeah. Come.
26:42[applause] So again, just to re-recapping Master Plan Part 3 it's it's really the point is to say that it is very very doable. It is happening. we need a a threefold increase in solar and wind. we need to [clears throat] really some of these are 29-fold increase in vehicle stationary and thermal battery production.
27:12actually in in a nutshell, the way to think about sustainability is the the the faster we can make our battery packs, the faster we can move to a sustainable energy economy. That's the fundamental limiting factor. So I'll just grab some water here. [cheering] So Yeah, this I can't emphasize that point that point enough. The the the the rate of of of lithium-ion battery production fundamentally decides the rate at which the world transitions to sustainability because the the batteries are needed for all forms of transport and for stationary storage given the intermittent nature of of of solar and wind power. So
28:12So the and we do invite people to critique our analysis because you know any given analysis is going to be to some degree wrong. So we that's why we we we put it out there looking for for critical feedback to say well, perhaps we got some things wrong. Perhaps some things need to be adjusted.
28:33but generally the the feedback we've gotten is that actually our analysis is quite accurate and we've not we've not seen any rebuttals thus far that would cause us to change our assumptions. So, it's it means roughly 240 terawatt hours of battery pack or 240,000 gigawatt hours of battery need to be produced.
28:56But, it'll only take 0.2% of land area for solar and wind. So, it's a very It's not like we need to carpet the earth with solar and wind. It's just literally a fraction of a percent. roughly a $10 trillion manufacturing investment, which relative to the global economy is actually a small a small number.
29:20Yeah, roughly roughly 10% of of the world economy. I think obviously I think 10% of the world economy is a small price to pay for a sustainable energy future. [applause] It's also true that even when factoring in the cost or the the emissions required to produce an electric vehicle, which for now are a little higher than gasoline vehicles, when you look at the emissions over time, electric vehicles absolutely win by a long shot. and as as we're seeing the the the cost
29:58and the emissions required to produce an electric vehicle are dropping rapidly over time. and we're going to get we're going to get to the point where an electric vehicle is lower is cheaper than a gasoline vehicle. So, [applause] So, in in 2022, we avoided releasing about 13 million tons of CO2, reduced manufacturing GHG per vehicle by 30%, and reduced water usage by by 15% despite massive increases in output.
30:36So, And we [laughter] We also made our factories safer, which is really important. So, we track the the injuries per per person and we believe at this point we are best in industry or have the lowest injuries per person. So, [applause] we're now actually quite a big company from a head count standpoint. This is this is a our direct employment numbers. So, this does not count contractors.
31:09and for every manufacturing job, depending on on how you count it, there are there are at least five sometimes up to 10 jobs created because you got to look at the the total supply chain as well as the all of the support functions. So, you know, when you have create a factory like this, you actually create jobs for teachers, lawyers, carpenters, electricians, and and and restaurants, and and everything that's that's required to support a person in a factory. And this is why countries and and states are so interested in having
31:46manufacturing facilities in their location. So, anyways, it's it's a lot of people gainfully employed doing doing very useful things. we received 3.6 million job applications last year or so. [applause] And and and and once again, the the the top two most desired companies for engineers
32:25on Earth were SpaceX and Tesla. So, [cheering] [applause] you know, at the end of the day, the competitiveness of any given company is a function of where are the most talented people interested in working. That that is the team that's going to win. In fact, if if I'd say this is general generally the case. if you look at any given company and say, "Where are the most smartest, most driven people going to work?" That company is going to win.
33:00so, whe- whether it's Tesla or or any other company. we're also excited to announce our next-gen drive unit. which is a a big reduction in silicon carbide. It's half the factory space. Notably, there are zero rare earth elements required. So. Yeah. And we're also changing to a 40-V 48-V low voltage architecture in the in the car.
33:37for the Yeah, this is a big deal, actually. So, the cars have been operating with 12-V batteries for basically about a century. So, for the first time in I think over 100 years, we're actually going to change from 12-V voltage of, you know, outside of the drivetrain to a 48-V architecture.
34:06and to first approximation, that that means we need only about a quarter as much copper for in in the in the car as would be needed for a 12-V battery. So, that's a big deal cuz people often worried about, "What you know, is there enough copper?" Yes, there is. So. [applause] and FSD beta is growing hyper exponentially.
34:36So, that that that chart is going to look like a wall, basically. and I just a question for for those in the room, are you have How many people have tried out FSD beta? Okay. [cheering] Great. [laughter] So, so what what what do you think of the latest build? [cheering] Right? So, it's it's really getting to the point where it's at least for me when I drive around, it's several days between interventions.
35:10And I think we're we're getting to the point where there's really just one last piece of of the system that needs to be a neural net, which is the the planning and control function. and and so we we expect to have that last piece become a neural net, so it'll be end-to-end from video into control out as a neural net.
35:35and yeah, so the thing the thing to appreciate is it's it's not that full self-driving will be as good as a person, it will be much much better. Like a lot, like over time and 10 times more 10 times safer than a person. It's not even going to be a contest, frankly. so this is this is a really big deal.
35:59and I think some people realize it, I think you guys probably realize it, but the being able to do a software update and have several million cars suddenly go from manual driving to autonomous, I think will be the single biggest asset value increase in history. so [applause] The the normal usage of a passenger car is is quite roughly 10 to 12 hours per week. Call it maybe an an hour and a half per day.
36:39And when you drive around you see lots of cars just parked in parking lots because out of the 168 hours in a week they're using less than less than 10% maybe 7% of the hours of the week a a car is in use. But once it is autonomous it it can be used probably I don't know 50 50 hours a week maybe more.
36:59So it's it it if it is effectively a fivefold increase in the value of a car overnight. I'm actually surprised that so few people realize this or or maybe they just don't believe it it's real. but but it is. So this is this is really an an insanely big deal. we're we're also the the largest EV maker in the world.
37:29so [applause] Yeah. And I think that will continue. and while at the same time being the highest margin of any car maker in the world. [applause and cheering] Now making electric vehicles profitably is is hard as illustrated by the difficulty of our competitors. a number of a number of our competitors are making EVs at a significant loss.
38:10so but we are not. We are actually making EVs profitably and almost no one else is. This is hard. So, [applause] And I'd say for this is it's a massive credit to the Tesla team for I I I can't tell you how hard manufacturing is. Like, you probably heard me say that prototypes are easy, manufacturing is hard.
38:38and and then and then manufacturing at scale with positive cash flow is excruciating. It's mega pain. but the Tesla team has done it. So, [applause] And as you can see, our free cash flow per year has been increasing steadily. so, you know, we're we're making good progress. I mean, it it it it should be said that And and I've made some of these comments that the the that interest rates make a a very big have a very big effect on the affordability of cars.
39:24So, the vast majority of people buy cars based on the monthly payment. So, it's like how much money how much is the monthly payment? and it's and it's it's not a it's not a question of of of value for money. It It's just do they actually have enough money? Can they afford it? So, so for the vast majority of people, it's just can they afford to pay the payment? As the interest rates increase and and credit tightens I could safely say that the you know, these various banks that have died are probably somewhat distracted from handing out auto loans. You know, it's like if they're on the way to the cemetery increasing their auto loan portfolio is
40:11not the first thing on their mind. So, this is going to be a challenging I'd say challenging 12 months. I want to be sort of realistic about it. that Tesla is not immune to the global economic environment. I I expect things to be just at a macroeconomic level difficult for at least the next 12 months.
40:39like Tesla will get through it and will do well and I think we'll see a lot of companies actually go bankrupt. so I I I don't I don't I want to make sure that this is not just the good news parade. It's important to to understand that no company is immune to the macro macroeconomic environment.
40:57But that said the the it it won't be darkness forever. I expect probably a year of difficulty globally for everyone. and then my my best guess is that the global economy turns around in roughly 12 months. and and then Tesla will be in an ex- extremely good position. So, anyone who is a long-term investor, I think will do extremely well.
41:27[applause] [cheering] And here I want to I give a a big shout out to the Tesla in internal software team. This is [applause] It it it's actually a really big deal that Tesla has such a powerful internal software team. I mean that that that software team is responsible for handling the entire customer experience from from buying the car, delivering the car, operating the factories, service and support. we internally wrote all of the insurance and financial software.
42:19the it's all the supply chain logistics stuff, the data centers and infrastructure. and the analytics and insight. This is all internally written Tesla software. I think there's there are almost no companies in the world can do this, because they do not have a a very talented internal software team. So, they're generally reliant on third-party enterprise resource planning software. So, but Tesla is not.
42:48This This is This is a hidden strength of the company, that that often doesn't get a lot of attention, but is incredibly powerful. so once again, I'd like to say thank them for their work. [applause] And as predicted, we are highly confident that Model Y will be the number one best-selling car on Earth this year.
43:15[applause] And and in fact, in Europe, the best-selling car of any kind yeah, what whatsoever. It's It's already was that in Q1, I should say. it's not like will be it it was that in Q1. and China, the best-selling SUV. and the United States, the best-selling non-pickup vehicle in the United States in Q1. So, you know, it's going well.
43:45[applause] I don't know if this is like the optimal picture, but the the it it it Tesla cars are actually the the safest cars in the world. We put immense effort into vehicle safety, and we keep updating the safety. So, we keep improving the automatic emergency braking system just with a software update. And we keep improving the airbag deployment with software updates. So, we close the feedback loop on like when we see an accident we analyze the accident.
44:23We say, "What can we do from a software standpoint?" Cuz there's actually quite a lot you can do to first of all avoid the accident cuz the best accident is no accident. And then if the accident occurs, how do we deploy the airbags and the the sort of the the the seatbelt pretensioners to minimize the probability of injury.
44:43and So, even even for cars that people have owned for many years we are behind the scenes continuously improving the safety of of of your car. And I mean, there are even some things that I think a lot of people aren't aware of that we we actually have automatic cabin cabin overheat protection.
45:04So never in the history of the company has a child or a pet died in a Tesla. [applause] [cheering] Okay. So, it's it's one of those like little known features, but because we have a large battery and we're monitoring things all the time, we can make sure that the cabin temperature never gets to lethal levels. and I think this is this this yeah, big deal. So.
45:39oh yeah, so and then while lots of cars will say they've got like, you know, five stars or or whatever there's there's nuance to that. when we do the the European NCAP sort of active safety tests we we got the highest scores that they've ever seen. So, yeah. [applause] So, we we we we got a a 98% score on the active safety system, which obviously isn't good enough.
46:16and we will but we we we we we have got a game plan for getting rid of the last 2%. [laughter] [applause and cheering] And then it's worth noting things like the total cost of ownership of a Model 3 is now comparable to that of a Toyota Toyota Corolla. So. [applause] superchargers, we're making good progress. So, our supercharger uptime is now 99.95%.
46:53So, yeah. [applause] And we have superchargers practically everywhere at this point. going to Megapack, cuz this is stationary storage is an important part of solving the sustainable energy problem. and the mega the the Tesla Megapack is now more competitive than a natural gas peaker plant. So, we have very strong demand for the Tesla Megapack. and we're we're going to make a lot of them.
47:27So. yeah. [applause] and a while back I predicted that the Tesla stationary battery pack business would actually grow faster than the automotive side of things, and that's exactly what has happened. you can see it's it's an it's an exponential curve growing at a rate that that is even even faster than our vehicle sales.
47:54So. Yeah, sorry. We're We're expanding Gigafactory Nevada for the Tesla Semi production line and for the 4680 in in in-house cell production. And yeah, some Megafactory stuff. So, the This is So, we're aiming for 40 gigawatt hours a year both in California and in China, but I think long long term, this will be much much greater than this. I mean, I would say it's I wouldn't be surprised if long term the stationary battery pack activity went well in excess of 500 gigawatt hours per year.
48:50So, the the the the demand is like quite quasi-infinite here. So. as we look ahead to tackle the what we see as the the choke points in the supply chain, one of them is lithium refining. and I mean, there've been a been a few times on Tesla earnings calls where I've said, "Can someone please just do lithium refining?" because it's it's there's a shortage of it.
49:25and but we we we we really see very little activity less outside of China for for lithium refining. So, it's not that we wish to take on extra problems, but since nobody else was doing it, we felt we had to do this. And so, we've just broken ground on a lithium refinery in Corpus Christi, which will be do you more lithium refining than I think probably everything outside of China.
49:54So, yeah. [applause] And we're signing a deal for our next Gigafactory in in Mexico, Giga Mexico. and I I I think that's going to be a fantastic factory. So, yeah, we're we're we're we're laying the groundwork for ultimately getting to 20 million vehicles a year. [applause] [cheering] Yeah.
50:32[cheering and applause] The So, Cybertruck is a hard car to make. it's because it's a such a radically new design, it actually you can't just use conventional methods of manufacturing. we had to invent a whole new set of manufacturing techniques in order to build an exoskeleton-based car instead of an endoskeleton-based car. So, it's it's extremely non-trivial to to build the Cybertruck.
51:07but we're making good progress on that. We have the So, so yes, it's I that that's the the the the the thing I when I in the in the factory, I I saw the the the Cybertruck line and see how how we're we're doing there and and sorry for the delay, but we're we're finally going to start delivering production Cybertrucks later this year. And I think the product, if anything, is better than expectations.
51:36[cheering] [applause] I mean, the Cybertruck is the car is the car I will be driving on a day-to-day basis. So, and you know, people always want to know what our next product is. [cheering] But, this is not the time to [cheering] We we obviously need to have a, you know, a proper dedicated product launch.
52:09I I I just want to emphasize that we are actually building a new product. We are actually designing a new product. we're not sitting on our hands here. so, there are two new products that I think you will be very excited about. [cheering] And both the design of the products and the manufacturing techniques are
52:38head and shoulders above anything else that is present in industry. So, yeah, in other words. [applause] I- I- if I were to guess, I w- I I would say of these these two new products, just these two new products alone, I I would say there's w- we'll probably make us this is just, you know, Elon's guess.
53:06so, you know, don't sue me. [laughter] but I you know, Elon's guess is that that that we'll probably make in excess of 5 million units a year of these two mo- models combined. So, [applause] [cheering]
53:53These are all real, by the way. [cheering] [applause] [cheering] I think the thing perhaps perhaps most most notable is if you look at the difference between the last time we showed Optimus and and this is a video that was taken basically yesterday and the Optimus team was up all night making this video.
54:58The Optimus team has done an incredible job. So just yeah. [applause and cheering] And it's the the the motors, the controllers, the electronics, and everything you see in the Optimus robot is a Tesla designed a system. So this is we actually tried to find drive actuators and and whatnot that that were off-the-shelf. We We found that there there weren't any. in order to make an an an effective humanoid robot, you actually have to design the motors and gearboxes and the and the electronics from scratch because it's a very different application from anything else that exists. so, we took our world-class motor and power electronics team and and said and and
55:52said, "Okay, we we we need to to design several actuators that are that don't exist in the world." And they did. So, Optimus is is is working quite well. And then, for full self-driving, a- as full self-driving gets closer and closer to generalized real-world AI, that same software is transferable to
56:19a humanoid robot. just like you know, we you know, humans can obviously walk around with their arms and legs, but but we can drive a car, fly a plane, steer a boat, ride a horse. if you have a generalized understanding o- o- if you have generalized real-world AI, which is what we are developing for full self-driving, it can be transferred to basically anything.
56:50and so so Optimus will use the same FSD computer as as the car. and the the Optimus stuff is is I think somewhat somewhat extremely underrated. Peo- peo- the the because the people just cannot comprehend the the the the the consequences. Now, obviously, we need to make sure that we don't have a Terminator scenario. That's very important.
57:18it's all fun and games until Terminator shows up. but if you say like if you have had a generalized humanoid robot, what would be the effect of ratio of humanoid robots to humans? Cuz I think basically everyone would want one. And and maybe people would want more than one. Which means the actual demand for for something like Optimus, if it really works, which it will, is I don't know. I mean, 10 billion units? It's It's It's some crazy number.
58:05it might be 20 billion units. If the ratio is, say, two to one on people, you know, humanoid robots versus people, it might actually be It's It's some very big number, is what I'm saying. and a number vastly in excess of the number of cars. so my prediction is that Tesla's long-term value will be a majority of long-term value will be Optimus.
58:33and and that prediction I'm very confident of. So, anyway. [applause and cheering] Let's see. So, I think with that we can do questions, perhaps? Oh, oh, yeah. What What is it? Oh, yeah. We also want to make sure, please retail investors, please please vote. normally retail investors for most companies don't vote, but you will We We really care about your vote, of of small investors, not just lot of large investors. So, please do register to vote. your vote is very important.
59:15and yeah, I think that's Yeah, that's that's it. So, [applause] yeah. [cheering] [applause] So, with that we can go to ad hoc questions from the audience. Okay, definitely the person in the Optimus outfit with the red cowboy hat. [laughter] Great outfit. Sir, You know, we'll try to get through as many questions as possible. And you know, you can hit me with your toughest questions. It's fine.
1:00:10[laughter] Topic has to be about Tesla. Okay, it is. Hey Dad, it's your son Optimus. [laughter] I wanted to I wanted to Do you Do you see it more likely that you guys will have an RV first or a cyber camper? good Well, we're we're we We don't have any plans to build an RV quite yet, although I I can certainly see how Cybertruck could be converted into an an RV.
1:00:41And we are adding a lot of sort of attached points to Cybertruck, so others can build things. You know, you know, somebody can have a startup or other companies can build things that are attachments that enhance Cybertruck and turn it into a camper essentially. cuz I think it'd be really cool to have all these like third-party things available for Cybertruck.
1:01:05Thank you. All right. I mean, you could also probably just You're that close, you just say it or something. I don't know. [laughter] Actually, we'd like you guys to queue up behind us. There's three mics. All right. All right. Yeah, so so yeah, queue up. And It it it's Try to make the question short so we can get through a lot of questions.
1:01:33So, Maybe. Oh oh, you mean just like a just like a Tesla team doing a podcast about Optimus. Oh, yeah. Yeah, sure. Yeah, you know, I I I wonder if maybe we could add a have a sort of like a an an addendum after the earnings call which which is like a you know, 30 minutes pod podcast on Optimus or something like for those who are interested. Okay, I think we'll we'll do that.
1:02:09So, [cheering and applause] All right. Hey, Elon. Thanks for doing this today. It's an awesome event. Thank you. Welcome. Thanks for coming. Hope you guys are having a good time. Yeah, having a great time. I I I'm pretty confident we have the funnest shareholders meeting by a long shot.
1:02:28Undoubtedly, yeah. It's amazing. So, So, I'm really excited about a lot of things with Tesla. It's hard to pick a good question that it's going to cover everything. But, with the energy side of the business, I think it's hugely underrated. I mean, we got tons of problems as I mean, California with rolling blackouts.
1:02:43Sure. How do you think you can incentivize people, you know, more for solar, battery backup, and we've got the referral program's been beefed up as well. So, you know, what are your thoughts there? How we can get people on board? Yeah, actually, one of the challenges in California and a few other places is that in order to have the Powerwall be be like an uninterruptible power supply for the house, we have to actually have a switch that disconnects you from the utility.
1:03:14Otherwise, you just you know, send energy into the grid and it doesn't do anything. and that's actually been quite difficult with the utilities, but I'd like to thank PG&E for for recently, helping us, get approval for installing the backup switch for Powerwall in California. so that people can have uninterrupted power in California. And we're working on that, in in, every jurisdiction. So, so I'm I'm confident that that that's that's a that's going to be a great long-term, situation. And then as I mentioned, Megapack, is is growing, very very rapidly.
1:03:53and yeah. So, things are I I I mean, things are really growing like ridiculously fast on the the energy storage front. So, yeah, I feel I feel good about that situation. All right, next question. or hello. Hey, Elon, a lot of people have just kind of been wondering, how are you doing? Just just wondering, what's up? How you doing? How you feeling? We saw you were partying in Cabo recently. So, just What? Me? Must have been someone else.
1:04:24Yeah, so how how are you doing? Just human to human, how you doing? Well, I have to say, it's been a bit of a you know, roller coaster situation. and you know, like it's it's actually pretty rare for me to be at at a at a party. so that that's like the you know, the first party I've been at in a in a while.
1:04:49and actually wasn't even going to go to it. Then my my brother talked me into it. [laughter] So, thanks, Kimbal. so, yeah, I have to say, like the sometimes the the the the the work pain level is quite excruciating. and then and then on top of that, I get dumped on in the press. So, it's like, you know, it's not exactly super fun.
1:05:12but you know, there was a you know, a short-term distraction cuz I had to do like some major open-heart surgery on on Twitter to ensure the company's survival. but that I think Twitter is now in a stable place and obviously very excited to have Linda Yaccarino join and so I I I think Linda's going to do a a great job running running Twitter.
1:05:38and I'll I'll provide obviously guidance on and you know, technology development, but as you can see I I think Twitter at this point has released more new features in the last 6 months than it has in the last 6 years. So you know, obviously this is not a Twitter meeting, but but but but but the the the larger point is that I feel at this point I don't need to devote incremental time. the amount of time that Twitter will take going forward is is relatively small compared to the last 6 months. so, yeah.
1:06:11[applause] you know, I'm I'm you know, apart from like there's like some macroeconomic things that we can't do anything about. overall I feel very good about the health of the companies and I'm increasingly optimistic about future. thanks for asking. It's kind of you to ask. Yeah.
1:06:31[applause] Have you guys considered stretching out the financing terms on your auto business you talked about earlier consumers don't necessarily look at the sticker price they're paying. They're all they're concerned about is monthly payment. Right. And one of the fundamental advantages of an EV is that it's substantially more durable. And so would you guys ever consider stretching terms out to 10 years? Especially in this rate environment given the shape of the yield curve, you know, extending the terms out that far, there you'd actually be capturing a lower cost of funds.
1:07:05Yeah, I mean the important thing to appreciate is that like that that that the vast majority of the the financing is from banks. So, it's kind of whatever the banks are prepared to do. I mean, this year, I guess we'll do on the order of a hundred billion dollars of sales. and and we don't have a hundred billion dollars in in cash to finance that. So, the vast majority of is got to come from you know, from banks.
1:07:32So, it's really the banks' choice of what they're willing to do. and it's it has been tricky with with with banks recently because as I mentioned earlier, it's if if you're a bank and you're not sure whether you're going to be alive or not, you're that's it's a pretty big distraction and increasing your auto loan portfolio is very much a secondary pri- priority. So, it's just important to appreciate it's not up to Tesla, it's it's up to the banks.
1:08:02and there's not there's not some easy solution where you can say they have a so-called a captive financing arm, etc. cuz you have to securitize those loans and the market for securitizing them is is is weak. There is no the I wish there was an easy path, but but there isn't. So, you know, I think just important to remember remember that the economy moves in cycles.
1:08:27and we've had a very long period of up cycle. The next 12 months will be, I think, difficult for everyone and I I think you know, when Berkshire Hathaway had their annual meeting, Warren and Charlie actually said like, "Hey, this this year the Berkshire companies are going to make less money." and so, and you know, they're they're they're very well-run organizations and I think that's just generally true for the economy. So, you know, but but if but but but but the it it's also important to remember that, you know, there there are good times and and there are dark times, but then the good times follow the dark times. So,
1:09:12my advice would be don't look at the markets for the next 12 months. [laughter] If if there's a dip, buy the dip. [laughter] and I I think you will not be sorry because the the the you know, the there's just generally sort of a there's an economic cycle to things, but things come back up. So,
1:09:37my guess is tough times for a year. Like like I said, just just just my guess. and and but then Tesla will emerge stronger than ever. and the the long term, if you say like net present value of future cash flows, will be incredible in my opinion. All right. Hey, Elon, meet Kevin here.
1:10:03Hey, how's it going? Good, thanks. So, Tesla is the largest position in my ETF, ticker PP. And that's for pricing power. Right. And I'm curious. I really think that some of the features that you highlighted here, like the over-the-air airbag deployments of improving features, the safety features, are great things that we could be advertising. For I've know you've heard this a many times before. I know you have.
1:10:28No, no, but Sorry, sorry, go ahead. Go ahead. Go ahead. Yeah. 525 bucks off every car this year is half of Netflix's advertising budget. 1,000 bucks is the entire Netflix advertising budget, and I see their ads everywhere. Why not advertise these things that you told us here? I feel like nobody else knows about this. I just talked to Gordo, the Tesla bear, and he's still talking about 2016 New York Times pieces. There's people are in the past, man.
1:10:54That's That's 7 years ago at this point. I mean, that's what I'm saying. Yeah, I it's actually I I talked to a lot of people and they still think that like Teslas are like super expensive. They're like, "No, actually the starting price for a Tesla is below the average auto price in the US." That's like And that's a good thing.
1:11:13I mean I think it's it's it's good that it's that way, you know, like we Our Our goal was always to make a cars that are affordable to the general public. so So I I I I hear you. I I mean I think and it's indeed ironic that, you know, what Twitter is like highly dependent on advertising. So you know, here I am it's like you know, not never used advertising really before and now have a company that's highly dependent on advertising. So I I guess I should say advertising is awesome and everyone should do it.
1:11:46[laughter] But but I I I I think I I hear your your sort of larger point, which is that there are amazing features and functionality about Teslas that people just don't know about. and So yeah. [applause] And and and although there's there's obviously a lot of people that follow like say the Tesla account and and and and the you know, my account, whatever on on Twitter.
1:12:18to some degree it is preaching to the choir. And the choir is already convinced. So I I think what you're saying does does have some merit and you know what? I I I believe in taking taking suggestions. So we'll we'll we'll try a little advertising and see how it goes. [cheering] [applause] Okay.
1:12:46[cheering] [applause] Okay. All right, I was I was I wasn't expec- I wasn't expecting that level of enthusiasm. But okay, it's it sounds like our shareholders if I read between the lines [laughter] subtle as it is are saying we should probably do some advertising. Okay, very well.
1:13:06All right. [applause] Hi Elon. Hi Tesla team. thank you for everything. So, my question is regarding you recently asked Steam to the car and that's great because I play video games. Is Tesla like currently developing the ability for other parties to develop apps for the car? For example, if if we can buy actually customizable themes for the operating system of the car? I think you you can make some revenue there.
1:13:33Like that's that's space themes for the car or Lord of the Rings or stuff like that. Do you have any interest in that right now? Well, I mean important So, I I think the the notion of having entertainment apps and whatnot becomes increasingly important as the cars achieve self-driving.
1:13:53because otherwise the the you you you can only really watch play games or watch movies or whatever when the car is stationary. you know, which you know, if it's charging or what whatever that then you can use it. But when it's when it's at the point at which it's full self-driving, I think it's very much about entertainment and productivity.
1:14:12so, I think that is something that might make sense in the future. the but overwhelmingly the focus right now from for for Tesla software apart from improving safety is achieving full self-driving and then getting full self-driving to be much safer than human driving. and then once as we once we achieve full self-driving, I think that there could be a lot of opportunity for for apps and whatnot.
1:14:41So, if it you know, even the car as it is today for a Model S and X, we've we've made a a sort of a a full Steam integration. So, you can actually just literally log on with a with a Model S or X with your Steam account and play any game that is on Steam, which is a lot of games.
1:15:00which is pretty wild. I I'm I think a lot of people don't don't know that. you can play with a keyboard and mouse. You can play with a joystick or controller. you can also play Netflix, YouTube, or anything else. and for my little son X, he he loves watching you know, actually loves loves watching Rocket videos in the on the in the rear screen of the the Model S and X.
1:15:26So, Yeah, so it is something that we will do long-term. But but but one of the things we already have a lot of stuff there and and we'll do more as self-driving becomes a a reality. So. Hey Elon, I'm John or Dr. Know-it-all on YouTube and I don't know it all. But anyway, my question is as you guys move completely or finish the transition in full self-driving from version 1.0 of the software to 2.0, what do you perceive of as the like potential stumbling blocks after that? Cuz it seems like that's the major you know, kind of step change that needed to happen. And so, you're almost done with that and what happens after that. And also, please let me interview
1:16:12about you you about full self-driving. Like, I'll I'll spend 2 hours with you about just that. So, thank you. Well, there's there's what what you can really think about full self-driving as a sort of a march of nines of reliability. So, you we we we we need to get to the point where the probability of injury is
1:16:34lower than that of the average human driver. And then ultimately, probability of injury is much lower. So, even at the point at which the car is capable of driving itself, there's still actually a lot of work to do cuz you know every year there's a million people that die in auto accidents. And I think probably everyone in this audience has friends and family that that have died or or been seriously injured in auto accidents.
1:17:01and and for every there's like a million people that died there's there's 10 million people roughly that are seriously like permanently injured. and and and and and so it matters that we you know if if if we can get that down by an order of magnitude, you know, there's like 900,000 lives saved per year, maybe 9 million severe injuries prevented per year.
1:17:24and so that that really I think morally has to be our primary goal. and I I guess I'm open to an interview on on FSD. 2 hours is a long time, but Okay. Okay. All right. I'll I'll I'll do an interview. It's going to be on Twitter, but sure. [laughter] All right. [applause] Hi. Hi Elon. I'm Josh Phillips, long-term retail investor. My question's about battery grade lithium supply in the next 10 years. you know, lithium experts all agree that mines are just not coming on online fast enough to meet battery supply. And actually Drew Baglino pointed this out recently, some mines have take 10 years plus to be permitted. And even GM has actually announced like huge lithium deals that
1:18:20will like secure away supply from the rest of the industry. So, what's Tesla's plans to get more supply of lithium at the mine level outside of refining, but also at a deeper level than a traditional off-take. Because as we know, he who controls the spice controls the universe. Sure.
1:18:39So, I I actually think that the the the industry analysis is is incorrect. And the the the constraint is fundamentally that of processing. So, Now, our refinery in Corpus Christi that we're building is you know, primarily oriented towards refining spod germaine, of which there is a truly vast amount in the world.
1:19:10About I think about 3/4 of our lithium comes from Australia. and frankly, you could increase the rate at which the mines are operating. and the limiting factor is is is is not how fast can you mine, but how fast can you can you process. So, the the the the the mines are only going to produce ore at the rate at which refineries can handle the ore, or they'll it's pointless.
1:19:36So, I I think my opinion I obviously I I I could could be wrong about this, but it is is that it's it's really not not about the lithium ore. Lithium is extremely common. Lithium is everywhere. In every every country's got lithium. It's It's not like oil. So, lithium is one of the most common elements on Earth.
1:19:55but taking the lithium ore and refining it to battery grade is extremely difficult because the purity levels required for a battery are extremely high. If you have even a small impurity, then you will degrade the life of the cell dramatically. So, you need ultra pure battery grade lithium. And that's why That's why our focus is on refining as opposed to to mining. Thank you.
1:20:20[applause] Hi, Elon. thanks. It's a great time here. you spoke about safety in your presentation of the vehicles. And I just want to say I'm thankful to you for building such a safe vehicle cuz I'm here today with my son. We drove in his Tesla 3 over here that he replaced with the one that he got rear-ended significantly and it destroyed the car.
1:20:44But we're able to be here today, so Great. Thank you very much. Absolutely. [applause] So which way do I look? Okay. Hey. How's it going? My name's Austin Gregory. I'll ask about this because I don't think it gets near enough love. I was wondering if there are any updates on the next gen Roadster actually.
1:21:10Yeah. I know I know it's sort of like the cherry on top and all that stuff. You guys have a lot of a lot on your plate if you've said before, but if there's any updates on like the release timeline, all that stuff, the SpaceX package maybe. Yeah, yeah. That'd be amazing. Thanks.
1:21:24Totally fair and reasonable question. where is that thing? [clears throat] so we expect to complete the engineering and design of the next gen Tesla Roadster this year and hopefully start hopefully start production this is not a commitment, but hopefully start production next year.
1:21:48So but I I I it is it is like it is like as you alluded to it's it's it's not even the icing on the cake, it's the cherry on the icing on the cake. so I would I wouldn't expect it to be like it's definitely not going to be a huge contributor to to revenue. It will be a modest contributor to profitability, but it will be sick.
1:22:13[laughter] [cheering and applause] So and you know, there's some value to me running two companies because the the next gen Roadster will have the SpaceX option package. So that and that that'll make it truly next level. So Hey Elon. Eva McMillan here. First of all, I want to thank you for making the absolute best, safest, and most fun cars in the whole world.
1:22:49Cars in in Florida, you could we are seeing many many more cars in Florida than a year ago. Many fact, cars have doubled. So the question is we all want Tesla insurance. Oh yeah. coming? Oh yeah, yeah. actually, that's a great question. Okay, let us as Zach Zach Zach says it's coming later this year. But I I have to tell you like the one of the thorniest things in a in the United States is car insurance or insurance in general because it's a state by state and every state's got different rules and regulations.
1:23:26so that's that's why it's not like rolled out nationwide. I think really this should be one national standard for insurance, especially car insurance. I I think it would improve the cost effective cost effectiveness of of car insurance. but our but our intent is to to roll out nationwide and ultimately internationally.
1:23:47but there is a staggering amount of paperwork that is that is needed to to to get it done. But as Zach said, we expect to offer Tesla insurance in in Florida later this year. Thank you. [applause] Thank you, Elon. Thank you, Elon. Alexandra Merz. I am Tesla Boomer Mama. Very happy to be here.
1:24:10[cheering and applause] so very grateful to the Tesla team and you, of course. Just want you to know we love the emotions every day. Joy and less joy, but we're there. Shareholders, it's not easy. We're here not because it's easy, but because it's hard. So, my question initially was on in-house financing. You already answered that, and I want to congratulate you to the double investment grade rating. But then I asked my Twitter subscribers, I'm sure you're going to appreciate, what they wanted to bring forward, and that question made the cut. Here we go.
1:24:42Automotive, including SAS, peak margin, the same for energy. What number, and by when do you have the best Elon guess that we get there? So, I'm not sure I fully understand the question. is the peak margin for automobile mar- for automobile, including SAS, and energy, and by when will we get there? Your guess. I know nobody will sue you. Sorry, what's SAS?
1:25:09Well, service software as a service. Oh, oh, okay. You you mean FSD full self-driving? FSD. What's going to be the top automobile margin? Oh, oh, yeah, yeah. I mean, that what I was saying earlier is it is that at the point at which you have a truly autonomous vehicle that can drive around with with no one in it, the utility of the vehicle, I I think, as a rough guess, is probably five times what it is today because, you know, people will drive an average of an hour and a half a day, perhaps, so 10 hours a week. but if the car is autonomous, it can probably I'm just this just speculation, but it probably can operate for a third of the hours in the week, which
1:25:55would mean 50 or more hours. That's a fivefold increase. But but the car costs the same. Sure, I understand, but what is going to be the margin for Tesla then? Because you're obviously thinking you're going to sell much more FSD then, right? You may have a system in place of robotaxis.
1:26:14You You have a cut with fleets. So, did you do any projections what the automotive peak margin for Tesla could be in 2 years, 5 years, and the same for energy? Well, I mean, this is definitely we're in sort of highly speculative territory here. But but but obviously if if you've got a car that costs the same and and has a I don't know a 20 or 25% margin and suddenly it is able to be used five times as much, then then you might have 80% margins.
1:26:53And the revenue would increase severalfold. That's why I say it's the it's probably going to be the biggest asset value step change in in history of Earth. Yeah. And And and it and energy does not have the energy just simply does not have such an increase in asset utilization. So, I would expect energy to be to remain at sort of 20 20 25% ish to to 30% margin.
1:27:28All right. So, you said ask the tough questions. You're still good, right? Yeah. All right, cool. First of all though, as a family man, I just want to say there's a couple of things like a loop for tether of us of a car seat going through the headrest, which would be super helpful.
1:27:43There's a couple other things that would be really nice. Model Y is the only one you can do that in today. Maybe we could fix that. But when we're thinking about 4680 battery cells, should we be thinking about the efficiency and the performance we're getting today, or is there some updated timeline path that we could see closer to battery day presented efficiency performance that we can look to in the future?
1:28:11well I think First of all it's really important to expe- like express the difficulty of you know, like for Tesla to go from from nothing to making a battery cell that we aspire to be better than any other battery cell on Earth even when compared to companies where the only thing they do is make a battery cell is is obviously not a trivial exercise.
1:28:36that that said, we do see a path to to, you know, high very high energy density and higher energy density and lower cost than than anything else out there. but it's but it's a hard path. It's a very hard path. and and normally this it it would be absurd for companies to attempt such a thing. And no other car company is is is really attempting to do anything like this in in a serious way.
1:29:02So, I I guess technically BYD cuz they started out as a battery company, but yeah, so it's it's a it's a very difficult thing to do. but I think we are tracking to success in that regard. So. Hey Elon, first of all, thank you for undoubtedly making the world a better place for my son to live in.
1:29:26Thank you. Kind of you to say. once Cybertruck is fully ramped in Austin, what is the target production? And also there's some rumors that you're thinking about stepping down as CEO, please say it ain't so. [laughter] It ain't so. so. [applause] [cheering] Yeah. yeah, so I I think Tesla's going to play an important role in in AI and AGI.
1:30:04And I I I think I need to oversee that, make sure it's it's good. So, cuz that's that's a thorny problem if there ever was one. you know, like I I think generally people do not or very few people, even in the AI community, do not appreciate just how much capability Tesla has in AI.
1:30:34it's by far the most advanced real-world AI. There's no one even close. and reality has the most degrees of freedom. So, I I got to make sure that's good. So, and you were saying to me? Cybertruck fully ramps here now. target production. So, it I said, you know, we'll we'll start production later this year. We'll start handing over cars later this year. There will be an S-curve of production, so it'll be slow at first and then and then ramping up. and I guess we'll see what the demand is like.
1:31:14so, I I mean, I think there's we're likely to do probably a quarter million a year, I think, maybe more. again, very much dependent on on on what the demand is like and and it's it's it you know, we we don't just need to ramp up production, but we also need to to improve the the the production cost efficiency.
1:31:39So, which is going to be also a very very hard thing. So, but I'd say it's, you know, a quarter million a year is is a reasonable guess. and it it might be 500,000. I I don't know, but we'll we'll we'll make as many as people want and can afford and and and and but and then like I said, it's it's going to be hard to make the the cost of affordable because it is a new car and new manufacturing method.
1:32:04So, in in the grand scheme of things relative to the production rate of all the other cars we make, it will be small. but but still very cool. yeah. Thanks. Elon, maybe three more questions? Sure. Thank you. Good day, Mr. Musk. Hello, my name is Delia Archer, and this question aims to address Tesla's efforts in developing skilled workforce, attracting diverse talent, and promoting career opportunities within the company. When I look around, I see a lot of young faces, and I think about internships and apprenticeships for students in high school and beyond. My question is, are there any plans to expand existing partnerships and explore new avenues for
1:32:52collaborations with educational institutions and nonprofit organizations to attract the next generation in varied career pathways? yeah, actually we with with all of our gigafactories, we work with the the local schools. actually from the high school level, community college, and university.
1:33:15because it's incredibly important to foster the talent for the factories. as you saw from the the employment numbers, we actually need a lot of people. Even though we've got a lot of automation, we still need a lot of people to operate the factories. and really matters that they have the the right training. So, we're big believers in in reaching out to educational institutions and because frankly it's in our interest to do so. so, thank you for asking the question.
1:33:43All right. [applause] Elon, Josh Fuller. I had originally thought of asking about the the fun police and pushing back on the boombox and allowing us to screech at people with goats again, but after seeing Optimus, I was inspired to ask does has anybody asked Optimus' opinion of of Mars and does he have a ticket yet?
1:34:09well, you know, Optimus is not a deep thinker at this point. So, Optimus is still, you know, figuring out how to do basic stuff. Like like it it it couldn't make you know, cook some eggs or something quite yet. Yeah, it's so we need to get Optimus to the point where it has reasonable agility and can and can do basic things.
1:34:39And you know, and we're aiming for it to to start off doing simple tasks that are sort of boring and repetitive or or dangerous. basically jobs people don't want to do. So, that that's our that's our goal. and and I'm confident we'll we'll achieve that goal. and then we're going to figure out how to make it at scale, make sure that the robots the robot is safe.
1:35:04I think it's going to be very important to have a local means of turning it off. So, safety is going to safety is going to be extremely important. I can't emphasize that enough. But but right now it it is not at an intelligence level where it's pondering questions like Mars. but perhaps it will be one day. So.
1:35:31Hi, good afternoon, Elon. My name is Luke Arsenal. I just want to start off by saying big thank you to you and the team of Tesla for all you guys do. it's been great to see and I'm I'm so happy for you guys. My question is with the rise in cybersecurity threats to operational technology and Internet of Things, what steps is Tesla taking to protect the company itself and its products from these threats? Also, as a suggestion, cuz I know you like suggestions, for the navigation, do you think we could do something on the UI that adds in when you're about to take the off-ramp to show a picture of that exit, kind of like other map systems do?
1:36:05and then also Oh man, this is a lot of questions. One of my friends wanted to get a shaka from you, if that's possible. Okay. Well, we are we are constantly improving the navigation system. and being robust to hacking is incredibly important. in fact, we conduct a lot of sort of third-party hacking contests to try to get the the best hackers in the world to break into our cars.
1:36:36and actually no one has yet actually broken into a a Tesla in a way that would allow you to really control the car in a significant way, but they have gotten to where they can like honk the horn and mess with the infotainment system. and I'd like to thank them for their efforts.
1:36:53so we take information security extremely seriously. for navigation, yeah, there's there's definitely things we can improve there. but really down the road navigation visuals are not going to matter very much because the car is going to take you wherever you want to go. yeah, as for pictures, I I I have to I have to decline, unfortunately, because otherwise if I if I say I I hate being I actually hate being rude to people, so if I say yes to one person, then it's it's it's like selfies for a zillion you know, it gets kind of gets kind of crazy, so but thanks for asking. Maybe we'll take one one or two more questions, actually. So Sure.
1:37:44I don't know. Okay, what's your It's your Hello. I I had a question on the efficiency. I don't know if you've been paying attention at Terra, and I was curious if you thought that was a space that you could see Tesla getting into as far as the not just an efficient manufacturing process, but also the actual car itself and range.
1:38:12I Or just buy I'm not so I'm not so familiar with with with them, but we have a lot of fish we we have a lot of products to to get done. it's not like this is it's it's never a shortage of ideas. I I ideas frankly I I find ideas to be somewhat trivial. but the execution of the idea is extremely difficult. And as I said, like prototypes are easy, production is hard.
1:38:39Production at and cash flow positive is excruciating pain that you could like at a level you cannot believe. So, it's it's not product ideas, they're irrelevant. Tough. All right, so that's what I think of them. it it it's sort of like the idea of going to the moon. That's not who can That's That's irrelevant.
1:38:58[laughter] Getting to the moon is the hard part. So, okay, what So, maybe one last question, yeah. Elon, John Lopez from Orlando here. I've got two things for you real quick. the first one is with ride share being a growing thing, especially with Tesla owners, are we going to get a guest mode option available for that?
1:39:17the second thing is is as an X plaid owner, I would really love to get track mode. As as a What's that, sorry? X plaid. well, plaid has plaid has track mode. The Model X plaid does not, though. Oh, oh, Model X. yeah, we can probably add that, sure. yeah, no problem. So, ride share is a bunch of these things will will will really not be relevant in a self-driving world.
1:39:51You can summon the car and take you somewhere and if you want to add someone to that that ride, you can if you want to. but overwhelmingly it is about achieving self-driving full self-driving and you know, I think we're really getting to to the final stages of of full self-driving where I mean, I drove for several days around Austin just dropping pins in random locations and I had zero safety interventions.
1:40:21So, it's it's really and even in in city of San Francisco which is very complex, I haven't had a safety intervention in a long time. So, I think it's it's really it's getting to the point where you know, I admittedly I have been optimistic about this in the past. but I think I think this time
1:40:43[laughter] All right, I won't say it. All right, fair enough. All right. [laughter] [cheering] Okay, I think it'll happen, you know. [cheering] So, All right, you want one last question? Sure. [applause] So, first congratulations on your decision to try advertising. We all want Tesla to pass Apple in market cap.
1:41:22I didn't realize people wanted it that much. [laughter] Okay. It's good that a lot of people agree. So, earlier this year you cut prices on everything 15 20%. More recently you've been raising prices on Model Y, Model Model Model X. And can you just talk about the pricing strategy of Model Y going forward and keep it separate from the decision to bring out the next gen vehicles that we all agree is going to, you know, blow the doors off on volume, but just on Model Y, it's the best-selling product in the in the in the world.
1:41:52Can you just talk about strategy pricing going forward on that? Thank you. I mean, it's pretty straight forward. I mean, we see what the demand is and then we adjust pricing to meet the demand. now, the the the thing is that what what is you know, what happens with with other car makers is that they're actually constantly adjusting pricing on cars.
1:42:13It's just not obvious. The In the In the case of Tesla, you can see immediately when there's a price change that occurs because we don't have any intermediaries. We don't have any dealers. There's no There's no manufacturer's manufacturer's suggested retail price. So, but the reality is actually at at car dealers, the prices are changing radically. I mean, last year there were there were significant premiums above MSRP at at sort of the conventional car dealer, you know, like so, you'd pay above MSRP last year. and then this year, I think things are below MSRP or close to it.
1:42:53And then the the manufacturers will then also offer incentives. So, what what's actually happening is our our very big price shifts by other car makers. They're just not obvious. whereas with Tesla, it's obvious. So, it And this is this is necessary because demand fluctuates a lot. So, something's got to be done to achieve a supply demand clear clearing point for for volume.
1:43:26So, just people are reacting to a something that's obvious as opposed to saying, "Oh, yes, every car company does this all the time." Tesla's no exception. That's the actual reality. So. All right. So. Last question. I'll I'll be really quick. Okay. So, I will say I have been driving FSD.
1:43:52My wife drives FSD. There's a a term called the wife test. My wife will only drive our kids on FSD. It's incredible. So, that's that's that's that's a great I mean, the these sort of significant other test is a very a good test. because for there were for a while there, there was like, you know, friends or significant others would be like, "Please don't put it on." [laughter] It's too scary. but then I guess it's getting to the point where I mean, it's it's going to be just smooth as silk.
1:44:21So, in fact, you'll be able to visually tell, and perhaps even now you can, visually tell if the car is being driven on FSD or manually. Because if it's being driven on FSD, it's it's smooth smooth and precise. And if you look at actual if you if you actually closely observe cars going down the highway, the people are constantly moving in their lanes. So, the the cars are like doing this all the time when they're manually driven. If it's if it's actually on FSD, it's it's dead center in the lane.
1:44:50and and it it will actually get to the point where it is not merely and maybe it is at that point, not merely safer than a than a person, but actually way smoother than than manual driving. I I just I really I can't say enough about the profundity of self a full self-driving. It is It is one just one of the the biggest changes in history that will occur. It's It's not some feature.
1:45:20It's It's like a profound It's really as profound as electrification. and and and and and that we have already have millions of the cars on the road that will literally achieve this with a software update. I mean, that's like head exploding emoji, you know? and I think probably a lot of people in this room realize it, but the rest most of the rest of the world doesn't.
1:45:46yeah, so [applause] given how incredible FSD is and how much, you know, a marriage you're giving to potential future margins of the car, as well as how much an impact on society. And given that they didn't even want you to say if it's going to happen this year, just, you know, not to jinx it. Sure. Would it be possible to have just for some type of public-facing, investor-facing kind of timeline, so that way, you know, we you mentioned V12's coming and that's, you know, AI front end. Could we could we just see kind of a timeline on this is the version we're at, these are the things we're working on, and this is when we're, you know, ETA. So we can
1:46:28kind of see that in real time instead of a maybe Elon time? No, so so the the thing that's been yeah you know, really difficult with full self-driving is that it's it's it very often seems like it's about to to happen until you realize that you're actually at a local maximum and that you need to re-architect elements of the software to get out of the local maximum to to then a higher local maximum. So it's been this sort of series of logarithmic curves. and that's why it's been like
1:47:05it seemed like, oh, if you just extrapolate this on a straight line, it gets to self-driving except that it it's it's not a straight line, it's a logarithmic curve and it sort of and you and you start hitting these like local maximums where you'll sort of asymptote to a certain capability and then have to re-architect things. So now I I I think we're we're we're finally at a high enough local maximum or will be this year where to to point where it is, I think, probably safer than than a person, meaning the probability of injury on average will be better than than a person even if the even if someone pays no attention to to the car. It It is worth saying that
1:47:44that right now for human-supervised Tesla's full self-driving, it is dramatically safer than manual driving. It's like I I think like four times safer, maybe more. so, it is already a massive safety improvement to have human-supervised FSD right now. and I think we will we will get to the point where we're we're always going to be in some amount of local maximum, but I think that we'll be in a local maximum that exceeds human human-level safety this year.
1:48:23yeah, but I guess we'll see. So, anyway, I I want to thank you all for for coming and [applause and cheering] so, [cheering] [applause] [cheering] so, [applause] just want to say like you know, your your support is super appreciated. and you know, the as they say, a friend in need is a a friend indeed. And supporting Tesla when the chips are down, that that those are the real friends and and you're them, so thank you.
1:49:10[applause] All right. Okay. [cheering]
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