萨。

萨姆。

欢迎来到得克萨斯的家。

Tesla 的家。

我叫布兰登·埃尔哈特。我是总法律顾问兼公司秘书。本次会议将分为两个部分。首先，我们将审议股东受邀投票表决的12个事项，以及任何其他已适当提交的事项。其次，我们的创始人兼首席执行官埃隆·马斯克将进行演讲。

在接下来的会议过程中，我们可能会讨论我们的业务前景，并作出其他前瞻性陈述。此类陈述是基于当前预期作出的预测。由于多种风险和不确定性，包括我们最近向美国证券交易委员会提交的10-K或10-Q表格中披露的风险和不确定性，实际事件或结果可能存在重大差异。此类前瞻性陈述代表我们截至今天的观点，此后不应不应予以依赖，并且我们不承担在今天之后对其进行更新的任何义务。

此时，我要感谢到场的Tesla团队成员和董事会成员。Tesla的独立审计师普华永道的一名代表也在场。但在开始之前，我想向大家介绍我们的董事会主席罗宾·德诺姆，她想讲几句话。

谢谢你，布兰登。大家好。作为Tessa董事会主席，我很荣幸欢迎各位来到我们的家园得克萨斯州，参加我们的年度参加我们的年度2024年度股东大会。当我说很荣幸欢迎各位股东时，我确实是这样想的。世界上没有其他公司像Tesla一样拥有如此多元、积极参与且给予支持的股东群体。没有其他股东群体像你们一样了解自己的公司，也没有哪个股东群体像你们一样致力于Tesla加速世界向可持续能源转型的使命。

这是董事会自豪地代表其利益的股东群体。这也是一个明白如下道理的股东群体：为了加速世界向可持续能源和可持续能源经济转型，Tesla需要开发最具革命性的技术。不仅是在汽车领域，还包括能源和人工智能领域。而且，我们过去一年的成就是实现我们使命的步骤。

一些亮点包括Model Y在2023年成为全球最畅销的车辆。

开始交付备受期待的Cybertruck，证明皮卡车卡车的力量和实用性也可以是电动且酷炫的。我们的储能部署量在2023年创下历史新高，这一部署趋势在2024年继续延续。通过使用我们的产品，Tesla客户在2023年避免排放了超过2000万公吨的温室气体。这相当于内燃机车辆行驶510亿英里所产生的排放。

自2022年以来，避免的内燃机车辆行驶里程增加了约180亿英里。但要让温室气体减排量继续呈指数级增长，我们不仅必须让尽可能多的客户用上我们的产品，还需要提高我们产品的利用率。为了进一步推动这一目标，我们发布了受监督的FSD，证明我们的自动驾驶方案具有可扩展性，也证明人工智能领域的成就，以及利用我们现有的生产线让下一代产品更快上市，对于实现我们的使命至关重要。

我们已经并计划在硬件和软件生态系统方面进行进一步投资，以进一步推进我们的自动驾驶之路，并为实现Robo Taxis确定方向。但可持续发展不仅仅是避免排放。它需要考虑从产品制造到产品寿命终结的整个生命周期。这就是我们负责任地采购，并设计制造流程、使每辆车的用水量低于行业平均水平的原因。

而且，我们连续第3年使用100%可再生能源为超级充电网络供电。

我们也在内部并与其他方合作建立电池材料回收生态系统，以确保我们正朝着真正的储能循环经济迈进。仅在2023年，我们就支持回收了足以制造43,000辆全新后轮驱动Model Y的电池材料。这一成果是可持续经济的典范。虽然可持续发展是我们的使命，但我们也以安全为目标。

我们的汽车经过精心设计，是世界上最安全的汽车之一；我们利用车队的真实世界数据，以及推送无线更新的能力，为我们所有车型优化安全性。我们对自动驾驶的重视也有助于实现我们的安全目标。根据2023年的数据，启用驾驶辅助技术的Tesla车辆每行驶564万英里仅发生1起车辆事故，相较于美国。

我认为这值得澄清。

相比之下，美国车队每行驶670,000英里发生1起事故。所以，驾驶Tesla不仅是在保护环境，也是在保护您自己以及与您同行的人。在推进使命方面又取得了1年的进展之后，我想借此机会感谢我们卓越的Tesla员工，他们的创造力、辛勤工作和奉献精神是我们成功的源泉。

非常感谢，而且由于大多数、大多数人也和你们每一个人一样是股东，我希望每位员工都为自己在Tesla成就中发挥的作用感到自豪。我为你们和董事会感到骄傲，我非常感激你们。最后，我想提醒我们的股东，你们的董事会正在倾听，我们在这里代表你们的利益。你们是。是的。

你们是公司的所有者，你们的声音、观点和意见对我们董事会以及公司的发展方向至关重要。我和其他董事是你们的受托人，并监督Tesla的使命、宗旨和战略。铭记这些责任，我们不断为道德行为、企业公民责任和公司治理设定最高标准。你们看不到董事们日常的工作，但我可以向你们保证，这是一届高度投入的董事会，我要感谢其他董事的奉献。

无论是在董事会会议室内、与我们的股东会面，还是与公司各处的团队成员密切合作。Tesla是独一无二的。谢谢。

Tesla的独特之处在于，我们的散户股东群体既活跃又庞大。我们希望今天在这个会场的各位以及在线参加会议的股东，继续充分利用我们的股东平台，加强你们与Tesla的互动。我要特别感谢我们规模不大但实力强劲的投资者关系团队和法务团队，以及参与开展今年代理权征集活动的每一个人。无论是Tesla内部还是外部的人。

仍有许多工作要做，但这让我们更加团结，坚信正确的事情永远值得为之奋斗。谢谢。现在我把时间交回去。谢谢。

现在我要把时间交还给布兰登。谢谢。

谢谢你，罗宾。现在我宣布会议开始。目前时间是美国中部时间下午3:40，我宣布投票现已开始。在过去几周里，我们已经收到了股东提交的投票委托书。不过，如果你们此时希望对自己的股份进行投票，可以通过虚拟会议门户进行投票，或者如果你们想亲自投票，前方备有选票和投票箱。Tesla董事会已任命Cardidio Group的代表托尼·卡迪迪奥担任选举监票人。

此外。此外，Broadridge Financial Services已证实，自2024年4月29日起，代理材料或代理材料互联网可用通知已邮寄或提供给截至2024年4月15日登记在册的所有Tesla股东。出席会议的股份占已发行股份的多数。因此，我宣布出席股份达到法定人数，我们可以继续举行会议。议程项目如下。

共有5项Tesla提案。选举2名第二类董事，即詹姆斯·默多克和和金巴尔·马斯克，任期3年，或直至他们各自的继任者正式当选并取得任职资格；在非咨询基础上批准高管薪酬。批准通过转换将Tesla的注册地从特拉华州迁至得克萨斯州。

批准此前已由股东于2018年批准、授予埃隆·马斯克的100%基于绩效的股票期权奖励。

批准任命PricewaterhouseCoopers LLP为Tesla 2024财年的独立注册公共会计师事务所。Tesla董事会已建议股东投票支持2名董事候选人中的每一位，并并支持其他4项Tesla提案中的每一项。我们还收到了代理声明中所述的7项股东提案。我要提醒各位，Tesla董事会已拟定一份反对这些提案的声明，该声明也载于我们的代理声明中。

那么，让我们转到那 7 项股东提案。第一项是关于将董事任期缩短至 1 年的咨询性投票。我们的董事会建议股东对此投反对票。正如委托书这份委托书中所解释的，这项股东提案由詹姆斯·麦克里奇先生提出，他已经准备了一段预先录制的发言。

谢谢。我们在 2021 年提出过一项类似提案。尽管董事会提出了一项相互竞争的提案，要求每隔 1 年而不是每年进行问责，仍有 55% 的人支持该提案。两项提案获得的支持都没有达到所持股份要求的 67%。标普 500 指数成分公司中有 90% 采用不分级董事会。大多数机构股东将再次支持我们的提案。请加入他们。最近的特拉华州法院案件指出了 Tesla 的 3 项法律缺陷。

董事并不独立，向投资者陈述的薪酬计划事实具有误导性，而且董事会未能进行适当谈判。补救措施的一个主要部分是取消董事会分级制，以便股东能够追究董事责任，或提交自己的提名。Free Float 的马特·莫斯卡迪称第 6 号提案是今天最重要的一票，因为它可能让投资者得以改变董事会，引入新的技能和专业精神。

约翰·查维丁的简单多数票是另一项至关重要的投票。另一位重要的支持者是内尔·米诺，她是包括代理顾问公司 ISS 在内的多家公司的联合创始人。马斯克溢价已经在消退。Tesla 是马斯克的流动存钱罐，因为它是上市公司。他的其他公司不是。要么他留得足够久，利用我们的股东资本为他的其他企业提供资金；要么如果我们否决他的薪酬方案并切断资金，他就会更早转移注意力。

轻点。硬着头皮接受吧。为一个不依赖超级英雄独裁统治的未来做好准备。我们需要从一个由亲朋好友主导的董事会，过渡到一个由工程、软件及其他相关领域的独立专家董事组成的董事会。马斯克是一位天才。看到他把注意力集中到别处，我很难过，但这正在发生。不要让他从 Tesla 抽走更多资源。我们需要过渡到一个更加民主的 Tesla，使其能够吸引那些追求未来更美好世界的客户、员工和股东。

再次强调，请投票支持第 6 号和第 7 号提案。取消董事会分级制，以实现问责。降低投票门槛，以获得更大的灵活性。请在投票结束前立即投票。你的一票可以对我们的未来产生重大影响。谢谢。

谢谢。谢谢你，麦克里奇先生。提醒一下，董事会的回应已载于我们的委托书中。第2项股东提案是一项关于我们治理文件中简单多数条款的咨询性投票。我们的董事会已建议反对，而理由并未载于委托书中。这项股东提案由约翰·奇维丁先生提出，他目前在线上介绍这项提案。先生。

查维廷，我想邀请你发言。你有3分钟。

3分钟。

大家好，我是约翰·查维丁。第7号提案，简单多数票。查尔斯请求董事会采取必要措施，将公司章程和细则中要求超过文明多数票的每一项投票要求，替换为要求对这类简单多数提案所投赞成票和反对票的多数。这包括作出必要的修改。用通俗的英语来说，这项第7号提案与 Tesla 董事会在2022年 Tesla 年会上支持的概念完全相同。

随后，2022年 Tesla 董事会的提案获得了参与投票的 Tesla 股东98%的支持。而如果董事会当时只稍微努力一下，鼓励更多 Tesla 股东投票支持他们的提案，董事会版本的这项提案早就会获得通过。这项简单多数票标准提案将有助于改善 Tesla 股东权利。Tesla 目前在股东权利方面得到糟糕的10分，而10分是可能的最差分数。

Tesla 董事会在2022年支持这一提案概念，今天也应该这样做。当 Tesla 有一位首席执行官埃隆·马斯克，而他正威胁要减少对 Tesla 的关注，并威胁除非获得560亿美元的薪酬，否则将不向 Tesla 提供先进技术时，肩膀权利和文明多数票标准就更加重要了。此前从未有分享者在被要求证明一笔骇人听闻的560亿美元首席执行官薪酬时承受如此大的压力。

此外，这笔560亿美元的薪酬并未承诺马斯克先生会将任何最低比例的时间投入 Tesla，也未承诺追求任何最低数量的先进技术。在特雷斯拉，投资公司 Gerber Kawasaki 的拉斯·格伯表示。我认为，Tesla 董事会是商业史上利益冲突最严重、独立性最低、能力最不足的董事会。请投票。赞成。简单多数票。

第7号提案。

谢谢你，查瓦丁先生。提醒一下，我们的董事会回应已载于委托书中。我们的第3项股东提案是一项关于每年报告反骚扰和反歧视工作的咨询性投票。出于委托书中所列的理由，我们的董事会已建议反对。这项股东提案由主要提交方纽约州共同退休基金提出，该基金已请求由代表 Neo Impact Capital 的克里斯汀·霍尔博士作为这项提案的共同提交方介绍这项提案。

赫尔博士，我想邀请你发言。你有3分钟。

谢谢。大家好，我是 NEA Impact Capital 的克里斯汀·赫尔博士。我在此代表主要提交方、纽约州审计长托马斯·D·保利，他是纽约州退休基金的受托人；截至今年3月，该基金持有的 Tesla 股票价值超过6.5亿。我也以第8号提案共同提交方的身份来到这里。该基金是 Tesla 的重要长期持有者，并明确专注于投资气候解决方案。Longview Large Cap Fund 通过其受托人 Amalgamated Bank，与 AP7 和 Arjuna 一同参与提交了这项决议。

我们公司的名称 NIA 意味着意图和目的。我们只投资于那些我们认为正在建设一个更美好世界、一个可持续且包容的未来的公司。作为 nia 的创始人，我曾为我们早期持有 Tesla 而感到自豪，我看到了汽车和清洁能源市场领域的潜力及创新需求。然而，当我们查看前备箱下面时，却发现了职场管理方面的失败，而这是任何企业最重要的组成部分之一。

员工需要被招聘、培训、激励并投入工作，以完成高质量的工作，而我们需要这些员工能够发挥他们的最高能力。要做到这一点，他们必须在工作中感到人身和情绪上都安全。投资者正在要求第8号提案中概述的报告。因为职场虐待、骚扰和歧视在 Tesla 没有容身之地。它们可能造成巨额成本，包括员工流失、缺勤率上升和生产率下降，以及法律费用，所有这些都可能影响股东对公司的投资。

它们还可能导致招聘高素质员工变得困难，而 Tesla 需要这些员工来保持创新，并处于快速发展的技术格局前沿。多年来，我们一直对 Tesla 的职场表达担忧；年复一年，更多骚扰和歧视指控不断出现，而 Tesla 的人员管理政策或实践几乎没有可见的改善。为了缓解我们的担忧。

Tesla 目前面临的诉讼和集体诉讼涉及数千名员工，指控包括经理和员工反复使用种族侮辱性称谓、在工作分配、薪酬和晋升方面隔离并歧视非白人员工，以及在员工报告自身经历时进行报复。正是由于这些担忧，我现在正式提出第8号提案，请求 Tesla 董事会委托编制一份独立报告，评估公司在这个革命性、创新性的工作场所中防止骚扰和歧视的工作成效。

谢谢。

谢谢你，霍尔博士。董事会的回应已载于委托书材料中。

让我们转向第4项股东提案，这是一项关于采用结社自由和集体谈判政策的咨询性投票。出于委托书中所列的理由，我们的董事会已建议反对。这项股东提案由 SOC Investment Group 提出，其代表特泽尔·帕特尔已准备了一段预先录制的发言。

大家好，我是 SOC 投资集团执行董事特贾尔·帕特尔，我在此提出第 9 号提案，这是一项股东提案，要求 Tesla 董事会采纳并披露一项不干涉和集体谈判政策，在其运营中维护国际劳工权利标准。本决议由其他共同提交者一同提交，我们担忧 Tesla 的现行政策未能清楚说明国际劳工法与当地劳工法之间的冲突将如何解决。

而且有明确迹象表明，尽管其中涉及运营、法律、人力资本管理和声誉风险，Tesla 仍未遵守其现行政策。我们注意到，Tesla 于 2023 年 4 月修订了其全球人权政策，明确提及工人自由结社和集体谈判的权利。但这些修订受到了致命损害。承诺尊重这些权利的声明包含“遵循当地法律”这一限定语，而非援引往往标准更高的国际人权标准，这表明 Tesla 仅在当地法律明确要求时才尊重这些权利。

许多迹象表明，Tesla没有尊重其员工的基本权利。在瑞典，Tesla因被描述为拒绝谈判的行为，面临了长达数月的罢工和抵制。此外，路透社在有关这一情况的报道中写道，Tesla有一项不签署集体协议的政策，如果属实，这将直接违反公司现行政策。在德国，Tesla面临拒绝谈判和工作条件恶劣的指控。

在美国，针对劳动权利侵犯问题，Tesla已收到全国劳资关系委员会全体成员和美国第五巡回上诉法院作出的不利裁决。就在上个月，全国劳资关系委员会发出了一份申诉书，指控Tesla工作场所技术政策中的限制措施非法阻碍员工加入工会或采取共同行动。

股东还应考虑Tesla面临的更广泛人力资本管理风险，包括有报道称，其健康与安全违规事件和受伤率高于同行，以及多起指控存在敌意工作环境的诉讼。结社自由和集体谈判被视为赋能性权利，这意味着它们使员工能够有效争取更多劳动权利，包括在不受歧视的安全健康工作场所工作的权利。

采纳一项全面的劳动权利政策，明确阐述对不干涉、诚信集体谈判以及遵守更高国际标准的承诺，将有助于Tesla更好地降低这些人力资本管理风险，以及我们在瑞典看到的这类运营和声誉风险。因此，我们敦促各位投票支持第9项议案。谢谢。

谢谢您，帕特尔女士。董事会的回应载于委托书。第5项股东提案是就报告电磁辐射和无线技术相关影响与风险进行咨询性投票。基于委托书中所述的理由，我们的董事会建议投反对票。这项股东提案由兰德罗·珀塞尔提出，她有代表出席。黛博拉·戴维斯博士准备了一段预先录制的信息。

我是黛博拉·戴维斯，环境健康信托基金的创始人。今天，我代表股东兰德里·珀塞尔就她提出的提案向各位发言，该提案寻求每年就其产品中的磁场和无线辐射风险发布报告。我也代表数百万幼童、婴儿、学步儿童和孕前女性发言；据美国儿科学会称，这些人更容易受到无线辐射的影响。

我是美国国家科学院环境研究与毒理学委员会的创始主任，并已发表超过220篇有关环境健康的科学文章。我与其他数百名专家共同得出结论：目前暴露于被称为EMF的非电离电磁场构成一种严重且未获认识的公共健康风险。Tesla车辆使用一个1,200磅、会发出磁场的电池，还使用Wi-Fi和雷达等会发出无线辐射的无线设备。

Tesla应提供有关实测暴露水平以及降低暴露所做努力的透明信息。世界卫生组织已得出结论，低水平EMF以及微波或蜂窝辐射都是，都是，引用原话，“可能的致癌物”，与某些杀虫剂和汽油属于同一类别。提供再保险的公司，如伦敦劳合社和瑞士再保险，长期以来一直认为电磁场属于不可承保风险，因其潜在的健康和环境损害而可与石棉相提并论。

Tesla股东有权知道公司如何就此类风险为自身投保。目前的水平可能合法，但这并不意味着它们是安全的。事实上，美国的法规是在超过四分之一个世纪前制定的。美国哥伦比亚特区巡回上诉法院认定，FCC未能考虑所有科学证据且拒绝更新无线辐射限值是，引用原话，“武断且反复无常的”，至今已有3年。Tesla未能向消费者告知汽车内部或或其其他产品中的磁场和无线辐射实测水平。

根据法律规定，婴儿和学步儿童必须被固定在座椅上，以便在碰撞期间保护他们，但他们对于能够更深入穿透其较薄颅骨的EMF却没有任何防护。为降低Tesla的风险，公司可以对其车辆进行屏蔽，并确保无线天线可以轻松关闭。硬件和软件变更可以降低暴露水平。总体而言，对于某些有毒化学品，Tesla已制定超出监管要求的极其严格的合规与危害筛查。

投资者可以鼓励公司在降低EMF方面发挥类似的领导作用。公司是时候在其产品发出的无线辐射和非电离辐射安全方面明确无误地展开竞争了。谢谢。

谢谢您，戴维斯博士。董事会的回应载于委托书。接下来是第6项股东提案，即就采纳目标并报告相关指标，以评估将可持续发展指标纳入高级管理人员薪酬计划的可行性进行咨询性投票。基于委托书中所述的理由，我们的董事会建议投反对票。这项股东提案由Tulip Share Securities Ltd.提出。

其代表康斯坦特·里基茨准备了一段预先录制的信息。

各位股东，下午好。感谢各位抽出时间。我是康斯坦丝·里基茨，今天我代表Tulipshare发言，支持第11号提案，敦促Tesla将可持续发展指标，包括高级管理人员的多元化程度，纳入高管薪酬计划。这项提案要求制定有力的量化目标并进行透明报告，从而既解决高管薪酬计划争议，又确保我们公司的长期稳定与成功。

领导层的多元化和独立性对于创新和增强决策能力至关重要。而高管薪酬中的可持续发展指标反映了与监管压力、投资者预期和负责任商业实践保持一致。尽管75%的标普500指数成分公司已将可持续发展指标纳入高管薪酬计划，Tesla却没有这样做。鉴于我们的使命是加速世界向可再生能源转型，这一点尤其令人担忧。

近期撤销我们首席执行官560亿美元薪酬方案的裁决，凸显了制定公平且经过协商、最符合股东利益的薪酬计划之必要性。Tesla的委托书并未提供实质性证据，说明被撤销的2018年奖励如何损害了股东民主，或批准该奖励将如何纠正先前的缺陷并带来不同结果。同样，对这项提案的反对意见也未能提供任何证据，证明现有高管薪酬结构中已纳入尊重既定人权、尽职调查流程和原则的绩效组成部分。

相反，它援引了其与高管薪酬完全无关的人权政策，并进一步坚持这一论点：目前存在争议的高管薪酬结构是激励实现我们公司使命的唯一手段。这种对现状的坚持，实质上无视了Tesla必须面对并克服的持续争议，包括销售疲软、竞争加剧、领导层精力分散、股价下跌30%、市值减少、大规模裁员、侵犯人权以及持续的法律挑战。

Tesla 的高管理应因创新且成功的业务获得适当报酬。然而，股东民主制度允许我们反对诸如要求获得公司 25% 的控制权，以及通过进一步发行股票来获得美国企业史上规模最大的薪酬方案。我们也有权投票决定，使领导层的薪酬与财务及可持续发展绩效相称。这并不意味着我们不尊重领导层付出的时间、精力和贡献。

恰恰相反，因为确保我们公司的长期生存能力与成功是我们作为股东的职责。

谢谢你，里基茨女士。董事会的回应载于委托书。让我们进入第 7 项股东提案，这是一项关于承诺暂停采购来自深海采矿的矿产的咨询性投票。基于委托书中所列的理由，我们的董事会建议投反对票。该提案由 As You Sow 提出，其代表伊丽莎白·利维也录制了一段预先录制的信息。

下午好，董事长、各位董事会成员，我叫伊丽莎白·利维，代表 As You Sow 就第 12 号提案发言。我们对火星表面的了解，比对海底的了解还要多。即便如此，一个新兴行业正在推进对深海的开发。我们认为，这一行业现在以及未来都会给 Tesla 带来风险。深海采矿是指疏浚海床以提取电池相关矿物的过程。

它将不加区分地摧毁栖息地，并消灭沿途的生命。海床采矿可能释放储存的碳和有毒沉积物羽流，毒害海洋食物链，并扰乱海洋生物的繁殖和迁徙。深海环境复杂，生命丰富。其生态系统生长缓慢且十分脆弱，几乎没有或完全没有能力从外部干扰中恢复。无论是否受到监管，启动商业规模深海采矿流程的计划都在推进。

Tesla 加入经合组织（OECD）和负责任采矿保证倡议（IRMA）的采矿协议并不能解决这一问题，因为两者目前都不适用于深海采矿。超过 800 名海洋科学与政策专家、25 个国家、72 个原住民团体和 48 家公司正在呼吁禁止、暂停或预防性暂缓深海采矿。其中包括 Tesla 的竞争对手宝马、雷诺、Rivian、大众和沃尔沃。就连通用汽车也已表示，其不打算使用深海开采的矿物。

我们公司对深海采矿的公开立场至关重要。Tesla 是电动汽车转型的代表，而其对深海采矿含糊不清的立场可能会推动资金流入这个备受争议的行业。反之，明确表态限制 Tesla 使用深海矿物，则可以推动电池矿物技术、回收利用和循环性的创新。借助这些技术，即使是未来最雄心勃勃的 EVU 情景，也可以在不损害深海的情况下实现。

Tesla 未能就深海采矿表明立场，再加上一名前 Tesla 董事会董事如今担任金属公司的董事会成员兼首席执行官特别顾问，而这家公司正是推动深海采矿的领头企业，这引发了担忧。这种关系似乎暗示 Tesla 支持金属公司的深海采矿活动。承诺暂停将明确预期并减少对公司的声誉损害，向客户和投资者保证，Tesla 确实会等到科学表明深海采矿不会造成重大或不可弥补的损害之后再行动。

谢谢。

谢谢您，利维女士。董事会的回应载于委托书中。请注意，现在是提交任何选票并使其计入统计的最后机会。

我宣布，投票现已结束。

根据我们收到的委托投票以及本次会议提交的投票，我想先行宣布初步结果。关于 7 项股东提案，我们的股东批准了 Tesla 董事会对所有提案的建议，但有关董事任期的第 6 号提案和有关简单多数投票制的第 7 号提案除外。

现在转到 Tesla 的 5 项提案。

第一，我们的股东批准选举 2 名第二类董事詹姆斯·默多克和金巴尔·马斯克，任期为 3 年，直至各自的继任者正式当选并具备任职资格。

第二，我们的股东以不具约束力的咨询方式批准了高管薪酬。

第三，我们的股东批准追认任命普华永道会计师事务所为 Tesla 2024 财年的独立注册公共会计师事务所。

第四，我们的股东批准将 Tesla 的注册地从特拉华州迁至得克萨斯州。

第五。

最后，我们的股东批准追认 2018 年经股东批准授予埃隆·马斯克的 100% 基于业绩的股票期权奖励。

会议现已闭幕。接下来，Tesla 的股东们，让我们欢迎埃隆·马斯克。

目标是让人们看到希望，相信存在一条通往完全可持续的全球经济的道路。相信我们正走在这条道路上，正加快沿着这条道路前进，而且只要我们不对此自满，它就会实现。

这确实是……的开端

Tesla 泡沫的终结。事实上，我认为这家公司可能会破产。

Tesla 的 Model Y 是全球最畅销的汽车，超过了丰田的 RAV4 及其卡罗拉车型。

关于 FSD 第 12 版，它意义深远。改进速度很快。它可能会带来史上最大规模的资产价值增值。当你能够实现无需监督的完全自动驾驶时，这听起来就像是故事股、自动驾驶出租车。我的意思是，你真的能靠这种，你知道，有点不切实际的预测来平衡收支吗？如果你问错了问题，就不可能得到正确答案。

我的预测是，Tesla 长期价值的大部分将来自 Optimus。而且我对这个预测非常有信心。

很少会有一种产品问世，它看似不可能实现，专家们也曾说它永远不可能被制造出来。而这正是其中一次。未来终于会看起来像未来。人们会问，比如，你为什么要把它做成防弹的？我会说，为什么不呢？

储能部署量，尤其是 Megapack，在第一季度创下历史新高，推动能源业务实现创纪录的盈利能力。这是制造机器的机器，而工厂就是产品。这座建筑是地球上迄今所见最先进的汽车工厂。

如果你把 Tesla 当作一家汽车公司来估值，那这个框架就是错的。如果有人不相信 Tesla 将解决自动驾驶问题，我认为他们就不应该成为这家公司的投资者。但我们会做到，而且我们正在做到。

萨姆。

嘿，各位，

欢迎，欢迎参加 Tesla 股东大会。

而我，我只想首先说一句，真他妈的，我爱你们。

是的，我们拥有最棒的股东群体。我的意思是，这简直不可思议。任何上市公司，都不可思议。哇。

所以，你知道，我们在这里召开了一场很棒的股东大会，来回顾公司的成就，并告诉大家我们要去往何方。而且，我觉得这不可思议。我认为我们不只是在为 Tesla 开启新的篇章，而是在开启一本新书。好了。

所以，那些一直密切关注 Tesla 的人明白，就是说，你们懂的，你知道，所以，所以，重要的是你们懂了。我的意思是，这个，所以，是的，我认为事情的发展方向绝对会令人震撼。

你知道，我们显然正在解决可持续能源问题上取得巨大进展，你知道，通过电动汽车、固定式储能、太阳能，然后我认为，就公司的价值而言，自动驾驶实在是件令人震撼的事情，就像我说的，你们明白，但我认为世界上大多数人还不明白。所以。

而且，你知道，在和许多那种大型机构投资者交谈时，他们往往在纽约之类的地方，而且不开车，所以我会问，你开过我们的车吗？你试过自动驾驶吗，你知道，12.3 版本？然后他们会说，哦，没有。好吧，那你应该试试。那会是一件值得做的事。而如果你只是把自动驾驶进展得有多好的各个点标在曲线上，并且就相信这条曲线，它正在以指数级的速度非常迅速地迈向无需监督的完全自动驾驶。

如果你只是，如果你只是。是的，这个，这个，我的意思是，是的，所以。

这，这，这，这真的就这么简单。事实上，我会请你亲自去做。就说，好吧，每发布一个版本，在你不得不进行干预之前，你能行驶多少英里？

真的就是那么，那么简单。而且每发布一个版本，你都会看到，会有一个，会有一个巨大的进步。巨大的进步。而且看起来是指数级的。所以。而且很明显，它实际上会达到一个比人驾驶汽车安全得多的程度。

所以。

而且我对，你知道，接下来的版本有一些了解，因为基本上对于每一个已经上路的版本，我们都能看到下一代版本是什么样，也能对再下一个版本有所了解。所以我们对它的发展方向有一些了解。而且确实令人惊叹。

毫无疑问，它的安全性将远远超过人类驾驶。

是的，毫无疑问。

而且我想，你知道，我提到过这一点，但有时候，重申一下并把一些点串联起来会有所帮助，因为，所以，因为有时候人们会想，好吧，你要怎样从这个庞大的车队，走到在自动驾驶的情况下真正将车队商业化？而实际上，它在某种程度上是 Airbnb 和 Uber 的结合。所以，会有一些由 Tesla 自己拥有的汽车。这有点像 Uber 的模式。

但对于由我们的客户拥有的车队，它会像 Airbnb 的模式。你随时都可以把自己的车加入车队或从车队中移出。所以你可以说，我要离开一周。只需在你的 Tesla 应用程序上点一下，你的车就会被加入车队，并且会在你离开期间为你赚钱。是的。

所以你可以说，你可以把它加入车队几个小时、几天或几周。无论你什么时候想要它回来，你就说，回来，然后车就会马上回来。

而且我非常有信心，它会远远超过这个价值，比如车主获得的收入将远远超过实际的月供。然后Tesla显然会从中抽取收入分成，但大部分钱会归车主所有。而且这真的会实现。这就是将会发生的事。所以你们，我是说，记住我的话，这，这仅仅是时间问题。当然，我承认，我有时有点乐观。

你知道，我，我，你知道，所以我并不是完全没有自知之明，但如果我不乐观，这一切就不会存在。这座工厂就不会存在，你知道，所以，所以，你知道，我哥哥就是这么说的，他很棒，他也在这里，刚刚再次当选。不，抱歉。这个东西的。

所以讲个小趣事

当我们

还是上学的孩子时。我哥哥会告诉我错误的公交车时间，这样我就不会错过公交车。

然后我会对他很不高兴，比如，你为什么告诉我公交车会比实际时间更早到？然后他说，嗯，因为否则你就赶不上公交车了。这是真实发生过的事。所以我想，我从，你知道，从出生起就有点病态地乐观，但不管怎样，这就是所有这些事情正在发生的原因。就像我必须在某种程度上病态地乐观，但我最终确实会兑现。

这才是重要的事。

所以。

是的，而且它真的会非常不同凡响。

我以前提到过这个，但，比如这个。关于Ark Invest，我建议阅读Ark Invest的分析。那是最准确的，可能有，你知道，更准确的分析，但就自动驾驶潜力而言，我所知道的最准确的那种分析，是，是Cathie Wood的Ark Invest所做的。所以，如果我没记错，他们，他们预测，仅仅基于车辆自动驾驶，不计入Optimus，Tesla的估值会超过5万亿美元。

是的，是的，我同意这一点。

所以，所以我认为，我认为，我认为，仅仅基于车辆自动驾驶，我们就能，就能让公司的价值增长10倍，超过10倍。

我相信这就是将会发生的事。是的，不，只是，你知道吗，现在，现在是下午4:20。刚注意到。

所以，是的，所以我认为那实际上，我认为那相当准确。

但还有这个。实际上，当你想到Optimus机器人时，事情会变得疯狂得多，它其实是一种人形机器人，旨在，你知道，能够做你想让它做的任何事，成为，你知道，你的伙伴。它可以待在你家里，可以算是帮你照看孩子，可以教他们，很棒的老师，你知道，它也可以做工厂里的工作。

比如我认为，最终的比例，比如说，你想要多少个超级有用的人形助手机器人？比如，谁不想要一个C3PO呢，你知道，你知道，但它是C3PO加R2D2，再加，你知道，再加再加

它

会非常棒，我认为，我认为世界上每个人都会想要一个，就真的是每个人。

而且，而且显然还会有机器人在工业领域制造东西。所以，我是说，我认为人形机器人与人类的比例可能至少会是2比1，差不多这样。肯定会是1比1。所以这意味着，大约会有100亿个人形机器人，也许，也许20或30。然后就是，好吧，假设，你知道，你大概制造，假设制造速度是，我认为制造速度最终可能会达到每年大约10亿个。

人形机器人，是真的。

而如果Tesla对此只有10%的把握，而且可能远远超过10%，还有，你知道，谁每年制造大约1亿个乐观主义者单位。我只是，我是说，作为参考，汽车行业每年大约生产1亿辆汽车。所以，你知道，大致处于相近的范围，至少在同一个数量级内。而且我认为，在规模非常大的情况下，我们可以把一台的成本做到大约10,000美元。它更小，会比汽车便宜。

所以，我是说，我认为，如果你以20,000美元左右的价格出售，这是在大规模产量下，Tesla基本上每年可以从中赚取大约1万亿美元的利润。

所以，是的，如果市盈率倍数是，比如，我不知道，20或25，差不多这样，那就意味着仅Optimus就能带来20万亿美元的市值，而且，而且自动驾驶汽车可能能带来5到10。所以，比如，我认为Tesla实现相当于当今市值最高公司10倍的估值，实际上是可以想象的，它在可能性范围之内。

所以当我说这就像我们正在开始一本新书时，我的意思是，它会成为最好的一本书。

下一个层次。下一个。

下一个。

下一个。下一个层次。

我们需要确保这些机器人对我们友善，而且，你知道，这非常重要。

所以，不管怎样，接下来让我进入演示环节。

吧啦，吧啦，吧啦，吧啦，吧啦，吧啦，吧啦，吧啦。

哦，天哪。

我认为，我们或许值得考虑一下，给未来的 Tesla 股东提案设定一个字数限制。

就像是，不能写成小说那么长。没错。我们不能把一整部小说当成 Tesla 提案。所以你们可以看到，我们的影响正在加速。我们开始真正显著地减少碳排放。所以，我们确实，你知道，在实现可持续发展目标方面取得了很大进展。而自动驾驶带来的一个影响，实际上会对碳排放产生更大的作用，因为我猜乘用车会从每周使用大约 10 小时，也就是每天大约 1 个半小时，增加到可能每周总时数的 3 分之 1，也许是 50、55 小时，这意味着同一辆车的使用频率会达到原来的 5 倍。

所以制造汽车需要一定数量的资源，但它的使用量会达到 5 倍。在我看来，我认为这很有可能，而这会对碳排放产生更加显著的影响。所以，你知道，我们非常重视可持续制造。我们的工厂很漂亮。比如你走在工厂里，实际上，你知道，与这些，比如你在媒体上读到的内容相反，工厂里的氛围很好，人们面带笑容、心情愉快。

就像，这挺不错的。

所以，所以，是的，所以，我是说，我们确实在乎。我们实际上非常在乎。这关乎做正确的事，而我们，你知道，我们不会做到完美，但我们非常在乎做正确的事。

所以我们的车辆节水、节能。

你知道，我们尽可能尝试使用可再生能源为工厂供能，而且，你知道，尽最大努力做正确的事。

我们的电池寿命延长了很多。反正我手里的东西太多了。电池寿命延长了很多，这就是。这样很好。

工厂里的安全状况良好。

我们正在提高电动汽车的可负担性。你知道，Model Y 可与 BMW X3 相媲美，但按月计算，它的实际拥有成本要低得多。

去年，我们的电动汽车销量位居全球第一。所以，是的，所以，是的，那么来看看。祝贺 Tesla 团队累计生产了 600 万辆汽车。

所以，我是说，Tesla 车队的规模确实正变得非常庞大。我是说，到今年年底将达到 700 万辆。超过 700 万辆。所以。

我们曾经。而我们位于弗里蒙特的工厂目前是北美产量最高的汽车工厂。所以。而且我们打破了它还是“新我”时创下的先前生产纪录。所以，实际上相当不可思议的是，我们在旧金山湾区有一座巨型汽车工厂。那里并不完全是建设汽车工厂成本最低的地方。就像在瑞士阿尔卑斯山或者什么地方一样，你知道。

但我们仍然设法大批量生产出了出色的汽车，这证明了我们在弗里蒙特拥有一支优秀的团队。所以祝贺弗里蒙特团队。

很多人曾说 Cybertruck 是假的，它永远不会推出，你知道，而现在我们正在交付大量 Cybertruck。我们的单周交付量创下了 1300 辆的纪录。

而且你知道，我认为 Cybertruck 确实很特别。

比如

人们有时会对 Cybertruck 有不同看法，但如果你真的想知道是否有一个……某样东西是否很酷、是否是一款出色的产品，那就给孩子看看，好吗？比如，不加掩饰，好吗？孩子就是不会掩饰，比如一个 5 岁、6 岁之类的孩子，甚至是 3 岁，然后问：你喜欢哪辆车？Cybertruck。

所以，大概就是这样你就知道了，而且终于有一样东西，它看起来就像未来。它。并且它开起来非常好，你知道，很舒适。它就是一款出色的产品。我认为它是，我认为它是我们最好的产品，而且其中有很多创新。它终于采用了 48v 低压系统，而不是 12v，还有 800 伏动力系统、世界上最大的铸件，你知道，它是防弹的。所有这些很酷的东西都可以，你知道，输出 F350 柴油版，而那是一款非常令人印象深刻的卡车。

而且它可以，你知道，它可以一边拖着一辆 Porsche 911，一边以比 Porsche 911 更快的速度跑完四分之一英里。

所以这太疯狂了。

所以，是的，如果你想想看，公司多久才能制造出真正特别、能打动你内心的产品？这种情况太罕见了，我认为这就是其中之一。是的。

而且我们推出了升级版 Model 3，它实际上是一款出色的汽车。我建议去试试。它，它确实是一款很棒的汽车。而且算上节省的汽油费后，它只要 216 美元，比如说，与类似的汽油车相比，总拥有成本基本上是 200 多美元。所以它确实非常划算，也是一款有趣的汽车。而高性能版，Model 3 高性能版，它就像，你知道，很快，还有 Porsche 911，它就是一款非常非常棒的汽车。

所以，是的，所以，是的。

当然，Model Y 成为了全球最畅销的车辆，而这是我们曾经预测过的。所以你知道，我想我在 2022 年说过，按销售额计算，Model Y 将成为世界上销售额最高的汽车，而到 2023 年，按销量计算，它将成为销量最高的汽车。结果确实如此。而且今年它将再次成为地球上最畅销的汽车。所以。

而且显然我们还有 Tesla Semi。你知道，我们正在小批量生产 Tesla Semi，就在上周，我批准了 Tesla Semi 的大规模生产计划。

所以我们将制造大量半挂卡车。我们将从，是的，这是某种，我认为它实际上会在财务上产生显著影响。它不是，它不是一件小事、小事。商业车辆的特点是，使用它们的公司极其客观。这不像是，实际上，我的意思是，显然我们会让它拥有很酷的造型，并成为一款很棒的汽车。但对于商业车辆，做出购买决定的公司只会看着它说，比如数字怎么样？

比如，用它运输的成本是低于还是高于，比如一辆柴油卡车？事实是，它的经济性比柴油卡车好得多。这基本上是不需要动脑筋的选择。如果你是一家运输公司，却不使用电动的，那款 Tesla 电动 Semi，你就是在亏钱。你为什么要这么做？不喜欢钱吗。好吧，但如果你确实喜欢钱，那么我建议使用 Tesla Semi。

所以，所以我认为这款产品，这款产品实际上将以令大家惊讶的规模销售，而且，而且实际上会在财务上产生显著影响并带来很多益处。而且对于，对于 CO2 和，而且对于可持续发展也是如此，因为半挂卡车实际上一直都在行驶，即便数量少得多，它们对碳排放总量也有，有非常大的影响。解决了。

所以很明显，我们有一些新、新产品，正在……之下研发

遮盖物

而且我认为这些，这些是，我认为这些将会相当特别。所以你知道，其中一些，我认为人们一开始可能会想，哦，它不会那么令人惊叹，但等着瞧吧，它会的，所以，所以我们的超级充电网络正在继续增长。你知道，有关超级充电站已死的传言被大大夸大了。

事实上，我们正在继续大幅扩展超级充电网络。事实上，我认为今年我们将安装更多、部署更多今年实际投入运行的超级充电站，数量会超过行业其他公司加起来的总和，仅供参考。

所以。

现在，对于超级充电站部署地点的资本效率，我们会更加谨慎，比如，你知道。但可以肯定的是，任何出现拥堵的地方，以及地图上我们遗漏的任何区域，我们都会在那里设置超级充电站。所以即便只算今年剩余时间，我们预计也会在超级充电站部署上花费大约 5 亿美元。所以这非常可观。这会是花得其所的 5 亿美元。

所以这是，你知道，数量很多。所以，而且我们，这实际上没有显示一批将在今年晚些时候和明年开放的新市场。所以很明显，我们希望 Tesla 遍布全球，而有很多国家，我们，你知道，在那些国家的 Tesla 数量为 0 或非常少，而显然，我们希望 Tesla 无处不在。所以有很多新市场是我们将在今年晚些时候和明年开放的。

我们也正在向其他公司开放超级充电站超级充电站，配备，那是用于旧式、略显、略显笨重的连接器的适配器。但我们的目标是帮助其他汽车公司。所以对于任何其他正在转向电动汽车的汽车公司，我们确实希望，我们确实希望通过我们的超级充电网络帮助它们。所以我们正在为它们提供适配器，并让它们能够使用 Tesla 超级充电网络，因为我们认为这样做比建立一个阻碍电动汽车发展的封闭花园更好。

这里有一点延迟，而且我们也在电池生产方面进行大量创新，使用的就是在这里制造的 4680。

而且这是，这是一个难题。你知道，比如有些公司整个公司就只生产电芯。那就是它们所做的全部事情。而且，但我们正在取得良好进展。比如你们看到的所有 Cybertruck 都使用 Tesla 4680 电芯。

而且我们有一条明确的路径，让 4680 成为，我们认为从，从制造效率的角度来看，可能最具竞争力的电芯。所以这是，这是，但它确实是，它确实是一个难题。我必须说，要擅长电芯制造，需要承受相当多的脑损伤。这就像一个脑损伤藏身处。

但我认为，如果，我不知道，地缘政治局势以某种方式发生变化，这确实会让 Tesla 具备韧性。但在电芯制造方面拥有一定的独立性是件好事。

所以，是的，我们正在稳步取得进展。而且我们还有正极材料精炼厂，你们可以在主工厂的后面看到它。所以你们大概可以在那张图片里看到它。那就是正极材料精炼厂。然后我们在得克萨斯州南部还有锂精炼厂。所以我们只是在确保自己基本上掌握了这块拼图。我的意思是，如果你看一段视频，比较 Tesla 如何进行比如正极材料和锂的精炼，以及行业其他公司如何做，那就是天壤之别。

我的意思是，在Tesla的精炼厂里，你差不多可以直接从地板上吃东西，而在其他精炼厂里，我不建议这么做。

所以，你知道，这，这确实是，我们在这里肯定正变得非常、非常垂直整合，

但是

我认为这是一项，这是明智的投资，它将，它将带来比人们意识到的更多回报。所以。

我觉得这东西没在工作。

所以今年，我们也正按计划完成数量庞大的能源部署。因此，部署能力有了巨幅提升。所以我们确实，我们似乎正朝着储能、部署以及固定式电池包同比增长约2%到300%的方向发展。所以规模非常庞大。

真正的限制因素是能否制造更多Megapack和更多Powerwall。所以我们正在提高Powerwall 3的产量；就个人层面而言，Powerwall 3确实是一款改变游戏规则的产品。任何特定技术通常都需要大约3次迭代，才能真正，才能达到那种让人觉得，好了，现在我们真的找到了最佳平衡点的状态。所以Powerwall 3是一款史诗级产品。

Megapack也是，

我会称它为

算是Megapack的第2次迭代，也是，是一款FA产品，而Megapack 3，我不知道，大概还要几年，我们，我们实际上将开始吸收越来越多的变电站的，那些，那些，那类电力电子设备。所以，因为我觉得，你会希望Megapack达到这样一种程度：你真的只需要把高压电力线路接上去就行。根本不需要变电站。

我们只要把它们接上。只要把它们放下来，把它放下来，接上，现在它就能工作了。

顺便说一句，这对公用事业行业来说简直令人难以置信。比如他们会说，什么？是的，只要把电线接上，它就能在极高电压下工作。所以我认为这实际上将会是一项重大的。所以总之，是的，这，这很棒，储能团队的工作非常出色。

服务及其他业务看起来也不错。所以我们现在实际上已经能从服务及其他业务中赚钱了。

所以这算是一件大事，对吧？是的。那么让我们看看。

是的，而且 Tesla 显然远不止是一家汽车公司。你们知道，我们在 Tesla 做了很多软件，你们知道，大约一半的，或者说我们所做工程中非常大的一部分，实际上是软件工程。所以我会说，Tesla 既是一家软件公司，也是一家硬件公司。这非常重要，因为汽车公司通常不是软件公司。所以这，这一点非常重要。比如我们用于储能的订单更好软件、控制汽车的所有软件、Megapack、Powerwall、太阳能的软件，显然还有用于 AI 和完全自动驾驶的软件，以及重要的保险、服务和碰撞业务，Tesla 还在内部编写了大量有助于这些业务的软件，比如我们内部所称的 Tesla 操作系统，它远远领先于其他任何公司拥有的系统。

我认为可能比任何一家《财富》500 强公司都要好。

Tesla 的内部软件，它就是好得多。所以，是的，它远远超出了人们通常对汽车公司的认知。

而且 Tesla 也是世界，现实世界 AI 领域的领导者。所以这确实非常重要，比如 Tesla 领先于 Google Matter，你知道，向任何现实世界软件开放，实际上查看、接收视频并根据视频做出决策。

没人能，甚至没有人接近。而且我会说，它每过1个月，甚至每过1周，都在变得更好。

所以，是的。

而且，而且值得注意的是，Tesla 实际上相当擅长芯片设计。所以，那个 AI 推理芯片，那是由 Tesla 设计并装在汽车里的。我们算是有硬件 3 AI 推理汽车，而在，你知道，过去1年里制造的汽车配备了硬件 4。我们刚刚完成了硬件 5 的设计，现在我们把它称为 AI5，因为它确实是，实际上仍然没有来自 Nvidia 或来自。

而且我非常尊重 Nvidia，来自任何公司、我们更愿意装进汽车里、并且比我们车上现有芯片更好的芯片。所以，所以我们从零开始设计芯片，就像我们从零开始开发 AI 软件一样，并且从无到有，拥有世界上最好的现实世界 AI 软件和最好的 AI 推理芯片。所以这，你知道，是一件大事。

而且车载芯片的能力非常惊人。我的意思是，现在所有汽车实际上都在训练。

我们让硬件 4 以模拟模式运行硬件 3。所以我们会继续在硬件 3 上取得重大进展。但今年晚些时候，我们实际上会分成两条路线，继续研究硬件 3、在硬件 3 上训练，但随后在硬件 4 上进行单独训练。那将是我们正在超级工厂南侧建设的那种训练集群。那将是专用于视频和推理的硬件。

而且你知道，硬件 4 的摄像头分辨率大约提高了4或5倍。取决于你如何计算这种硬件 4，它比硬件 3 好大约3到8倍不等。但到目前为止，你们看到的一切都只是硬件 3。而且在达到硬件 3 的极限之前，我们还有很长的路要走。所以我认为硬件 3 会实现惊人的成果，但我认为硬件 4 可能会好大约5倍。

然后是硬件 5，大约18个月后推出，它的能力是完整硬件的10倍。它就像。所以如果你了解 Nvidia 硬件，它，它算是一台 B200 级别的计算机。

所以你知道，大概就是这样。所以我们接下来要逐步做的，就是提高汽车可靠性中有多少个9。

而我，而且，而且当然，那也会应用到 Optimus 上。所以同一款芯片会装进，你知道，硬件 4 会装进选……。是，是一个 Optimus 硬件 AI5，我们称之为从硬件 5 改称 AI5，将在大约18个月后装进 Optimus 和，和所有汽车。它确实拥有惊人的算力。而且这是非常、非常节能的算力。所以它，它必须如此，因为如果你是在像人形机器人或汽车这样的移动应用中，就不能像在数据中心里那样消耗10千瓦的功率，你知道。

所以它必须非常节能。

所以，那个，你知道，大致来说硬件 3 和 4。只有几百瓦。

是的，是的。这很难。

现在，硬件 5 可能将能够达到大约7或800瓦，但它的功率会根据所处场景的复杂程度而波动。所以如果它静止在停车场里，那就像，你知道，相比之下。你知道，不需要想太多，就像人一样。比如，如果你身处复杂的交通场景，就必须比仅仅在空旷道路上巡航时思考得更多，你知道。

但我认为未来可能有意思的一件事是，比如在某个时候，我认为 Tesla 车队可能会，你知道，超过1亿辆汽车。而且如果每辆汽车都有1千瓦的高效推理算力，我认为这里有一个，有一个，我认为算是类似 Amazon Web Services（AWS）的机会。因为如果你有1亿辆汽车，每辆都有1千瓦的高效推理算力，你就有100吉瓦的算力，比如100吉瓦的计算机非常多。

而且它分布在世界各地。所以即便如此，当汽车不处于自动驾驶模式时，我认为它可能会，你知道，每周做50、60小时的自动驾驶出租车工作，但每周有100小时，它，它没有工作，它可能是静止的。所以。所以这里有100小时、100吉瓦的推理算力，我认为我们应该利用它。

为什么不呢？你知道，当人们审视 Amazon 时，它起初，我会说，是一家在线书商，而它在，你

知道，

已经发展成这个令人难以置信的、可以买到任何东西的地方，然后他们，然后是 Amazon Web Services，比如，好吧，他们拥有所有这些只在某些时候达到峰值使用率的计算机，但在非峰值使用时，他们要拿这些计算机做什么？而且有时 Amazon 服务器会有大约10%的时间处于停机状态。所以那时他们说，好吧，我们来做 Amazon Web Services。然后 Amazon Web Services 变得比 Amazon 其余所有业务加起来还更有价值。

而且，你知道，我认为未来 Tesla 在这方面存在某种相当重大的机会。

你知道，再说一次，这确实，确实没有人把这一点考虑进去，但我认为它实际上会相当重要。

我们的训练也不再受算力限制。所以我会向团队了解情况。就像是，对于训练和推理，有没有什么我们可以做的事情来加快进展？而目前这并不是限制因素。事实上，目前的限制因素是干预之间的行驶里程太长了，以至于需要相当长的时间才能弄清楚哪个版本比另一个版本更好，因为它们都不需要任何干预。

所以，就像，你知道，如果干预之间开始达到数千英里，或者要行驶大约10,000英里才出现一次干预，那么，普通人1年也只开大约10,000英里。而且，如果是在城市环境中，平均速度是每小时20英里，我的意思是，坦率地说，我们的专业测试驾驶员都觉得相当无聊。你知道，他们会说，好吧，我开了整整1周，一次干预都没有。

就像这周最精彩的时刻会是：太好了，一次干预。终于，就快到那个程度了。

所以，在这方面，拥有一个庞大的车队实际上极其重要，因为我们可以部署一个新的FSD模型，让它在影子模式下运行，然后看看，看看它的表现如何。

你知道，把人类驾驶汽车的方式与新的自动驾驶版本进行比较，然后，然后在影子模式下分析这种差异，就像影子知道，或者视情况而定，并不知道一样。然后，然后借助如此庞大的车队迅速获得数十亿英里的数据，从而能够进行评估。基本上，那个，那个数据引擎极其有用。实际上，如果没有数百万辆汽车在道路上行驶，就不可能解决自动驾驶问题。

所以，所以这实际上就是，就像，想出巧妙的方法来测试下一个版本——FSD版本——到底有多好，目前实际上是一个限制因素。然后当然，就像我说的，我们正在这里建设另一个训练数据中心，它将专用于Hardware 4训练。

所以我们将在今年晚些时候把硬件 3 和硬件 4 的训练分开。我们会继续改进硬件 3，但也会释放硬件 4 的全部能力。

所以，是的，我想在座的大多数人都试用过版本 12。

我们确实说过，大概，无监督的完全自动驾驶。完全自动驾驶会是版本 12。所以我们实际上只是随意地把版本号维持在 12，然后，比如，把它叫作 12.4、12.3、12.4、12.5，但实际上它们真的就像版本 13、版本 14。不过不管怎样，这只是一个任意的命名。所以 12.4 实际上就像是一个与 12.3 完全不同的版本，而 12.5 是一个完全不同的版本 12.4。

所以你们会看到真正巨大的改进，我认为有时后续版本之间会有 10 倍的改进。

然后，正如我提到的，对于现有车队，它的运作方式是，你可以按自己的意愿把你的车加入车队或从车队中移出。

而且也会有 Tesla 自有的车辆。所以它会是类似 Uber 加 Airbnb 的组合。

但会相当疯狂的是，只需一次软件更新，然后整个车队突然就都可以使用了。就像突然之间你有了700万辆、1000万辆汽车，最终会有数千万辆能够自动驾驶的汽车，而且它们不再是每周使用10小时，而是每周可以使用50或60小时。

所以，你知道，身处帕洛阿尔托，周围是一堆立方体，然后一个人形机器人就这样走过去，这相当疯狂。

我们在很短的时间内让 Optimus 取得了巨大进展，从有人穿着套装假装机器人跳舞，到一个相当东拼西凑的机器人，再到如今一个真正在工厂里执行有用任务的机器人。所以，我们在弗里蒙特工厂有2台 Optimus 机器人，基本上在执行这项任务，也就是从生产线末端取下电芯，并把它们放进运输容器。还有，是的，实际上有相当多这样的机器人在我们帕洛阿尔托的办公室里四处走动。

所以，这个，而且我想我们大概还有一次主要的硬件修订，应该会在今年年底或明年年初之前完成。然后明年我们会让 Optimus 进入有限量生产，在我们的工厂中使用这些有限量生产的 Optimus，我们会在那里测试产品，算是，你知道，你知道，就像我说的，有点像吃我们自己的狗粮，或者不管它在电子领域对应的说法是什么。

所以，但是我，我觉得就明年而言，我，我的预测是，明年会有超过、超过1000台，也许几千台 Optimus 机器人在 Tesla 工作。

而且，而且，而且此后规模会非常迅速地扩大。所以我们会消除这些问题。它会，就是，自主程度将大幅提高。你真的只需要跟它说话，说，请完成这项任务，或者我要给你展示一些东西。现在照着做，我正在

给你展示的东西

而且，你知道，最终它可以观看一段像是某个人做某件事的视频，然后仅仅通过观看那段视频来学习，并执行那项任务。所以，是的，所以它确实会非常了不得，而且我确信这一预测：人形机器人与人类的比例将会大于1比1，因此世界上将有超过100亿个人形机器人。

可能是20或更多。而 Tesla 将遥遥领先。你们看到有很多机器人初创公司，但我认为，作为一家机器人初创公司，要做 Optimus 实际上非常具有挑战性，因为我们发现，为了让 Optimus 运作起来，我们不得不从第一性原理出发，从零开始设计机器人的每一个部件。所以电机、齿轮箱、传感器、电力电子设备、通信系统，一切都必须从零开始。

我们发现基本上什么都没有，没有供应链。所以，尽管世界上生产了许多电动机，但人形机器人所需的那些类型的电机、传感器和齿轮箱并没有供应链。

而且，我的意思是，你们在这里看到的是我们当前一代的手和手臂，但我们的下一代，也就是拥有11个自由度的，我们的下一代拥有22个自由度。它将能够弹钢琴。

所以真的就是，哇。

当然，就像我说的，我们需要确保不会出现《终结者》那样的情景。这非常重要。

所以人形机器人的安全性将非常重要。

但由于它需要如此多从零开始的设计，要从零开始设计每一台电机、齿轮箱、传感器和电力电子设备，对一家初创公司来说，要复制这一切非常困难，如果不是不可能的话。但在 Tesla，我们拥有世界上最优秀的电气工程能力。我认为在齿轮箱、电动机和电力电子设备方面，我们拥有世界上最优秀的机械工程能力。

所以我们有资源去做这件事。它非常适用。然后你还必须拥有那个，那个大脑，你需要那个，你需要那个、那个、那个，你需要一台高能效的推理计算机，我们已经为汽车配备了这种计算机，也会在 Optimus 上使用它。你需要 AI，真正的，你需要在现实世界 AI 方面做到最好，而 Tesla 在现实世界 AI 方面是最好的。所以你需要所有这些，你需要手里有一副非常强的牌，才能制造出一款有吸引力的机器人。

然后你还需要非常擅长规模化制造。所以，为了不让机器人的成本达到数十万美元，为了让它的成本达到比如10或20,000美元，实际上需要针对制造进行设计，并且非常擅长制造。

根据我的经验，与大批量制造相比，原型很容易。原型很容易。相对而言，生产很难。所以 Tesla 拥有生产能力，拥有工程能力，也拥有 AI 硬件和软件能力。

甚至在我见过的对 Optimus 最乐观的估计中，那些 Optimus 乐观主义者，我认为也低估了这个机器人将能够做到的事情的规模。

你知道，就像我在演示的、演示的一开始所说的，我，你知道，我同意 Ark Invest 的分析，即自动驾驶交通是一个所谓的5万亿到7万亿美元市值的局面。我认为 Optimus 是、是一个25万亿美元，确切地说是25万亿美元市值的局面。

所以。

现在我不想，我不想轻描淡写、轻描淡写实现这一目标所需要的一切。我的意思是，要实现它需要完成极其庞大的工作。真的超级困难。但我们正在沿着这条路飞速前进。我们会让它实现。所以谢谢大家。

我想这样我们就可以进入问答环节了。

抱歉。对。哦，我想我们已经把麦克风排好了。

该用麦克风了。

嘿，埃隆，去年我感谢了你

感谢你让这个世界变得更美好，让我的儿子能生活在其中。而且我刚才没有特别说明白，我指的是你为《第一修正案》和我们的权利所做的一切。谢谢。

历史不会忘记这一点。还有你所作的牺牲和你所冒的

风险终将获得回报，而且已经获得了回报。

谢谢。

我非常感激。你知道，我认为，这个，你不得不说，比如，为什么，为什么《第一修正案》如此重要，是因为，你知道，人们从一些地方来到这里，在那些地方，如果他们说出自己的想法，就会被关进监狱或杀害，或者干脆被社会排斥、遭到孤立。所以，言论自由对于美好未来绝对至关重要。而且，是的，这就是，你知道，收购 Twitter，并试图把它打造成一个言论自由平台的原因。

而且我认为，这个，这就像是，我认为对文明的未来非常重要。所以谢谢。

嘿，埃隆，我叫罗尼。我在你的右边。最近怎么样？

嘿，我在伊利诺伊州奥兰帕克从事 Tesla 销售工作。我一直想知道，你认为 Tesla 有没有可能推进把软件捆绑成一个套餐，从而，你知道，可能提高 fsd 的采用率，你知道，也许把它与高级连接服务和未来的软件选项捆绑起来？

我认为我们肯定希望鼓励人们尝试 fsd。而其中一件，我认为可能算是合理的事情是，当我们取得重大改进时，每次我们对 fsd 作出重大改进，都给人们，你知道，给所有客户一次，比如为期两周的免费试用。所以，比如，嘿，我们改进了 FSD。你知道，再试两周。你知道，在，在，在 12.4 版本中，我们终于取消了方向盘提醒，我知道它快把人逼疯了。

就，就，就像是谢天谢地，总算有点小恩惠。

所以你知道，即使只是这样的事情，没有方向方向盘提醒，而且显然干预次数大幅减少，比如每次干预之间能行驶远得多的里程。所以我认为我们会，你知道，我认为每当我们取得重大改进时，都会继续算是向人们提供 FSD 免费试用。

哦，抱歉。是的。当你听到那么响的，扬声器时，你就会觉得，耳朵不管用了，你知道。

下午好。

是的，其实有两个问题。首先，有了560亿美元之后，你现在大概有钱带孩子去迪士尼乐园了，那是你准备去的地方吗？

嗯，我是说，顺便说一句，值得强调的是，那是我必须持有5年的 Tesla 股票。所以你知道，比如它就像是，它并不是真的现金，我不能拿钱跑路，我也不会想那么做，但是。

而且实际上，我已经和孩子们去过迪士尼乐园和迪士尼世界很多次了。如今我去那里而不引起骚动有点困难。

你知道，也许我就该，比如穿上一套米老鼠服装直接进去，然后，你知道，假装我是其中一个。

所以，随着我们推进自动驾驶，并逐渐达到这样一个门槛，也就是事情正发展到非常先进的水平，最终将会出现一些严峻的后果。这是取得突破的一部分。

对。

我只是好奇你对此是怎么想的，还有你是否考虑过，你知道，为了拥有一个更好、更安全的未来，我们很不幸将不得不经历的一些早期后果？我只是想知道，你是指可怕的机器人之类的吗？

不，不，是汽车方面。

汽车方面，搭载完全自动驾驶的汽车。好的。比如我们看到一些公司发生了一些不幸的事故，然后，哦，对，就直接关门了，而且你知道。是的，这不是计划。但这些确实是开发这项技术所带来的真实后果。我只是想知道你对此是怎么想的。

是的，嗯，我们确实会，我们确实尽力非常谨慎地部署 FSD，而且你知道，我们确实有很多汽车。所以呢？就像，当有很多汽车，并且已经行驶了，你知道，数十亿英里时，不可避免地会出现大数定律的情况，也就是说，发生坏事的概率很小。但当你看每英里的安全性时，因为我们积累了大量里程，很明显，每英里的安全性优于人类驾驶。

所以，所以当，当我们，你知道，我们可以向监管机构展示这一点，并且显然可以向公众陈述我们的理由，说实际上，你知道，人类驾驶并不完美。你知道，大约每行驶 8000 万英里就有 1 人死亡。所以，比如，美国每年约有 35,000、40,000 人死于汽车事故。重要的是，我们是否正在让这个数字变小？只要我们正在让那个，那个数字变小，我们做的就是正确的事。

谢谢。

嘿，埃隆，我是巴什。我有 1 个问题。

当然。

所以我们知道，我们知道。我们爱你们所有人。我们发自内心地爱着。我们非常爱你，埃隆。

谢谢。

我的问题是，我的问题是，我们都知道 Tesla 有一天会成为世界上最有价值的公司。

是的。

我们只是想知道，你是否对时间以及检验点是什么有任何预测。什么时候，什么时候我们可以说这就是检验点，是，是解决 FSC，还是比如 Optimus 落地会是你，你知道，从你的角度来看所认为的那个节点？

嗯，正如我之前提到的，我认为值得看一下，

你知道，

Ark Invest 的分析，我认为那是最准确的分析。他们的预测是，我想，到 2029 年市值将达到 5万亿至7万亿美元。所以在 5 年内会达到今天的 10 倍以上。我认为这是准确的。

谢谢。谢谢。

嘿，埃隆。

嘿。

超级粉丝。

而且我认为历史会证明，

你对人类产生了深远影响。

所以我认为这是不言而喻的。然后只有几个问题。我想，我通过

把车放到 Turo 上，治好了我的 Tesla 瘾，而且我发现，大约一半的租客是第一次驾驶 Tesla，而

另外一半人在试驾之后真的很想买一辆。所以我想问几个问题。

我们大概什么时候可以期待延期计划恢复？如果它会恢复的话？另外，我们能不能也推出一款载客

厢式车，比如一款 12 座的厢式车，而且是在不久之后？

我会把这个加到我的清单上。

所以，那个，没错。所以推荐计划。

是的，我想我们大概会在下个季度重新审视推荐计划，而且非常重要的是，要看看，一辆 Tesla 经全面考量后的成本究竟是多少？所以有了推荐计划，实际上会提高汽车的价格。所以，你知道，对最终客户而言，怎样的价格上涨才更好，是因为推荐计划而略微涨价，还是不涨？但我想我们会审视这件事，我想是在，你知道，下个季度，然后，然后看看它，看看它是否合理。

我认为某种推荐计划确实是合理的。

而且，你知道，但我认为，对很多人来说，遥遥领先的，遥遥领先的最大问题是，人们就是买不起这辆车。这是最大的问题。就像，那就是需求的限制因素。

你可以想象，在极限情况下，如果有，比如说，你知道，一个18，就像，如果 Model Y 的价格，比如低于20,000美元，它仍然可能有大约500万辆。那会很疯狂。所以现在这在今天还不可行。但我认为，努力找出制造更实惠车辆的方法，才是真正改变游戏规则的关键。比如，对。对大多数人来说，问题真的不在于他们是否想要一辆 Tesla。他们确实想要一辆 Tesla。

他们只是没有足够的钱买得起。所以我们必须让它价格亲民。这至关重要。谢谢。

嘿，埃隆。我只是想自我介绍一下。我叫哈姆扎。首先，谢谢你把我介绍给安拉贝克。多亏了你，我和他现在关系很亲近，而且多亏了你，我搬到了奥斯汀。所以我的问题基本上是关于 FSD 转移的。我认为这应该成为一项持续进行的安排。我认为这不仅对你有好处，可以让更多人加入地面车队，并获得，你知道，统计数据和信息，而且对我们 Tesla 车主来说，这也会让我们想购买下一款车型并把它转移过去。

你知道，例如，一直让我没有行动的原因，就是想等 FSD 转移再次开放。那么，你为什么没有把它实施为一种，比如，持续性的安排，是有什么原因吗？这对我们来说做起来相当复杂。它确实给销售流程增加了不少复杂性。但是，是的，而且这很棘手，因为我们总得以某种方式赚取收入。

所以，你知道，转移这件事很棘手。

抱歉，再延长1个季度。

好，再延长1个季度。

好吧，就这么定了。

再1个季度。

好吧。所以，是的。

好吧。

哦，埃隆，谢谢你接待我们。如果你能看到我，我正不受控制地发抖。真的非常敬仰你。作为股东、消费者，你知道，过去2天里，我把自己持有的那点微不足道、少得可怜的你的股票增加到了3倍。仅仅是参与其中，知道我们要来这里，也感谢你接待股东，这对我们意义太重大了。不客气。

你绝对是我的偶像。提出这个问题，是因为我把你当作一个人来关心，但从股东的角度来说，要释放 Optimus 的这些价值，我们需要你。而且面对全球所有这些不确定性，你又如此引人注目。只是想确保你正在尽一切努力照顾好自己。为了公司的未来，注意安全和健康。

我这么说是因为我关心你这个人。但既然这是一次股东大会，本着股东价值的精神。嗯，谢谢你。我很感谢这些善意的话。

是的，我想我可能需要锻炼，而且你知道，还有比如不要被暗杀之类的，你知道，我是说，最近情况变得有点疯狂，你知道，比如粗略地说，一个杀人狂试图杀你的概率，与有多少杀人狂听到过你的名字成正比。好吧，所以他们经常听到我的名字。我就想，好吧，我在名单上了，你知道。

所以，是的，过去大约7个月里，确实有2个杀人狂前来，心想着要杀掉我和其他一群人。所以不只是我。而且他们并没有明确说出什么实际问题。他们只是在从事杀人狂这一行。所以。

所以我确实需要稍微谨慎一些。

确实如此。

这不太好，因为它迫使我表现得更有点疏远，你知道，而且你知道，比如，现在我已经不再给东西签名了，因为只是，你知道，某种事情。因为比如，那肯定会成为一种像杀人狂那样的方式，我就签个东西，然后开枪打他们之类的。所以之后我，你知道，我不得不表现得更疏远之类的，而我并不愿意这样。但我认为这似乎是明智的，就像约翰·列侬，他当时唱着类似这样的话，嘿，我们大家就不能彼此友善一点吗？

然后他被他的一个粉丝开枪打了，你知道，就像，好吧，我们会尽量避免这种事。

所以不管怎样，我很感谢这些善意的话。我会努力活下去，并且。

好的。

嗨，埃隆，我前几天读了一篇文章。Tesla 的产品时间线上是否仍有某个正在开发的家用 HVAC 系统，一个家庭互动系统？据我所知，我们没有家庭互动系统。不，不，我们不想要 HVAC 系统。

哦，HVAC。

哦，HVAC。对，对，当然，对。在那些可以说更有用的事情清单里，比如说，做一个很棒的、非常节能的 HVAC 系统会很好，你知道，它有内置的 HEPA 过滤器，并把我们在汽车上学到的很多东西应用到家用 HVAC 上。

因为我。我其实觉得外面也许有很棒的家用 HVAC 系统，但我没有，我还没找到。这挺不可思议的。

还有就是内置过滤系统，而且你知道，能够在一定程度上也控制湿度，并且像我说的那样，超级高效又安静，这些都是我们在汽车上做的事情。

这会是一件相当棒的事情。也许我们可以把它和 Powerwall 之类的东西结合起来。

它在待办事项清单上的优先级不是特别高，但有朝一日做这个会很酷。

嗨，埃隆，我是桑迪普。我只想感谢你制造了 Cybertruck。

不客气。

对，孩子们很喜欢它。真的吗？我家 4 岁和 6 岁的孩子，他们很喜欢它。

没错。我说的就是这个，孩子们。简直太不可思议了。

就像我上周去了迪士尼。没人看迪士尼。所有人都在看萨巴斯。

对。

有个有趣的故事，如果你允许的话。我们当时在让女儿做好准备，我们去了德斯廷，然后从那里开车去迪士尼，接着我们在让女儿准备好去。她 4 岁。就是要习惯排队，而且你知道，排队必须非常有耐心。她就说，爸爸。我们为什么不直接说，我们会让你看看我们的 Cybertruck，然后我们就可以插队了。我的意思是，她已经习惯了，因为我们当时在驾驶超级充电站，然后所有人都会走过来拍照。

当然。

这是个有趣的故事。

对，对。

然后我有一个请求。你会允许 6 岁的孩子参观工厂吗？也许未来某一天可以。现在的年龄限制是 12 岁。好的。尽管我们有推荐奖励，年龄限制仍是 12 岁。你会允许孩子们参观工厂吗？也许他们可以看看机器人。

当然，当然。

我们可以办个活动之类的。

对。

也许我们可以弄一个更糟的，比如担心儿童安全之类的，你知道，以防孩子跑进机器人工作单元之类的地方。但是，但是对，我想我们可以降低这个年龄，也许安排一种更有限的参观，让年幼的孩子可以参加。当然，我们会做的。

好的。

看。

好的。

唐纳德·特朗普一直是电动汽车的强烈批评者。

但上周，他说自己是 Tesla 的忠实粉丝，也是你的忠实粉丝，这让我们很意外。

你对他说了什么？

嗯，你知道，我很有说服力。

所以。

对，其实我也不完全清楚他为什么。这是个好问题。我的意思是，你知道，我和他有过几次交谈，而且他确实，他确实会毫无征兆、无缘无故地打电话给我。我不知道为什么，但他确实会。而且，而且他打电话时非常友善，你知道，然后我就说，你知道，你知道，我觉得电动汽车对未来很有好处。美国是电动汽车领域的领导者，你知道，你知道，购买美国货之类的。

而且，而且我觉得他现在其实有很多朋友都拥有 Tesla，他们，他们都会喜欢它。他还是 Cybertruck 的超级粉丝，所以我认为这些可能都是促成因素。

嘿，埃隆，在你右边。

是的。

我叫基努，长期以来一直是你们品牌、公司以及一切的粉丝。我想知道，当你们把驾驶员从无人驾驶汽车的驾驶座上移走时，你们计划如何处理干预？我确信干预会在某个……某种频率下发生，即使很少见。

是的，所以，我的意思是，很显然，我们希望确保如果车内没有驾驶员，需要严重干预的概率极低。你知道，我们希望确保汽车在骑行者或行人等弱势道路使用者周围极其谨慎。

所以，我们真正希望伤害行人或骑行者的概率低得难以置信。比如，西装，远低于某个人驾驶汽车时。然后，还有一些干预发生时，汽车就只是，比如，我不知道，在一条单行道上卡住了，而那里有一种疯狂的施工情况，但在这种情况下，它基本上会联系总部并说，救命，我需要，我卡住了。

然后我想，我们或许只需，你知道，远程接入汽车，像操控一辆巨型遥控车那样驾驶它，就能解决许多这类问题。

这是我的猜测。我认为我们会这么做。

嘿，埃隆，我是来自哈萨克斯坦的安瓦尔别克。感谢安装超级充电桩。我只是想感谢 Tesla、Boomer Mama 和索耶·梅里特所做的一切努力。他们确实组织了……

在 Luster Pearl East 有一场庆祝活动后的派对，如果你们愿意，可以过来庆祝大家为社区付出的一切努力。祝贺你，埃隆。谢谢。好的。

埃隆，感谢你今天主持活动。自从你举办电池日以来，大约已经过去4年了，而长大后，那里曾有非常雄心勃勃的目标，要达到1太瓦时的电池产能。在你之前的演讲中，你提到了扩大这项新技术规模的巨大困难。我认为，作为股东，我们最感兴趣。你能再多告诉我们一点吗？因为就可负担性而言，你也提到这是一个巨……巨大的负担。

这些4680电芯能获得制造税收抵免，而且在消费者一侧也能获得大额税收抵免。如果你们能扩大它的规模，让我们能够满负荷生产 Cybertruck

并且也开始把它们放回

Model Y 和 Model 3 中，这可能会大幅提高可负担性，并且在规模方面也会很棒。

当然。

所以，在目前这个阶段，实现可负担性实际上更多取决于汽车的其他部分，而不是电池包本身。供应商电芯的每千瓦时价格已经大幅下降。所以有几年，供应商锂离子电芯的成本出现了巨幅飙升，因为每家汽车制造商都下了巨额电池包订单。比如巨大、巨大的订单。所以，锂之类的价格，以及所有原材料和电芯的价格都疯涨了。

它翻了一倍，比如电芯价格翻了一倍。我认为锂价上涨了差不多5倍、5倍，或者说500%。就是上涨了疯狂的数字。所以从那以后，需求，比如供应商，电芯供应商增加了供应，而其他汽车制造商的订单则下降了。

所以，我的意思是，值得注意的是，电动汽车现在处境艰难。你知道，我记得通用汽车最近宣布将其电动汽车削减50%。福特宣布了削减计划。我认为几乎每家制造商都宣布缩减其电动汽车计划。至于 Tesla，我们会增加，但这并不是一个轻松的市场。

但是，要降低汽车的成本，电池包确实是拼图的一块，但其实其他一切也很重要。所以我确实会这样形容，比如，我们要让汽车变得更实惠。这就像《权力的游戏》，但玩的是便士。《便士的游戏》。所以，引用一下，按一级近似计算，一辆汽车有10,000种独特零件。如果每种零件的成本是5美元，那么你就得到了一辆50,000美元的汽车。所以，要真正降低汽车成本，你必须一便士一便士地削减，让这10,000种零件的成本从5美元变成3美元之类的。

大致就是这么回事。

现在，关于我们自己的电芯生产，我们确实看到了在今年年底前实现成本持平的路径。

实现成本持平是一条非常艰难的道路。目前，我们的4600成本高于供应商。现在，它们的成本比供应商目前的更高，但比供应商一年前的更低。所以你知道，电芯供应有点像一种野兽饥荒式的情况，有点像DRAM芯片，你知道，DRAM行业会从，比如说，你知道，供过于求变成供应不足。DRAM的价格变化得非常疯狂。销售方面也有点像这样。

所以。但是我们预计，到今年年底，即使与供应商目前低得多的销售价格相比，我们也能实现成本持平。

您好，马斯克先生。我从2012年起就是长期投资者。很好。而且我在2018年投票支持了您的薪酬方案。再次感谢您今年所做的一切。还要向Tesla Boomer Mama致意。Alexandra，

我曾有幸交谈

在2013年和您短暂交谈过。您当时谈到了一款卡车，后来它成为了Cybertruck。您是一个有回报的人。我在世界各地都有朋友，包括亚洲、加拿大和欧洲。而且我知道Cybertruck一直在巡展。Cybertruck获得了怎样的反响？它什么时候会，或者说是否有可能销售，以及什么时候会在国际市场销售？是的，所以我想，也许，你知道，我们或许能在明年某个时候让它通过其他市场的认证。

但可以肯定的是，今年它只会在北美。我们确实是按照北美的要求设计这款车的，因为如果你一开始就采用所有国际要求的超集，那会给、给、给Cybertruck带来很多限制，坦率地说会让产品变得更差。

所以我认为，我们需要制造一个特别版本，例如符合中国标准或符合欧洲标准的版本。

但在Cybertruck实现更高产量的生产之前，增加这种复杂性确实没有意义。在成本方面仍有很多工作要做，要降低Cybertruck的成本。你知道，我通常会说，比如，从原型到，比如说，从原型到生产的难度，要高出100倍，从、从，你知道，从原型到生产要困难100倍。

然后，一旦实现生产，要把价格改善20%。比如把商品成本降低20%，比最初实现生产还要难。这简直是巨大的痛苦。就像，你知道，坦率地说，这不是世界上最有趣的工作。你知道，比如一分一分地抠成本并不是。坦率地说，这就像洗碗，你知道，就像要在这里洗很多碗。制作原型特别有趣。那真的很有趣。

而且，让生产运转起来算是、算是挺酷的，确实挺酷。然后，降低成本的苦活就是一种苦活，比如，你知道，那是艰苦的工作，你知道，它也有一些回报。就像洗碗也算有某种回报。我想，你会得到干净的碗。

但这、这确实很艰难。

所以那确实是非常艰苦的工作。我只想说，反响真的非常热烈。比如我有朋友，你知道，给我发消息说他们喜欢Cybertruck的设计，而且它、它绝对很吸引眼球，所以。哦，是的，我认为从需求角度看，世界其他地区肯定会有大量对Cybertruck的需求，毫无疑问。比如无论它去到世界上的什么地方，都会引来一大群人。

所以，所以这不是需求问题。而是。实际上，我们必须实现大规模生产，熬过降低成本的苦活。我无法。我很难夸大降低成本有多么困难。它真的非常艰巨。然后我们必须让这款车重新通过认证，其中包括做出一些设计变更，以满足、符合其他市场的要求。

好的。

下午好，埃隆。我喜欢你的鞋。我不知道其他人能不能看到。是的，在那边。Cybertruck。

是的。

我的问题是关于Foundation系列，你们是否计划生产某个特定数量，或者非Foundation系列会在什么时候推出，有没有时间表？是的，我们会开始生产，我们实际上很快就会结束Foundation系列，并在下个季度的某个时候推出非Foundation系列。

你好，埃隆，感谢你为股东创造价值，也祝贺你的薪酬方案。我是Michael。我在季度投资者电话会议上提了很多问题，我想问你，能否进一步介绍一下你们在训练FSD时观察到的神经网络缩放定律？比如，为了让全自动驾驶软件的可靠性连续增加每一个9，你们需要多少数据、多出多少推理算力。多出多少训练算力。

你们目前又在收集多少数据？谢谢。是的，这是个很好的问题。

我的意思是，这确实是我们每周都会在Tesla讨论的事情。我花很多时间与Tesla AI团队一起工作，而且我想可能，比如，你知道，比如，占了很大一部分，我的意思是，这确实是我工作中很大的一部分，就是，好吧，我们需要让汽车实现自动驾驶，你知道，而且，是的，你需要在芯片层面解决推理、训练和软件方面的问题。数据存储也面临巨大的挑战，因为有PB级的视频数据不断传入。

而且即使对于，比方说，一个大型Nvidia集群，实际上要让一个大型Nvidia集群真正用于训练并正常工作也非常困难。并不是打开箱子插上电就行。完全不是那样。要让一个大型训练集群运行起来、真正运行起来，然后以较高的正常运行时间工作，需要付出巨大的努力。

所以我的意思是，就我们看到的各个版本而言，比如从版本12、3开始，只要忽略12，把它看作3、4、5、6。实际上就应该这样理解，忽略12那部分，就把它称为版本3、版本4、版本5，那是版本6。应该这样理解。每个版本之间，我不知道，改进幅度至少会翻倍，在某些情况下，我认为会提升5到10倍。

你知道，尤其是当我们消除，比方说，我认为我们已经接近消除几乎所有与静态物体的碰撞。所以你知道，比如弯道和中央隔离带，我们已经非常接近做到与静态物体的碰撞几乎为零。所以一旦静态碰撞降到零，每两次干预之间的行驶里程就会获得巨大提升。

然后你会遇到不同类别的动态物体，而不同类别和不同事物显然会变得非常复杂。如果你遇到一个复杂的十字路口，到处都是交通信号灯，比如有人撞了一下其中一盏交通信号灯，现在它指向了错误的方向，或者交通信号灯坏了，还有人不遵守交通法规，因为他们就是在闯红灯，或者他们对交通信号灯感到困惑。

所以你会遇到这些复杂得多、解决起来困难得多的情况。

所以，然后就像我之前说的，事实上，随着系统变得更好，要判断哪个 AI 模型更好就会变得更难。就像，就像，因为现在，你知道，比如，好吧，两次干预之间已经相隔数千英里了。

我们怎样才能尽快判断哪个 AI 模型更好？而且当，当你制作这些不同的 AI 模型时，它们，它们显然并不是超级确定性的。所以，因为某个新模型消除了一个问题，却造成了另一个问题。所以。

我们正试图通过结合模拟、上传模型并让它们以影子模式运行来解决这个问题。所以，并非每个人都有完全自动驾驶其实还有点帮助，因为我们可以看到，我们可以让它以影子模式运行，并看看这个新模型与用户的操作相比会怎么做？所以，既然我们有，你知道，数百万辆汽车可以用来做这件事，这就能让我们得到 AI 模型预测会采取的操作与用户会采取的操作之间的差值。

如果你把它们之间的误差大致加总起来，就能看出，哦，把这个模型和那个模型上传到车辆后，这个模型累积的误差更大。我们把每个模型都上传到了 100,000 辆汽车上。

但这就是目前最大的限制因素。不是训练，也不是数据。实际上是测试 AI 模型，然后想出巧妙的办法来判断新模型是否更好。比如我们知道，有些特定的十字路口比较困难。所以我们，如果只是比如，所有模型在一条标线清晰、没有十字路口的道路上行驶都会表现得很好，它们在那种情况下都会表现得很好。

但是，但是它们在复杂的……也会表现得好吗。所以，所以我们已经在美国找出了数千个十字路口，我们会，我们实际上会用新模型，让汽车在那些，那些非常困难的十字路口进行尝试。比如你知道，那个挺有名的，比如查克左转。

是的，

所以，如果你想象一下查克左转。嗨，查克。

我想知道邻居们怎么想。

但全国还有许多这样复杂的交通状况。所以我们已经找出了数千处这样的地点，然后派车前往这些地点来比较模型。

然后正如我提到的，我们还没有释放硬件 4 的任何能力，这会在今年晚些时候进行。

所以我认为，两次干预之间的行驶里程将继续呈指数级改善，而且还会加速。

埃隆，我想知道，既然你曾

把Tesla称为基本上是一个带轮子的机器人。所以我有点想知道，随着这个、随着这个逻辑转移到Optimus上，Tesla会使用什么样的

任何构建在

当前Tesla机器人能力之上的大型语言模型吗？还有，如果会的话，Optimus会有某种个性吗？这会是最终用户能够控制的东西吗？所有这些问题的答案都是肯定的。

全都是肯定的。

所以，我们正走向一个疯狂的未来。我是说，疯狂、疯狂、疯狂。

你知道，汽车将会自动驾驶，人们会拥有非常有帮助的人形助手机器人。

而且，是的，你将能够定制个性、定制声音，并且，而且机器人确实也会逐渐了解你，知道你的偏好。你甚至不必说出口，因为我们会有点理解你。我认为会发展到这样一个阶段：没有一个机器人伙伴反而会显得奇怪，就像，你知道，有点像没有、没有你自己的C3PO，还有，你知道，R2D2。

所以，是的。

嗨，埃伦，我是马克斯，超级粉丝，谢谢你接待我们。我想问你，你怎么、你认为，把Optimus部署到不同类型的工厂、行业中，需要多少整合工作，比如独特的技术人员，或者只是让它们去观察？你对此有何预想？嗯，我认为，如果快进几年，它确实会成为一种相当通用的机器人。比如说从现在起2年后，比如明年它会，就像今年它仍然相当特定，你知道，针对特定任务的训练。

所以，要执行一项特定任务，我们必须进行相当多的训练来解决那项任务。

但随着时间推移，我认为你会逐渐积累这些训练基元，比如说，像行走就是一个训练基元。

拿起某样东西、识别一个物体并抓住它，会是一个基元。然后这些基元会泛化到任何事物上。所以你可以说，拿起一个杯子，拿起一件衬衫，把衬衫叠好。

我认为这些事情会相当快地实现泛化。所以我预计可能不是明年。但到2026年，它确实会成为这样一种东西：你可以让它执行相当广泛的任务。广泛的任务，包括它以前没有专门做过完全相同事情的任务，我猜大概是2026年。

没错。你是否预见会有一个更便宜的版本，用于那些不需要比如行走之类能力的任务，或者说，这样硬件就会简单得多或便宜得多？

嗯，我们将专注于规模化制造一款出色的人形机器人，真正实现规模经济，并以非常高的产量制造这种人形机器人。如果你真的很擅长制造，就能把几乎任何东西的成本降到其构成要素的水平，本质上就是重量。如果只看原材料的重量乘以这些原材料的价值，不包括电路板，你确实可以说，如果它重量更轻，成本就会更低。

但我要说的是，对我们来说，让Optimus实现非常高的产量，要比拥有种类繁多但产量较低的机器人更好。

重新批准你的薪酬方案所带来的最棒的事情之一，就是听到

现任和前任员工的所有正面反馈，关于

你为Tesla所作的贡献。

万一发生不幸，你被一辆没有全自动驾驶的汽车撞了，

你能否花几分钟或者

几秒钟，夸夸他们，关于

你的董事会和你的其他领导层，以向我们保证，即使没有

你，Tesla依然拥有一个价值30万亿美元的不可思议的未来。

嗯，我认为没有我，Tesla也有一个美好的未来。我认为我是推动那个未来的一个有益加速因素。

而对于技术来说，你是否领先非常重要。所以，像是，重要的不是创新，而是创新的速度。

你知道，你能否用1年完成其他公司用2年或3年完成的事情？

所以，可能我最大、最有帮助的作用，就是充当创新的加速因素。

而且，你知道，只要有足够的时间，那些正在做人形机器人的各家初创公司最终都会成功。其中会有一家成功。所以真正重要的是，我们能否比其他所有人快得多？我们的产品能否比他们的提前几年完成，并且做得更好？

因为如果不是这样，那么，你知道，他们将获得那样的估值，而不是Tesla。他们将制造出有价值的产品，而不是Tesla。所以，你知道，我认为我可能最有用的地方，就是加快技术和创新的步伐，并实现规模化。

是的，

基本上就是这些。但现有形式的销售、员工、Tesla 董事会，还有你们大家给予的支持和帮助，确实让我深受鼓舞，也令我非常感激。你知道，我的心离开了你们。

你好，伊莱恩。

杰西卡，我的名字是，而且我已经

在南得克萨斯州全职住在房车里大约3年。

好的。

我只是想感谢你的星舰，感谢星舰项目以及你在那里所做的一切。

除了 Tesla，你还在 SpaceX 之类的地方工作吗？

不。

好的。

好的。

但你住在南得克萨斯州的一辆房车里？

是的。

好的。

你好。

在我住的房车营地，如果没有牵引拖车，他们完全不允许 Tesla 或电动汽车进行车内露营。我觉得这方面存在巨大的机会。另外，举个例子，对于那些开着 Tesla 进城观看星舰发射的人，如果他们能在某个地方停一夜并进行涓流充电。

好的。

也许还能使用，你知道，卫生间设施，或只是类似的简单设施。这只是我想提出来的一件事，因为我，我知道很多人来到营地后都很失望。他们不能这么说。

好的，明白了。只是，只是一个想法。我会问问，看是不是有某个地方可以允许人们待在自己的车里，而在 Tesla 里，你确实经常可以相当舒适地待在车里，而且。是的。还可以充电。

好吧。

谢谢。好吧。我想，我觉得这必须是最后一个问题了。或者也许是最后2个。对。好吧，那么最后2个问题，你和2位先生，还有。

嘿，埃隆。

我叫斯蒂芬。我由衷感谢你为人类所做的一切，而关注你的公司和你所做的事情，是有史以来最令人激动、最能激发希望的一部电影。所以非常感谢你，也谢谢你。是的，谢谢。今天我的问题是关于锂精炼的。你过去曾表示，这在未来会成为一个限制因素。我知道 Tesla 已经在得克萨斯州建造了精炼厂。我只是有点好奇你目前对精炼的看法。

你认为即将投产的产能足够吗？还是 Tesla 在这方面需要扩展到得克萨斯州的10以外？

如果把中国算在内，锂精炼产能是足够的；如果，如果在这方面应该，如果存在地缘政治挑战，那么锂可能会出现短缺，但中国非常擅长精炼，所以。

但我认为，仅仅是 Tesla 弄清楚锂精炼、了解需要做什么，然后，你知道，一旦我们了解它如何运作，就可以从那里扩大规模，而且这，我认为，你知道，谨慎应对地缘政治风险，并确保即使在最糟糕的情况下 Tesla 也没问题，是相当重要的。好了，谢谢。

这是最后一个问题。

嗨，Lon。谢谢你。祝贺你再次取得重大胜利。我是 Nikhil，我常驻硅谷。正如你提到的，要提高人们对 FSD 的认识并赋予人们权利，因为大多数人除非真正体验过，否则并不十分了解它正变得多么安全、多么便利。所以我创办了 Tesme 公司。这是一个供 Tesla 车主提供 FSD 学习体验的平台，我们很希望探讨如何在你的支持和指导下帮助扩大规模。

我们也很希望进行后续交流，与你讨论更多有关我们可以如何合作的细节。谢谢。

好的，当然。

好的，当然，当然，当然，当然。我真的想提问。

请跟进一下。

我不，我不太确定具体该如何回答那个问题。但我的意思是，我们确实，我们确实专注于改进 FSD，而且我认为，当它真正好用到不可思议、驾驶得像一位五星 Uber 司机时，它就会得到广泛采用。我认为采用速度会，会非常、非常快。

好的，我确实有一个非常好的建议。

好的，是什么建议？当然，

我把这个建议告诉了你们 FSC 团队中的一员，但我不知道他们是否把它传达给了你。基本上，随着 FSC 不断改进，我们不太确定你们将以怎样的速度推进边界线。因此，关于你们究竟何时能运营机器人出租车，滞后依然存在。于是我想到，在受监督的 FSD 和机器人出租车之间，存在一个可以商业化的中间阶段，也就是我所说的机器人租赁汽车。

好的。

这个想法是，你会，人们会到你们的平台上订一辆租赁汽车，而你们会在夜间把它送到他们的车道前。然后，汽车唯一会在无人陪同的情况下行驶的时候，就是夜间没有人的时候。你们想花多长时间都可以，然后你们把它放在车的前面，此后它就处于受监督状态。所以这会让你们更早实现商业化，也更有保障。

所以这就是我的建议。

好的，嗯，我很感谢你的这个想法。

我认为，正如我所说，FSD 的进展会非常惊人。你会看到，每推出一个新版本，正如我所说，理解 FSD 的最佳方式是忽略 12，只看那个点号数字从版本 3 变成 4、5、6，你会看到每个后续版本都有彻底的改进。你知道，而且我的意思是，如果目前的趋势持续下去，情况将会是，你真的会发现自己在一年的时间里都不需要干预之类的。

所以即使在复杂情况下也是如此，谢谢大家的支持。

所以。

所以谢谢大家。

好了，好了，Tesla 年度股东大会到此结束。

我们会想办法，别走，他们

全都非常有趣。我们来做些体育博彩。今天有冰球比赛，明天有 NBA 总决赛。

如果你愿意玩一点体育博彩，就留下来吧。我们会玩大约 5、10

分钟，同时我会下注一笔现实中的赌注。

等这里结束，我们会继续那个。

当然，不用谢聊天框。

好的，很好。

我这就开始，就像我说的，请来点胆量。如果这是你感兴趣的内容类型，我们可以在聊天中的那个里刷一个 1。

好的，很好。所以你们现在应该能看到我的屏幕了。

其实，让我这样做，让我那样做，让我确保这一切都是全屏的。

好的，所以我就快速下一注，如果你们感兴趣，可以留下来，从中找点乐子。顺便说一句，刚才大约持续了 2 个半小时，不过你是在找聊天框吗？我们来在 Underdog Fantasy 上快速下几注。显然，我们是一家财经和新闻媒体公司，但我们也喜欢，你知道，冒险押上一点钱。而且显然，这不是财务建议，不过顺便说一句，我想我上次在直播中这么做时，当着你们所、所有人，或者说，你知道，当时在这里的各位的面下了这一注，而我们赢了。

这是我下的一注。这是一张投注单。是 Al Horford，低于 9.5。他拿了 8 分。昨晚，是 Stanton，高于 7 个梦幻积分。他拿了 19 分；而这位我不会念名字的先生，小鸟球数低于 3 个，他打出了 0 个。所以看到这个还挺棒的。让我。

把那个粘贴上。在那里弄。

好的。好的，我们来看看能不能再做一张投注单。

很多次。开始吧。我们选了他。乔·瑞安。

他们预计他会打出一场精彩的比赛。我不知道自己是否非常相信这一切，但我们要押乔·瑞安低于这个数。

我们来押她的篮板数低于7.5个，这里也押低于。我准备冒险押50美元。下注。确认。

我们看看是否会确认。好了。太棒了。很好。

把这个记到我的投注追踪器里，因为就像股票一样，如果我要下一个贝塔，我就想追踪它，看看自己做得怎么样。我开了这个账户。顺便说一下，对于那些不熟悉我的人，我在 Benzinga 交易股票和期权，所以你们可以，你们知道，关注我们。我们也有一些很棒的交易竞赛活动正在进行。那应该会非常有趣。但我会追踪自己的投注。我会确保，你们知道，我正按进度走向获胜，到目前为止，进展不错。

我开这个账户时有，我想，1000，我当时有的，对吧。所以让我把这个叉掉。我有1000美元。你们可以看到。我刚刚下了50美元的注。我可以进入实时。选他们。我可以进入结果。不好意思。选他们。我不是每次都赢，但也算比较接近。几天前我赢了这一笔。输了、输了、输了、输了1笔。我们看看情况会怎样。只要我，你们知道，每6次赢1次，我觉得自己就做得还不错。

但基本上，我喜欢用来大致预测的方法，就是看看体育博彩公司认为某个人会得多少分。所以我想刚才是，它去哪儿了？去哪儿了？去哪儿了？去哪儿了？

所以我给你们举个例子，好吗？

塞韦里诺，他，我押了，我们会选他的低于30.5。其他所有人，市场大致把它定在28.5左右。所以我们有一点，2、3分的优势，市场在说，嘿，看，28。我们这边是30.5。这就是我们想继续下注的，再说一次，你们知道，要负责任地博彩，确保你们明智地对待自己的行为。但帮我个忙，帮我个忙。这是我想要的。

我想做这个节目。我觉得你们看不到我的脸。让我，让我给你们看看我的脸。

等一下。

好了。你们好。

你好。

好了，这是我想做的。顺便说一下，我们希望你们喜欢刚才的 Tesla 直播。我们会尽己所能，确保你们在金融市场方面得到知识，从而能用自己辛苦赚来的钱作出知情决定。我们也想提供一些体育博彩内容，对吧？橄榄球赛季就要来了。梦幻体育会非常火爆，而且还有大学篮球。然后当然会过渡到10月，到时 MLB 棒球季后赛也会在那里举行。

而且我们想，我们想做更多体育博彩内容。但我需要你们这样做来帮我，这样我就能去告诉，这样我就能去告诉老板们。这是我们应该做的内容类型。如果你们喜欢这种体育博彩类内容，可以直接在聊天区刷屏发送 underdog。你们也可以给我发电子邮件。如果你们还想要任何其他类型的内容。我也会把我的，把我的电子邮件地址发在那里。

但请告诉我们，这是不是你们想要的体育博彩内容类型。我已经下了那一注。也许我还会再下一注。也可能不会再下。看看吧。亚特兰大梦想队今晚有比赛。

那里押低于。

也有一些电子竞技比赛。不过我不知道自己是否想押那个。不过，是啊，为什么不呢？来看看爆头数高于。我们就押那里高于，然后再押这个。好了，那么听着，那些是。你想在这一项押低于吗。

不过它们是在同一个东西上的，所以我不知道那要怎么操作。

我不知道我喜不喜欢这个。

我不知道我喜不喜欢这个。

看看吧。

好了。

我想我就要。因为你知道，类似于股市，我想仅仅为了下注或交易而强行去做。所以我们就用这张投注单吧。就这样。我们刚刚现场下了一笔50美元的投注。

如果大家有任何问题，如果你们有任何类型的内容，请告诉我们。我们正在推出。我们暂时还不能完整公布，但先稍微预告一下：我们有一个交易竞赛，希望能在接下来几周内公布，你将有机会赢得最高30,000美元。它会像是一个为期3个月的，它会是3场各为期1个月的竞赛，总价格池为30,000美元。所以那会是有趣、令人兴奋的事情。

顺便说一下，那是现金。

而且你知道，我们总是一些 Benzinga Pro，以及这方面的所有内容。我们有很多交易节目。我们8点开始盘前准备，然后从9点一直直播交易到12点。说实话，太酷了。如果大家有任何想法、想看的任何类型的内容，请给我们发邮件。不过投注单就是这样。如果大家跟单，祝你们好运。如果不跟，也没关系。反正我可能会输，不过祝你们好运。

希望大家喜欢这场直播。我们下次见。

再见。

Sa.

Sam.

Welcome to Texas home.

The home of tesla.

My name is Brandon Earhart. I'm the General Counsel and Corporate Secretary. And this meeting is going to be in two parts. First, we're going to cover the 12 items that stockholders have been asked to vote on, as well as any other matters that have been properly presented. Second, our founder and CEO Elon Musk will provide a presentation.

During the course of the following sessions, we may discuss our business outlook and make other forward looking statements. Such statements are predictions based on current expectations. Actual events or results could materially differ due to a number of risks and uncertainties, including those disclosed in our most recent 10k or 10q as filed with the SEC. Such forward looking statements represent our views as of today, should should not be relied on hereafter and we disclaim any obligation to update them after today.

At this time, I'd like to thank members of the Tesla team and the board that are here in attendance. A representative from PricewaterhouseCoopers, Tesla's independent auditor is also present. But before we begin, I'd like to introduce you to our board chair, Robin Denholm, who would like to say a few words.

Thanks Brandon. And hello everyone. As chair of Tessa's board, I am honored to welcome you all to our home, Texas for our annual for our annual 2024 annual shareholder meeting. And when I say I'm honored to welcome you, our shareholders, I truly am. No other company in the world has such a diverse, engaged and supportive set of shareholders as Tesla. No other shareholder base understand its company like you do, nor is as committed to Tesla's mission to accelerate the world's transition to sustainable energy like you are.

It is a shareholder base whose interests the board are proud to represent. It is also a shareholder base that understands that in order to accelerate the world's transition to sustainable energy and a sustainable energy economy, Tesla needs to develop the most revolutionary technologies. Not only in autos, but in energy and artificial intelligence. And as well, our achievements this past year are steps towards fulfilling our mission.

Some highlights include the Model Y becoming the best selling vehicle in the world in 2023.

Beginning deliveries of the much anticipated Cybertruck, proving that the muscle and utility of a pickup truck truck can be electric and cool. And our energy storage deployments reaching an all time high in 2023, with the deployment trend continuing in 2024. By using our products, Tesla customers avoided releasing over 20 million metric tons of greenhouse gas emissions in 2023. That is the equivalent of of 51 billion miles driven by ICE vehicles.

An increase of about 18 billion avoided ICE miles since 2022. But to continue the exponential growth in greenhouse gas emission avoidance, we not only have to get our products into as many customers hands as possible, but we also need to increase the utilization of our products. And to help further with that goal, we released FSD supervised demonstrating that our approach to autonomy is scalable and that accomplishments in artificial intelligence, along with leveraging our existing production lines to get our next generation products to market faster are essential to achieving our mission.

We have made and plan to make further investments in hardware and software ecosystem to further advance our path to autonomy and set a course to realization of Robo Taxis. But sustainability is not just about avoiding emissions. It needs to take into account the whole life cycle from the manufacturing of the products to the products end of life. That is why we source responsibly, we design our manufacturing processes to use less water per vehicle than the industry average.

And for the third year in a row, we power our supercharger network with 100% renewables energy.

And we are also working to build a battery material recycling ecosystem both internally and with others to ensure that we are moving towards a truly circular economy for energy storage. In 2023 alone, we supported the recovery of enough battery materials to build 43,000 new rear wheel drive Model Y's. That result is the epitome of a sustainable economy. While sustainability is our mission, we also have the goal of safety.

Our cars are engineered to be some of the safest in the world and we take real world data from our fleet and our ability to push over the air updates to optimise safety for all our vehicle models. Our attention to autonomy also plays a role in our goal of safety. Based on 2023 data, Tesla vehicles with driver assist technologies engaged have only a one vehicle accident per 5.64 million miles driven compared to the U.S.

i think it's worth clarifying.

Compared to the US vehicle fleet of one accident per 670,000 miles driven. So not only are you protecting the environment by driving a Tesla, you're also protecting yourself and those who ride with you. With another year of progress in furthering our mission under our belt, I want to take this moment to thank our phenomenal Tesla employees whose creativity, hard work and dedication are the source of our success.

As huge thanks and as most as most are also shareholders alongside each and every one of you, I hope each employee takes pride in the role that you have played in Tesla's accomplishments. I am proud of you and the board and I appreciate you tremendously. Finally, I Want to remind our shareholders that your board is listening and we are here to represent your interests. You are. Yes.

You are the owners of the company and your voices, your perspectives and your input are critical to us as a board and to the direction of the company. My fellow directors and I serve as your fiduciaries and oversee Tesla's mission, purpose and strategy. With those responsibilities in mind, we continuously set the highest standards for ethical behaviour, corporate citizenship and corporate governance. You don't see the day to day work of the directors, but I can assure you this is a highly engaged board and I would like to thank my fellow directors for their dedication.

Whether it be inside the boardroom, meeting with our shareholders, or working closely with our team members throughout the company. Tesla is unique. Thank you.

Tesla is unique in how active and large our retail shareholder base is. And we hope you here in this room and our shareholders joining us virtually today, continue to take advantage of our shareholder platform to enhance your engagement with Tesla. I want to give a special shout out to our small but mighty IR team and legal team and everyone involved in running this year's proxy campaign. Both inside and outside of Tesla.

There is still much to do, but it has galvanised us around the strong belief that the right things are always worth fighting for. Thank you. And now I'll hand it back. Thank you.

Now I'm going to hand it back to Brandon. Thank you.

Thank you, Robin. So I will now call the meeting to order. The time is now 3:40pm Central and I declared that the polls are now open. We have already received over the past few weeks voting proxies from our stockholders. However, if you wish to vote your shares at this time, you may do so or through the virtual meeting portal or if you want to vote in person. Ballots and ballot boxes are available at the front. Tesla's Board of Directors has appointed Tony Cardidio, a representative of the Cardidio Group, to serve as the inspector of election.

In addition. In addition, Broadridge Financial Services has certified that starting on April 29, 2024, the proxy materials or notice of Internet availability for proxy materials were mailed or provided to all Tesla stockholders of record on April 15, 2024. We have a majority of the outstanding shares represented at the meeting. So I declare that there is a quorum present and that we may proceed with the meeting. The items on the agenda are as follows.

There are five Tesla proposals. The election of two Class 2 directors, James Murdoch and and Kimball Musk, to serve for a term of three years or until their respective successors are duly elected and qualified to approve executive compensation on a non advisory basis. To approve the redomestication of Tesla from Delaware to Texas via conversion.

To ratify the 100% performance based stock option award to Elon Musk that was previously approved by stockholders in 2018.

To ratify the appointment of PricewaterhouseCoopers LLP as Tesla's independent registered public accounting firm for the fiscal year 2024. Tesla's board has recommended that our stockholders vote for each of the two director nominees and and for each of the four other Tesla proposals. We have also received seven stockholder proposals that are described in the proxy statement. I would like to remind Tesla's board has prepared a statement in opposition to these proposals which also appear in our proxy statement.

So let's turn to those seven stockholder proposals. The first one is an advisory vote regarding reduction of director terms to one year. Our board has recommended that our stockholders vote against this. As explained in the proxy this proxy this stockholder proposal is proposed by Mr. James McRichey who has prepared a pre recorded message.

Thank you. We presented a similar proposal in 2021. 55% supported it even though the board put up a competing proposal accountability every other year instead of every year. Neither receive the required 67% of shares owned. 90% of the S&P 500 have a declassified board. Most institutional shareholders will again support our proposal. Join them. The recent Delaware court case identified three legal deficiencies at Tesla.

Directors were not independent, the facts of the pay plan presented to investors were misleading and the board failed to negotiate properly. A major part of the cure is declassifying the board so shareholders can hold directors accountable or submit their own nominations. Matt Moscardi of Free Float calls proposal number six the most important vote today since it may enable investors to change the board, bringing new skills and professionalism.

John Chavedin's simple majority vote is the other crucial vote. Another important endorser is Nel Minow, co founder of companies including Proxy Advisor iss. The Musk premium is already eroding. Tesla is Musk's liquid piggy bank since it's publicly traded. His other companies are not. Either he sticks around long enough to use our shareholder capital to fund his other ventures or he shifts his attention sooner if we reject his pay package and turn off the money.

Tap. Bite the bullet. Prepare for a future that doesn't depend on a superhero dictatorship. We need to transition from a board dominated by friends and family to one of expert independent directors in engineering, software and other relevant fields. Musk is A genius. I'm saddened to see him focus his attention elsewhere, but it is happening. Don't let him siphon off more from Tesla. We need to transition to a more democratic Tesla that can attract customers, employees and shareholders seeking a better world ahead.

Again, vote for number six and seven. Declassify the board for accountability. Lower the voting threshold for more flexibility. Vote now before the polls are closed. Your vote can make a significant difference in our future. Thank you.

Thank you. Thank you, Mr. McCritchie. As a reminder, the board's response was contained in our proxy statement. The second stockholder proposal is an advisory vote regarding simple majority provisions in our governing documents. Our board has recommended against and aren't set forth for the reasons in the proxy. This stockholder proposal is proposed by Mr. John Chivedin, who is on the line to present this proposal. Mr.

Chavetin, I would like to invite you to speak. You will have three minutes.

Three minutes.

Hello, this is John Chavedin. Proposal seven, simple majority vote. Charles requested the Board of Directors take the necessary steps so that each voting requirement in the charter and bylaws that calls for a greater than civil majority vote being replaced by a requirement for a majority of the votes cast forward against such proposals for a simple majority. This includes making the necessary changes. In plain English, this proposal 7 is the exact same concept that the Tesla Board of Directors supported at the 2022 Tesla annual meeting.

The 2022 Tesla board of directors proposal then won 98% support from the Tesla shareholders who voted. And the Board of Directors version of this proposal would have been adopted earlier if the Board of Directors had only made a slight effort to encourage more Tesla shareholders to vote for their proposal. This simple majority votes standard proposal will help improve Tesla shareholder rights. Tesla now scores a dismal 10 in shareholder rights, with 10 being the worst possible score.

The Tesla Board of Directors supported this proposal concept in 2022 and should do so today. Shoulder rights and a civil majority vote standard are all the more important when Tesla has a CEO, Elon Musk, who is threatening to give less attention to Tesla and is threatening to withhold advanced technology from Tesla unless he gets a $56 billion paycheck. Never before have sharers been under such duress when asked to prove an outrageous $56 billion CEO paycheck.

Plus, there is no promise with the $56 billion paycheck that Mr. Musk will devote any minimum percentage of his time to Tesla or pursue any minimum amount of advanced technology. At Tresla, Russ Gerber of the investment firm Gerber Kawasaki said. I believe the Tesla Board of Directors is the most conflicted, least independent and most incapable board of directors in the history of business. Please vote. Yes. Simple majority vote.

Proposal 7.

Thank you Mr. Chavadin. As a reminder our the Board's response is set forth in the proxy. Our third stockholder proposal is an advisory vote regarding annual reporting on anti harassment and discrimination efforts. Our board has recommended against for the reasons set forth in the proxy. This stockholder proposal is proposed by the New York State Common Retirement Fund as the lead filer which has requested the Dr. Kristen hall as represented for Neo Impact Capital as co filer for this proposal present this proposal.

Dr. Hull, I would like to invite you to speak. You have three minutes.

Thank you. Hello, I'm Dr. Kristen Hull of NEA Impact Capital. I'm here on behalf of lead filer New York State Comptroller Thomas D. Paoli, Trustee of New York State Retirement Fund, owner of Tesla shares valued at over 650 million as of March of this year. And I'm here as a co filer of Proposal 8. The fund is a significant long term owner of Tesla and has an explicit focus on investing in climate solutions. Longview Large Cap Fund through its Trustee Amalgamated Bank, AP7 and Arjuna join us in filing this resolution.

Our firm name NIA means intention and purpose. We invest only in those companies we see building toward a better world, a sustainable and inclusive future. As the founder of nia, I had been proud in our early ownership of Tesla, I saw the potential and the need for innovation in the automotive and the clean energy market space. However, when looking under the frunk, we find failures in workplace management one of the most important components of any business.

Employees need to be hired, trained, motivated and engaged to do high quality work and we need those employees to be able to contribute to their top ability. To do so, they must feel both physically and emotionally safe at work. Investors are requesting the report outlined in Proposal 8. Because workplace abuse, harassment and discrimination have no place at Tesla. They can result in substantial costs including employee turnover, increased absenteeism and reduced productivity, as well as legal costs, all of which can impact shareholders investments in the company.

They may also lead to difficulties in recruiting high quality employees, something that Tesla needs to remain innovative and at the forefront of a quickly moving technological landscape. We have expressed concerns about Tesla's workplace for years and year after year more harassment and discrimination allegations have emerged with few visible improvements in Tesla's people management policies or practices. To assuage our concerns.

Lawsuits and class actions before Tesla now involved thousands of employees with allegations including repeated use of racial epithets by managers and employees, segregation and discrimination against non white employees in job assignments, pay and promotion, and retaliation when employees report their experiences. These concerns are why I now formally move proposal 8 requesting that Tesla's board commission an independent report on the effectiveness of the company's efforts to prevent harassment and discrimination within this revolutionary and innovative workspace.

Thank you.

Thank you, Dr. Hall. The board's response is set forth in the proxy materials.

Let's turn to the fourth stockholder proposal which is an advisory vote regarding the adoption of freedom of association and collective bargaining policy. Our board has recommended against for the reasons set forth in the proxy. This stockholder proposal is proposed by SOC Investment Group whose representative Tezel Patel has prepared a prerecorded message.

Hello, I'm Tejal Patel, the Executive Director of the SOC Investment Group and I Hereby move proposal 9, a shareholder proposal requesting that Tesla Board of Directors adopt and disclose a non interference and collective bargaining policy that upholds international labor rights standards in its operations. This resolution was submitted with other CO filers and we are concerned that Tesla's existing policy fails to clearly explain how conflicts between international and local labor law will be resolved.

And because of clear indications that Tesla is not adhering to its current policy despite the operational, legal, human capital management and reputational risks involved. We recognize that Tesla revised its global human rights policy in April 2023 to explicitly refer to the rights of workers to free association and collective bargaining. But these revisions were fatally compromised. The statement committing to respect these rights includes the qualifier in conformance with local law rather than referencing the often higher international human rights standards suggesting that Tesla respects these rights only as explicitly required by local laws.

Numerous signs suggest that Tesla is not respecting its workers fundamental rights. In Sweden, Tesla has faced months long strikes and boycotts over what has been characterized as a refusal to bargain. Moreover, as part of its reporting on the situation, Reuters wrote that Tesla has a policy of not signing collective agreements which if true, would directly contravene the company's current policy. In Germany, Tesla faces allegations of refusal to bargain and poor working conditions.

In the United States, Tesla has had adverse decisions by the full National Labor Relations board and the U.S. 5th Circuit Court of Appeals over labor rights violations. Just last month the NLRB issued a complaint alleging that restrictions in Tesla's workplace technology policy served to unlawfully discourage employees from joining unions or acting concertedly.

It is also relevant for shareholders to consider the broader human capital management risks Tesla faces, including reports of higher health and safety violations and injury rates than peers and numerous lawsuits alleging of hostile work environments. Freedom of association and collective bargaining are considered enabling rights, meaning that they empower workers to effectively advocate for additional labor rights, including the right to a safe and healthy workplace free from discrimination.

The adoption of a comprehensive labor rights policy that clearly articulates a commitment to non interference, good faith collective bargaining and adherence to higher international standards would help Tesla better mitigate these human capital management risks as well as operational and reputational risks like we are seeing in Sweden. Therefore, we urge you to vote for item nine. Thank you.

Thank you, Ms. Patel. The board's response is set forth in the proxy. The fifth stockholder proposal is an advisory vote regarding reporting on effects and risks associated with electromagnetic radiation and wireless technologies. Our board has recommended against for the reason set forth in the proxy. This stockholder proposal is proposed by Landreau purcell who's represented. Dr. Deborah Davis has prepared a prerecorded message.

I am Deborah Davis, founder of Environmental Health Trust. I am speaking with you today on behalf of shareholder Landry Purcell regarding her proposal seeking an annual report on the risks for magnetic fields and wireless radiation in their products. I'm also speaking on behalf of millions of young children, infants and toddlers and pre women who are more vulnerable to wireless radiation, according to the American Academy of Pediatrics.

I'm the founding director of the Board on Environmental Studies and Toxicology from the National Academy of Sciences and I published more than 220 scientific articles on environmental health and have concluded, along with hundreds of other experts that current exposures to non ionizing electromagnetic fields called EMF constitute a serious and unrecognized public health risk. Tesla vehicles operate with a 1,200 pound battery that emits magnetic fields and also uses wireless equipment such as WI fi and radar emitting wireless radiation.

Tesla should provide transparent information on measured exposures and efforts to reduce them. The World Health Organization has concluded that both low levels of EMF and microwave or cellular radiation are are quote, possible carcinogens, the same category as some pesticides and gasoline. The companies that provide secondary insurance, like Lloyd's of London and Swiss Re, have long considered electromagnetic fields to be uninsurable, comparable to asbestos due to potential health and environmental damages.

Tesla shareholders have a right to know how the company insures itself against such risks. The fact that current levels may be legal does not mean they are safe. In fact, US regulations were set more than a quarter century ago. It has been three years since the U.S. circuit Court of Appeals for the District of Columbia determined that the FCC had been, quote, arbitrary and capricious in its failure to consider all of the science and refusal to update wireless limits, Tesla has failed to inform consumers about the measured level of magnetic fields and wireless radiation within the automobile or or their other products.

Infants and toddlers by law are strapped into seats to protect them during crashes, but they have no protection from EMF that can go more deeply into their thinner skulls. To mitigate risks for Tesla, the company could shield its vehicles and ensure wireless antennas could be easily turned off. Hardware and software changes can lower exposures. Generally, Tesla has developed extremely rigorous compliance and hazard screenings beyond regulatory requirements for certain toxic chemicals.

Investors can encourage the company to take a similar leadership role regarding reducing emf. It's time for the company to unambiguously compete on safety regarding wireless radiation and non ionizing radiation emitted by its products. Thank you.

Thank you, Dr. Davis. The board's response is set forth in the proxy. Let's move on to the sixth stockholder proposal, which is an advisory vote regarding adopting targets and reporting on metrics to assess the feasibility of integrating sustainability metrics into senior executive compensation plans. Our board has recommended against for the reason set forth in the proxy. This stockholder proposal is proposed by Tulip Share Securities Ltd.

Whose representative, Constant Ricketts, has prepared a pre recorded message.

Good afternoon fellow shareholders. Thank you for your time. I'm Constance Ricketts and today I'm speaking on behalf of Tulipshare in support of Proposal 11 urging Tesla to integrate sustainability metrics, including diversity among senior executives, into executive compensation plans. This proposal calls for robust quantitative targets and transparent reporting to both resolve executive pay plan controversies and ensure our company's long term stability and success.

Diversity and independence in leadership are essential for innovation and enhanced decision making. While sustainability metrics in executive compensation reflect alignment with regulatory pressures, investor expectations and responsible business practices. Despite 75% of S&P 500 companies incorporating sustainability metrics into executive pay plans, Tesla fails to do so. This is especially concerning given our mission to accelerate the world's transition to renewable energy.

The recent ruling that voided our CEOs $56 billion pay package highlighted the need for fair and negotiated compensation plans that serve stockholders best interests. Tesla's proxy statement does not offer substantive evidence of how the voided 2018 award harm stockholder democracy or how its ratification will cure prior deficiencies and lead to a different outcome. Likewise, the opposition to this proposal fails to show any evidence of an existing performance based component integrated into the executive pay structure that respects established human rights, due diligence processes and principles.

Instead, it cites its human rights policy that is wholly unrelated to executive compensation and doubles down on the argument that the currently contested executive pay structure is the only means of motivation to achieve our company's mission. This commitment to the status quo essentially dismisses the ongoing controversies that Tesla must face and overcome, including softening sales, advancing competition, split focus in leadership, a 30% drop in stock price, reduced market cap, mass layoffs, human rights violations and ongoing legal challenges.

Tesla's executives absolutely should be duly compensated for innovative and successful business. However, stockholder democracy allows us to oppose demands like 25% control of the company and the largest pay package in corporate American history. Via further share issuance. We also have the right to vote for our leadership's compensation to be commensurate with financial and sustainability performance. This does not mean we do not respect our leadership's time, energy and contributions.

Quite the contrary, as it is our duty as stockholders to ensure long term viability and success for our company.

Thank you, Ms. Ricketts. The board's response is set forth in the proxy. Let's move on to the seventh stockholder proposal which is an advisory vote regarding committing to a moratorium on sourcing minerals from deep sea mining. Our board is recommended against for the reasons set forth in the proxy. This proposal is proposed by AS you so, whose representative Elizabeth Levy has also recorded a pre recorded message.

Good afternoon Chairman, members of the Board, My name is Elizabeth Levy, representing as you sow, speaking on proposal number 12. We know more about the surface of Mars than we do about the bottom of the ocean. Even so, a new industry is moving forward to exploit the deep sea. We believe this industry poses risks to Tesla now and into the future. Deep sea mining is the process of dredging the ocean floor to extract battery related minerals.

It will indiscriminately destroy habitats and obliterate life in its path. Seabed mining could release carbon stores and toxic sediment plumes, poisoning marine food chains and disrupting sea life breeding and migration. The deep sea is complex and abundant with life. Its ecosystems are slow growing and fragile, with little to no ability to recover from outside disturbance. Plans are moving forward to launch the process for commercial scale deep sea mining with or without regulation.

Tesla's membership in the OECD and IRMA mining protocols does not resolve this issue as neither currently applies to mining in the deep sea. Over 800 marine science and policy experts, 25 countries, 72 indigenous groups and 48 companies are calling for a ban, moratorium or precautionary pause on deep sea mining. This includes Tesla's competitors, BMW, Renault, Rivian, Volkswagen and Volvo. Even General Motors has stated it does not plan to use deep sea mine minerals.

Our company's public stance on deep sea mining matters. Tesla is the face of the EV transition and its ambiguous position on deep sea mining could drive investment in this controversial industry. Conversely, a clear stance restricting Tesla's use of deep sea minerals can drive innovation in battery mineral tech, recycling and circularity. With these technologies, even the most ambitious future EVU scenarios could be met without harming the deep sea.

Tesla's failure to take a stand on deep sea mining, combined with the fact that a former Tesla board director is now serving as a board member and special advisor to the CEO of the metals company, the very company leading the charge on deep sea mining raises concerns. Such a relationship appears to imply that Tesla supports the metals company in its deep sea mining activities. Committing to a moratorium will clarify expectations and reduce reputational harm to the company, assuring customers and investors that Tesla will indeed wait until science shows deep sea mining will not cause significant or irreparable damage.

Thank you.

Thank you, Ms. Levy. The board's response is set forth in the proxy. Please note that now is the final opportunity to submit any votes in order for them to be counted.

I declare that the polls are now closed.

Based on the proxies that we have received and the votes submitted at this meeting, I'd like to announce the results on a preliminary basis. With respect to the seven stockholder proposals, our stockholders have approved the recommendations of the Tesla board on all of them, except for Proposal 6 regarding director terms and Proposal 7 regarding simple majority voting.

Now moving on to the five Tesla proposals.

First, our stockholders have approved the election of our two Class 2 directors, James Murdoch and Kimball Musk, to serve for a term of three years and until their respective successors are duly elected and qualified.

Second, our stockholders have approved executive compensation on a non binding advisory basis.

Third, our stockholders have approved the ratification of appointment of PricewaterhouseCoopers LLP as Tesla's independent registered public accounting firm for the 2024 fiscal year.

Fourth, our stockholders have approved the redomestication of Tesla from Delaware to Texas.

Fifth.

And finally, our stockholders have approved the ratification of the 100% performance based stock option award to Elon Musk that was approved by stockholders in 2018.

The meeting is now closed. And with that, Tesla shareholders, let's welcome Elon Musk.

The goal is to give people hope that there is a path to a fully sustainable global economy. That we are on that path, that we are accelerating that path and that so long as we don't get complacent about it, it will happen.

This was really the beginning of the

end of the Tesla bubble. I actually think the company could go bust.

Tesla's Model Y is the world's best selling car, beating out Toyota's RAV4 and its Corolla model.

Regarding FSD version 12, it's profound. The rate of improvement is rapid. It might be the biggest asset value appreciation in history. When you can do unsupervised full self driving, that just sounds like a story stock autonomous taxi. I mean can you really balance your checkbook with, you know, sort of pie in the sky predictions like that? If you ask the wrong question, the right answer is impossible.

My prediction is that a majority of Tesla's long term value will be optimus. And that prediction I'm very confident of.

It's very rare a product comes along that is seemingly impossible, that experts said would never be made. And this is one of those times. Finally the future will look like the future. People say like, why'd you make it bulletproof? I'm like, why not?

Energy storage deployments, the Megapack in particular reached an all time high in Q1, leading to record profitability for the energy business. This is the machine that builds the machine and the factory is the product. And this building is the most advanced car factory that earth has ever seen.

If you value Tesla as an auto company, this is the wrong framework. If somebody doesn't believe Tesla is going to solve Autonomy, I think they should not be an investor in the company. But we will and we are.

Sam.

Hey guys,

Welcome, welcome to the Tesla shareholder meeting.

And I, I just want to start off by saying hot damn, I love you guys.

Yeah, we have the most awesome shareholder base. I mean it's just incredible. Any public company, it's incredible. Wow.

So, and you know, we've got a great sort of shareholder meeting here to recap the achievements of the company and tell you about where we're going. And it's, I think it's incredible. I think we're not just opening a new chapter for Tesla, we're starting a new book. Okay.

So those of you who've been following Tesla closely understand like, you get it, you know, and so, so like, important thing is you get it. I mean it's so the, yeah, where things are going I think is just going to be absolutely mind blowing.

You know, we're obviously making great progress in solving the sustainable energy problem, you know, with electric vehicles, with stationary storage, with solar and then I Think in terms of the value of the company. Autonomy is just such a mind blowing thing that, like I said, you guys understand, but I think most of the world does not yet understand. So.

And you know, in talking to a lot of the sort of big institutional investors, when they're often in like New York and they don't drive cars, so I'll be like, have you driven our car? Have you tried self driving, you know, the version 12.3? And they're like, oh, no, okay, well, you should try it. That would be a good thing to do. And if you just plot the points on the curve of how well autonomy is progressing and just believe the curve, it's headed towards unsupervised, full self driving very quickly at an exponential pace.

If you just, if you just. Yeah, it's, it's, I mean, yeah, so.

It's, it's, it's, it's really as simple as that. And in fact, I'd invite you to just do it personally. Just say, okay, with each release, how many miles do you drive before you have to intervene?

It's like literally that, that simple. And with each release you'll see that there's a, there's a big improvement. Big improvement. And it looks like an exponential. So. And it's very clear that that will actually go to the point where it is actually far safer than a person driving the car.

So.

And I have some insight into, you know, the next releases because we've got basically for every release that's on the road, we kind of see what the next generation release is and a little bit about the release after that. So we have some insight into where it's going. And it's really amazing.

No question whatsoever that it will far exceed human safety.

Yeah, no question.

And I think, you know, I have mentioned this, but like, sometimes it's like helpful to sort of reiterate and connect some of the dots because, so, because sometimes people wonder, well, how do you go from this big fleet to actually monetizing the fleet in an autonomous situation? And it's actually a combination of like, it's like Airbnb and Uber to some degree. So like, there'll be some cars that Tesla owns itself. It's kind of like an uber fashion.

But then for the fleet that is owned by our customers, it will be like an Airbnb thing. You can add or subtract your car to the fleet whenever you want. So you can say like, I'm going away for a week. Just one tap on your Tesla app, your car gets added to the fleet, and it just makes money for you while you're gone. Yeah.

So you can say, you can add it to the fleet for a few hours, for a few days, for a few weeks. Whenever you want it back, you say, come back, and car will come right back.

And I'm highly confident that it will far exceed the value, like the revenue made by the owner of the car will far exceed the actual monthly payment. And then Tesla will obviously take a rev share on that, but most of the money will go to the owner of the car. And this is actually going to work. This is what will happen. So you, I mean, mark my words, this is, this is simply a matter of time. Now, admittedly, I'm a little optimistic sometimes.

You know, I'm, you know, so I don't have complete lack of self awareness, but if I wasn't optimistic, this wouldn't exist. This factory wouldn't exist, you know, so, So, you know, that's what my brother, who's awesome, he's here, just got reelected. No, sorry. This thing's.

So like a little anecdote

when we

were kids in school. My brother would tell me the, the wrong time for the bus so that I wouldn't miss the bus.

And then I'd be upset with him, like, why'd you tell me the bus was going to come earlier than it was? And he's like, well, because otherwise you'd be late for the bus. This is an actual thing that would happen. So I guess I've been sort of pathologically optimistic from, you know, from birth, but that's the reason all this stuff is happening anyway. It's like I got to be somewhat pathologically optimistic, but I do deliver in the end.

That's the important thing.

So.

Yeah, and it's really just going to be something else.

I mentioned this before, but, like the. For Ark Invest, I'd recommend reading the Ark Invest analysis. That's the most accurate there may be, you know, more accurate ones, but the most accurate sort of analysis of the potential of autonomy, the most accurate that I'm aware of is, is Cathie Wood's Ark Invest. So, and if I recall correctly, they, they predicted somewhere over a $5 trillion valuation for Tesla just based on vehicle autonomy, not counting Optimus.

Yeah, Yeah, I agree with that.

So, so I think, I think, I think just based on vehicle autonomy, we can, we can 10x more than 10x the value of the company.

I believe that's what will happen. Yeah, no, it's just, what do you know, it's, it's 4:20pm Just notice that.

So yeah, So I think that's actually, I think that's pretty accurate.

But there's this. It actually gets way crazier when you think about the Optimus robot, which is really a humanoid robot that is intended to, you know, be able to do anything you want it to do, to be, you know, your companion. It can be at your house, it can sort of babysit your kids, it could teach them great teacher, you know, it can do factory stuff.

Like I think that the ultimate ratio of say, how many super useful humanoid helper droids do you want? Like, who doesn't want a C3PO, you know, you know, but a C3PO plus R2D2 plus, you know, plus plus

it

would be pretty awesome, I think, I think everyone in the world is going to want one, like literally everyone.

And, and then there'll be obviously robots in industry making stuff. And so I mean, I think the ratio of humanoid robots to humans will probably be at least two to one, something like that. One to one for sure. So which means like somewhere on the order of 10 billion humanoid robots, maybe, maybe 20 or 30. And so then it's like, okay, well let's say, you know, you kind of make, let's say the build rate is, I think the build rate will be probably something ultimately like a billion a year.

Humanoid robots, like actually.

And if Tesla just has a 10% sure of that, and it might be a lot more than 10% and there's you know, who make like 100 million optimist units a year. I just, I mean for reference, the auto industry is roughly 100 million vehicles per year. So you know, sort of similar ballpark, at least within an order of magnitude. And I think we could make one for a cost of maybe at really high scale of about $10,000. It's smaller, it'll be less expensive than a car.

So I mean I think if you sold for sell for $20,000 or something, this is at large scale volume, Tesla would basically make about a trillion dollars of profit a year from that.

So yeah, If the price earnings multiple is say, I don't know, 20 or 25, something like that, that would mean a 20 trillion dollar market cap from Optimus alone and, and probably 5 to 10 from autonomous vehicles. So like, I think it's actually conceivable it's when within the realm of possibility for Tesla to achieve a valuation 10 times that of the most valuable company today.

So when I say this is like we're starting a new book I mean, it's going to be the best book.

Next level. Next.

Next.

Next. Next level.

We need to make sure these robots are nice to us and, you know, that's very important.

So, anyway, so with that, let me get into the presentation.

Blah, blah, blah, blah, blah, blah, blah, blah.

Oh, man.

I think it might be worth us just putting a word limit on future shareholder Tesla proposals.

It's like you can't have a novel length. It's correct. We can't have a whole novel as a Tesla proposal. So as you can see, our impact is accelerating. We're starting to make a real noticeable dent in carbon emissions. So we're really, you know, making a lot of progress towards achieving our goal of sustainability. And one of the effects of autonomy is actually is going to be an even greater effect on carbon emissions because I suspect that we'll go from passenger cars having about 10 hours of usage per week, about an hour and a half per day, to probably a third of the hours in the week, maybe 50, 55 hours, which means the same car will be used five times more often.

So you need a certain amount of resources to make the car, but it'll have five times the usage. In my opinion, I think that's quite likely, which will be an even more dramatic impact on carbon emissions. So, you know, we care a lot about sustainable manufacturing. Our factories are beautiful. Like you walk around the factory, it's actually, you know, contrary to these, like what you read in the press, it's like a good vibe in the factory, like people are smiling and happy.

It's like, it's nice.

So, So, yeah, so, I mean, we do care. We actually do care a lot. It tells about doing the right thing and we, you know, we're not going to be perfect, but we do care a lot about doing the right thing.

So our vehicles are water efficient, energy efficient.

You know, we try whenever possible to use renewables for powering the factory and, you know, do our best to do the right thing.

Our batteries are lasting a lot longer. I get too many things in my hand anyway. Batteries are lasting a lot longer, which is. Which is good.

Safety is good in the factory.

We're improving the affordability of EVs. You know, model Y is comparable to BMW X3, but its actual cost of ownership on a monthly basis is much lower.

We were number one in EVs globally last year. So, yeah, So, yeah, so let's see. Congrats to the Tesla team on making 6 million vehicles all time.

So, I mean, the Tesla fleet is really becoming very substantial. I mean it's going to be 7 million vehicles by the end of this year. Over 7 million vehicles. So.

We had. And our factory in Fremont is currently the highest volume auto factory in North America. So. And we broke the prior production record from when it was new me. So it's, it's actually pretty wild that we have a giant car factory in the San Francisco Bay area. It's like it's not exactly the cheapest place to have a car factory. It's like that in the Swiss Alps or something, you know.

But we still managed to make great cars at high volume and that's a testament to the great team we've got in Fremont. So congrats to the Fremont team.

And a lot of people were saying the cybertruck is like fake, it's never going to come out, you know, and now we're shipping a lot of cybertrucks. We hit a weekly record of 1300.

And you know, I think with the cybertruck it really is something special.

Like

people sometimes have like different opinions on the cybertruck but if you really want to know if there's a, if something is cool, if it's a great product, like show it to a kid, okay? Like no filter, okay? The kid's got like no filter, like a five year old, six year old, something like that, or even three year old and say which car do you like? Cybertruck.

So it's like that's how you know, and it's finally something that it just looks like the future. It's. And it drives so well, you know, it's comfortable. It's just a fantastic product. I think it's, I think it's our best product and it's a lot of innovation. It's got a 48v low voltage system finally instead of 12v 800 volt drivetrain, the world's biggest castings, you know, it's bulletproof. All the cool things can, you know, output F350 diesel, which is a very impressive truck.

And it can, you know, it can do a quarter mile faster than a Porsche 911 while towing a Porsche 911.

So that's insane.

So yeah, If you think about it like how often do companies make products that move your heart that are really special? It's so rare and I think this is one of them. Yeah.

And we launched the upgraded 3, which is actually a fantastic car. I recommend trying it out. It's, it's really a great car. And it's only $216 after a gas Savings like total cost of ownership is basically $200 something dollars when compared to a gasoline car, similar gasoline car. So it's really a great deal and a fun car. And the performance, model three performance is it's like, you know, fast and Porsche 911, it's just a great, great car.

So yeah, so yeah.

And of course Model Y became the best selling vehicle globally and this is something that we predicted. So you know, I think I said in 2022 Model Y would be the biggest car in the world by dollar volume sales and at 2023 it would be the biggest in unit volume. And it was. And again this year it will be the best selling car on earth. So.

And we obviously we've got the Tesla Semi. You know, we're in low volume production of the Tesla semi and just last week I approved the plans for volume production of the Tesla semi.

So we're going to make a lot of semi trucks. We're going to start off with yeah, it's something, I think it'll actually move the needle financially. It's not like it's not a small, small thing. And the thing about like commercial vehicles is that companies that use them are super objective. It's not like it's actually, I mean obviously we're going to make it have cool style and be an awesome car. But for commercial vehicles the companies that make the buying decisions, they just look at it say like what are the numbers?

Like is this cost less to transport or cost more to transport than than say a diesel truck? And the thing is the economics are much better than diesel truck. It's kind of basically a no brainer. If you're a transport company, you don't use an electric, that Tesla electric semi, you're just losing money. Why would you do that? Do not like money. Okay, but if you do like money then I recommend using the Tesla semi.

And so, So I think this is, this is really going to actually sell at a scale that people will be surprised at and, and will actually move the needle financially and do a lot of good. And also for, for CO2 and, and sustainability because semis are driven all the time actually even at, at much smaller numbers they have a, a very big effect on total carbon emissions. Solved.

So obviously we've got some new, new products that we're working on under the

covers

and I think these, these are, I think these are going to be pretty special. So you know, some of them, I think people maybe at first may think oh it's not going to be that amazing but just wait it will be so, So our supercharged network is continuing to grow. You know, rumors of the death of a supercharger are greatly exaggerated.

We are in fact continuing to grow the supercharged network significantly. In fact this year I think we'll put more, deploy more superchargers this year that are actually working than the rest of industry combined, just FYI.

So.

Now we are going to be more careful about like you know, the capital efficiency of where we deploy superchargers. But for sure, any place that has congestion, any sort of missing parts of the map that we're missing, we're going to put the superchargers there. So even for the remainder of the year we expect to spend about half a billion dollars on, on supercharger deployment. So it's very significant. It'd be a well spent half a billion.

So this is, you know, it's a lot. So, and we're, this is actually not showing a bunch of new markets that will be opening up later this year and next. So we're obviously what we want Tesla to be worldwide and there's a lot of countries that we, you know, have zero to very few Teslas and we obviously, we want Tesla everywhere. So there are a lot of new markets that we're going to open up later this year and next.

We're also opening the Supercharger Superchargers to other companies with the, that's the adapter to the old somewhat, somewhat clunky connector. But our goal is to be helpful to other car companies. So any other car companies that are going electric, we do want to, we do want to be helpful to them with our supercharged network. So we're providing them with adapters and enabling them to have access to the Tesla supercharging network because we think it's better to do that than to create a walled garden that inhibits EV growth.

There's a bit of latency here and we're also innovating a lot in battery production with the 4680 that's built right here.

And it is, it's a hard problem. You know, like there are entire companies that just make battery cells. That's all they do. And, but we're making good progress. Like all the cybertrucks that you see use the Tesla 4680 cell.

And we have a clear path to the 4680 being, we think probably the most competitive cell from its, from a manufacturing efficiency standpoint. And so it's, it's, but it is, it is a hard Problem. I have to say there's, there's quite a lot of brain damage required to be good at cell manufacturing. It's like a brain damage hide.

But I think it does, it does give Tesla resilience in the face of if there are changes in the, I don't know, geopolitical situation somehow. But it's good to have some independence in cell manufacturing.

So, yeah, we're making steady progress. And we also have the cathode refinery, which you can see behind the, the main factory. So you can sort of see it in the picture there. That's the cathode refinery. And then we've got the lithium refinery in South Texas. And so we're just making sure we've kind of got the piece of the puzzle. And I mean, if you were to look at sort of a video of how Tesla does say cathode and lithium refining and how the rest of industry does it, it's night and day.

I mean you can sort of eat off the floor in the, in the Tesla refinery and I would not recommend doing so in the others.

So, you know, it's, it's really, we're definitely going very, very vertical here,

but

I think this is a, it's a wise investment that will, that will pay off more than people realize. So.

I don't think this thing's working.

So this year we're also on track to complete a massive number of energy deployments. So that is a gigantic increase in deployment capability. So we really, we seem to be tracking to sort of a 2 to 300% year over year growth in energy storage, deployment and stationary pack. So it's giant.

And the limiting factor really is being able to build more megapacks and build more Powerwalls. So we're ramping up production of the Powerwall 3, which is really a game changer for, at the personal level, Powerwall 3. It usually takes about three iterations of any given technology to really, for it to be something where it's like, okay, now we're really hitting the sweet spot. So Powerwall 3 is like an epic product.

The Megapack is also,

I'd call it

sort of iteration two of Megapack is, is also an FA product with, with the Megapack 3, which is, I don't know, probably a couple years away, we, we'll start actually absorbing more and more of the substation of the, of the, of the sort of power electronics. So because I sort of think you want to get to the point with the Megapack where you can literally just take the high voltage power Lines and plug them in. Just there's no substation.

We just plug them in. Just drop them down and drop it down and plug it in and now it works.

Which is like mind blowing for the utility industry by the way. Like they're like what? Yeah, just plug the wires in and it'll work for very high voltage. So that I think is actually going to be a major. So this is anyway, so yeah, this is, this is great, excellent work by the energy storage team.

Services and other are also looking good. So we're actually now making money on the services and other stuff.

So this is kind of a big deal, right? Yeah. So let's see.

Yeah, and Tesla's obviously way more than a car company. You know we do a lot of software at Tesla, you know about, I don't know, roughly half the or a very huge percentage of the engineering we do is actually software engineering. So Tesla's I'd say as much a software company as it is a hardware company. This is a very big deal because car companies are not software companies normally. So this, this matters a lot. So things like our order better software for energy storage, all the software that controls the cars, the megapacks, the power walls of the solar obviously for AI and, and full self driving or big deal insurance service and collision that Tesla also writes a lot of software internally that, that helps that like we call like the Tesla operating system internally that is head and shoulders above what any other company has.

I think probably better than any Fortune 500 company.

The Tesla internal software is, it's just way better. So it's, yeah, it's just far more than what people normally think of as a car company.

And Tesla is also the world, the leader in real world AI. So there's really this is a big deal like Tesla is ahead of Google Matter, you know, opening anyone on real world software actually look, taking in video and making decisions based on video.

No one can, no one is even close. And it's getting better I'd say with each passing month, if not each passing week.

So yeah.

And, and it's also worth noting that Tesla is actually pretty good at chip design. So the, the AI inference chip that's, that was designed by Tesla that's in cars. We had sort of our hardware three AI inference cars made in the, you know, over the past year have the hardware 4. We've just completed design on hardware 5 which we're now calling AI5 because it really is there still actually is not a chip from Nvidia or from.

And I have a lot of Respect for Nvidia from any company that we would prefer to put in our car that is better than what we have in the car. So, So we started from scratch in chip design, just says we started from scratch in AI software and have the best real world AI software and the best AI inference chip in the world from nothing. So this is, you know, a big deal.

And the capability of the chips in the car is dramatic. I mean right now all the cars are actually training.

We have hardware 4 run hardware 3 in emulation mode. So we'll continue to make significant progress in hardware 3. But later this year we'll actually bifurcate, continue working on hardware three, training on hardware three, but then do separate training on hardware four. That'll be the sort of training cluster that we're building at the south side of the Gigafactory. And that'll be dedicated hardware for video and inference.

And you know, so the hardware 4 has cameras that have about four or five times better resolution. And depending on how you count the sort of hardware 4, it's about anywhere from three to eight times better than hardware 3. But everything you're seeing thus far is just hardware 3. And we still have a long way to go before we reach the limits of hardware 3. So hardware 3 I think will do amazing things, but hardware 4 I think will probably do about 5 times better.

Then hardware 5, which comes out in about 18 months or so is 10 times more capable than hardware full. It's like. So if you know, Nvidia hardware, it's, it's sort of a B200 class computer.

So you know, that's like this. So what we'll just progressively do is improve how many nines of reliability the car gets.

And I, and, and then of course that will go into Optimus as well. So the same chip will go into, you know, hardware four will go into opt. Is, is an Optimus hardware AI5, which we're calling switching from hardware five to AI5 will be in Optimus and in, in all cars in about 18 months. And it's really just a staggering amount of compute. And it's very, it's very power efficient compute. So it's, it's got to be because if you're in a mobile application like a humanoid robot or a car, you can't just be sucking down 10 kilowatts, you know, like you can in a data center.

So you've got to be very power efficient.

So the, you know, sort of hardware 3 and 4. Only a few hundred watts.

Yeah, yeah. This is hard.

Now, hardware 5 will be able to go probably up to about 7 or 800 watts, but it'll power fluctuate contingent upon the complexity of the scene that it is in. So if it is in a parking lot stationary, it's like, you know, compare. You know, don't have to think very much just like a person. Like, if you're in a complicated traffic scenario, you've got to think a lot more than if you're just cruising along, you know, on an empty road.

But something that I think is potentially interesting down the road is like, at some point, the Tesla fleet, I think, will probably be, you know, over 100 million vehicles. And if each vehicle has a kilowatt of efficient inference compute, I think there's a, there's, I think there's sort of an Amazon Web Services AWS type opportunity. Because if you've got 100 million vehicles with a kilowatt of efficient inference compute, you've got 100 gigawatts of compute, like 100 gigawatts of computers a lot.

And it's distributed all over the world. So even so, when the car is not in autonomous mode, which I think probably is, you know, doing robotaxi work, maybe 50, 60 hours a week, but at 100 hours a week, it's, it's not, it's probably stationary. So. So there's 100 hours of 100 gigawatts of inference compute, which I think we should use.

Why not? You know, when people looked at Amazon, which started out, I would say, as an online bookseller, and it's on, you

know,

has grown to be like this incredible place where you can buy anything, and then they, and then Amazon Web Services, like, well, they got all these computers that only really see peak usage sometimes, but what are they going to do when it's not peak usage? And sometimes Amazon servers are down like 10%. So that's when they said, well, let's do Amazon Web Services. And then Amazon Web Services became more valuable than the entire rest of Amazon.

And, you know, I think there's, there's some kind of opportunity there that's pretty significant for Tesla down the road.

You know, again, that's really, nobody's really factoring that in, but I think that that actually will be quite significant.

We also are no longer compute constrained for training. So I check in with the team. It's like, is there anything we could do to improve the pace of progress with respect to training and inference? And currently that is not the limiting factor. In fact, the limiting factor right now is that the amount of miles between interventions is so long that it takes quite a while to figure out which version is better than the other version because none of them are requiring any interventions.

So it's like, you know, if you start getting to like thousands of miles between interventions or you're like 10,000 miles to get an intervention, then like well the average person only drives about 10,000 miles in a year. And if it's in an urban environment and the Average speed is 20 miles an hour, I mean our professional test drivers get pretty bored frankly. You know, they're like okay, I drove all week and there was no intervention.

Like the highlight of the week would be like yes, an intervention. Finally it's like getting to that point.

So this is where actually having a giant fleet is extremely important because we can deploy a new FSD model and run it in shadow mode and see what, see how well it performs.

You know, compare how the human drives the car versus the the new self driving build and, and then analyze that delta in shadow mode like the shadow knows or doesn't know as the case may be. And, and then be able to assess by getting billions of miles very quickly with the giant fleet like that. Basically that, that data engine is incredibly helpful. Like actually it's not possible to solve the self driving problem without having millions of vehicles on the road.

So, so that's actually the, it's like figuring out clever ways to test which how good the next build FSD build is is actually a limiting factor right now. And then of course like I said, we are building another training data center right here which will be dedicated to hardware 4 training.

So we'll bifurcate hardware 3 and hardware 4 training later this year. We'll keep improving hardware 3, but we're going to uncork the full capability of hardware 4 as well.

So yeah, I think most people here have tried out version 12.

We did say like unsupervised full self driving. Full self driving would be version 12. So we are actually just keeping the version arbitrarily at 12 and then like calling it 12.4, 12.3, 12.4, 12.5, but it's actually really like version 13 version 14. But anyway this is an arbitrary designation. So 12.4 is actually like a whole different version than 12.3 and 12.5 is a whole different version 12.4.

So you'll see really giant improvements I think sometimes factor of 10 improvements it between successor versions.

And then as I mentioned, the way it'll work for the existing fleet is you'll Be able to add or subtract your car to the fleet as you'd like.

And there will also be Tesla owned cars. So it'll be a combination of like Uber plus Airbnb.

But it'll be pretty wild that there'll be a software update and then the whole fleet suddenly becomes accessible. It's like Suddenly you've got 7 million, 10 million cars that can ultimately tens of millions of cars that can do autonomous driving and instead of being used 10 hours a week, can be used 50 or 60 hours a week.

So it's pretty wild to just, you know, be in Palo Alto with a bunch of cubes and then a humanoid robot just walks past.

We've made a massive amount of progress with Optimus in a short period of time, from someone pretending to be a robot dancing in a suit, to a pretty hodgepodgey robot, to a robot that is actually doing useful tasks in the factory today. So we have two Optimus robots in our Fremont factory that are doing basically this task which is taking cells off the end of the line and placing them in a shipping container. And yeah, we actually have quite a few of these cruising around our offices in Palo Alto.

So there's, and I think we've got kind of like one major hardware revision which should be done by end of this year or early next before. And then we'll move into a limited production next year of Optimus limited production for use in our factories where we'll test out the product, kind of, you know, you know, as I say, sort of eat our own dog food or whatever the electronic equivalent of that is.

So, but I, I think like next year, I, my prediction is next year we'll have over, over a thousand, maybe a few thousand Optimus robots working at Tesla.

And, and, and things are going to scale up very rapidly from there. So we'll iron out the bugs. It'll like, the degree of autonomy will be radically better. You'll just literally be able to talk to it and say, please do this task or I'm going to show you something. Now do that, the thing that I'm

showing you

and you know, get to the point where it can watch a video of something like a person and then learn just by looking at that video and do that task. So, yeah, so really it's going to be quite something and I'm confident of the prediction that there will be more like the ratio of humanoid robots to humans will be greater than one to one, so that there'll be more than 10 billion humanoid robots in the world.

Probably 20 or more. And Tesla is going to be by far the leader in that. You're seeing a lot of robot startups, but I think it's actually very challenging to do optimus as a robot startup because what we found to make optimus work, we've had to design from first principles, from scratch every part of the robot. So the motors, the gearbox, the sensors, the power electronics, the communication system, everything had to be done from scratch.

We found that there's basically nothing, there's no supply chain. So even though there are many electric motors made in the world, there's no supply chain for the types of motors and sensors and gearboxes that are needed for a humanoid robot.

And I mean, what you're seeing here is our current generation hand and arm, but our next gen, which has 11 degrees of freedom, our next generation has 22 degrees of freedom. It will be able to play the piano.

So it's really like, wow.

Now, of course, like I said, we need to make sure you don't have the Terminator scenario. That's very important.

So safety of the humanoid robot will be very important.

But because it requires so much ground up design, designing every motor, gearbox, sensor, power electronics from scratch, it's very hard for a startup to, if not impossible for a startup to replicate that. But at Tesla, we have the world's best electrical engineering. I think we've got the world's best mechanical engineering for gearboxes and for electric motors, power electronics.

So we have the resources to do that. It applies quite well. And then you also have to have the, the brain, you need the, you need the, the, the, you need a power efficient inference computer which we've got for the car and we'll be using an optimus. You need AI, real, you need to be the best in real world AI and Tesla is the best in real world AI. So you need all of these, you need a very strong hand of cards in order to make a compelling robot.

And then you also need to be very good at scale manufacturing. So in order to have the robot not cost like hundreds of thousands of dollars, in order to make it cost like 10 or $20,000, actually need to design for manufacturing and be very good at manufacturing.

And in my experience, prototypes are easy compared to volume manufacturing. Prototypes are easy. Production is hard, relatively speaking. So Tesla has the production capability, it has the engineering capability and it has the AI hardware and software capability.

And even the most optimistic estimates that I've seen for Optimus, the Optimus optimist, I think undercount the magnitude of what this Robot will be able to do.

You know, as I said at the beginning of the, of the presentation, I, you know, I agree with the Ark Invest analysis that autonomous transport is called sort of a 5 to 7 trillion dollar market cap situation. Optimus, I think is, is a 25, literally 25 trillion dollar market cap situation.

So.

Now I don't want to, I don't want to trivialize, trivialize what's necessary to get there. I mean, it's an immense amount of work that is required to get there. Like super difficult. But we are moving very fast down that road. We're going to make it happen. So thank you.

I guess with that we can go to Q and A.

Sorry. Yeah. Oh, I guess we've got mics lined up.

Time for the mic.

Hey, Elon, last year I thanked you

for making the world a better place for my son to live in. And I wasn't clear in particular, I was referring to what you've done for the First Amendment and our rights. Thank you.

History will not forget it. And the sacrifices you made and the

risks you took will pay off and have paid off.

Thank you.

I very much appreciate that. You know, I think it's, you have to say, like, why was it, why was the First Amendment so important is because, you know, people came here from places where if they spoke their mind, they'd be put in prison or killed or just shut out from society, ostracized. So freedom of speech is absolutely fundamental to a good future. And yeah, that's the reason for, you know, acquiring Twitter and trying to just making it a free speech platform.

And I think it's, it was like, I think very important to just the future of civilization. So thank you.

Hey, Elon, my name is Ronnie. I'm on your right side. How's it going?

Hey, I work for Tesla Sales in Orland Park, Illinois. And I was always wondering, do you think Tesla could ever move forward towards bundling software as like a package to, you know, possibly increase take rate for fsd, you know, maybe bundle it with premium connectivity and future software options?

I think we definitely want to encourage people to try fsd. And one of the things that sort of, I think probably makes sense is that when we have a significant improvement, every time we have a significant improvement in fsd, give people, you know, give all customers a, like a two week free trial. So, like, hey, we've made an improvement to FSD. You know, try it for another two weeks. You know, with, with, with 12.4, we finally get rid of the steering wheel nag, which I know is driving people crazy.

Like, like, like thank heaven for small mercies.

So you know, even just something like that, not having the steer steering wheel nag and obviously having a far lower intervention count, like far more miles between interventions. So I think we'll, you know, I think keep sort of offering people free trials of FSD every time we make a significant improvement.

Oh, sorry. Yeah. When you hear the loud, the speakers you're like, ears don't work, you know.

Good afternoon.

Yeah, two questions actually. First, after $56 billion, you can afford to probably take your kids to Disneyland now, is that where you're going?

Well, I mean, by the way, it is worth emphasizing that is Tesla stock that I have to own for five years. So you know, like it's like it's not possible to like, not actually cash and I can't cut and run, nor would I want to, but.

And I've actually been to Disneyland Disney World many times with kids. It's a little tough for me to go these days without causing a ruckus.

You know, maybe I should just like wear a Mickey Mouse outfit and just go and you know, pretend I'm like one of the.

So as we work on autonomy and we're getting to that threshold where things are getting to a seriously advanced level, eventually there are going to be some harsh consequences. Part of breaking through.

Right.

And I'm just curious where your mind is at on that and have you thought about, you know, some of those early consequences that we're unfortunately going to have to have in order to have a better, safer future? I'm just wondering, you mean like scary robots or what?

No, no, in auto.

In auto, in the cars with full self driving. Okay. Like we've seen some companies that there's been some unfortunate mishaps and oh yeah, just straight up like closed shop and you know. Yeah, that's not the plan. But those are real consequences of, of developing this technology. And I'm just wondering where your mind is at on that.

Yeah, well we are, we do try to be very careful with the deployment of fsd and you know, we do have a lot of cars. So what? Like inevitably when there are a lot of cars and you've got, you know, billions of miles, you do have the law of large numbers where okay, there's a small chance of something bad happening. But when you look at the sort of safety per mile, because we've got a lot of miles, it's very clear that the safety per mile is better than human driving.

So, so when, when we, you know, we can show that to regulators and obviously make our case to the public and say like, actually, you know, human driving is not perfect. You know, there's, there's a person killed roughly every 80 million miles. So like, like 35, 40,000 people are killed every year in the US from auto accidents. What matters is like, are we making that number smaller? And as long as we're making that, that number smaller, we're doing the right thing.

Thank you.

Hey Elon, this is Bash. I have one question.

Sure.

So we know that, we know that. We love you all. We love from our bottom of the heart. We love you so much, Elon.

Thank you.

My question is, my question is like, we all know Tesla will be the most valuable company in the world one day.

Yeah.

We just want to know if you have any prediction when and what is the inspection point. When, when we can say this is the inspection point is, is it solving the FSC or is it like Optimus on ground will be the point from your, you know, perspective?

Well, as I mentioned before, I think it's worth looking at the,

you know,

the Ark Invest analysis, I think that is the most accurate analysis. Their prediction is I think five to $7 trillion by 2029 market cap. So more than 10 times what things are today in five years. I think that is accurate.

Thank you. Thank you.

Hey Elon.

Hey.

Huge fan.

And I think history will show that

you've had a profound impact on humanity.

So I think that's just given. And just a couple of questions. I guess I fixed my Tesla addiction

by putting cars on Turo and I found that about half my renters the first time they're driving a Tesla and

then the other half really want to buy one after they've tried it out. So I guess a couple questions.

When could we expect maybe the deferral program to come back? If it would come back? And could we also get a passenger

van, like a 12 passenger van sometime soonish?

I'll add that to my list.

So the, yeah. So the referral program.

Yeah, I think we'll re examine the referral program probably next quarter and like it is very important to sort of look at the, what is the fully considered cost of a Tesla? So with the referral program it effectively increases the price of the car. So you know, is the price increase what is better for the end customer, a slight price increase because of the referral program or not? But I think we examine that, I think in, you know, next quarter and, and see if it, see if it makes sense.

I think some Kind of referral program just makes sense.

And, you know, but I think for a lot of people, like, by far, like, by far the biggest thing we see is that people just can't afford the car. That's the biggest thing. Like, that's the limiter on demand.

You can imagine in the limit, if there was, say, you know, an 18, like, if model Y costs, like less than $20,000, it was still, like, probably 5 million units. Like, it'd be insane. So now that's not feasible, you know, today. But I think working on figuring out ways to make more affordable vehicles is really the game changer. Like, for. For most people, it is really not a question of whether they want a Tesla. They do want a Tesla.

They simply do not have enough money to afford one. So we have to make it affordable. That's essential. Thank you.

Hey, Elon. I just wanted to introduce myself. My name is Hamza. First off, thank you for introducing me to Anrabek. He and I are now, like, close, thanks to you, and I moved to Austin thanks to you. So my question is basically for FSD transfer. I think that should be something that should be ongoing. I think not only on the benefit of you for getting more people on the floor fleet and getting, you know, statistics and info, but also as Tesla owners, it makes us want to get the next model and transfer over.

You know, for instance, the thing that's been holding me back is wanting to wait for the FSD transfer again. So is there a reason why you haven't implemented that for, like, ongoing time? It is pretty complex for us to do this. It sort of adds a fair bit of complexity to the sales process. But, yeah, and it is tough because we've got to earn revenue somehow.

So, you know, transfer is tough.

Sorry, one more quarter.

Okay, one more quarter.

All right, you got it.

One more quarter.

All right. So, yeah.

All right.

Oh, Elon, thank you for hosting us. If you can see me, I'm trembling uncontrollably. Really look up to you. Shareholder, consumer, you know, tripled my humble, tiny little portion of your shares the last two days. Just being engaged, knowing we're coming here and appreciate you hosting shareholders, and it just means so much to us. You're welcome.

You're an absolute idol. This question comes because I care about you as a human being, but in the context of being a shareholder, unlocking these values of Optimus, we need you. And just with all the global uncertainty, you're so high profile. Just want to make sure you're doing everything to take care of yourself. Safety and health for the future of the company.

I say that because I care about you as a person. But since this is a shareholder meeting in the spirit of shareholder value. Well, thank you. I appreciate the kind sentiments.

Yeah, I think I probably need to work out and you know, and like not get assassinated or something, you know, I mean, it is getting a little crazy these days, you know, like to first approximation, the probability that a homicidal maniac will try to kill you is proportionate to how many homicidal maniacs hear your name. Okay, so they hear my name a lot. I'm like, okay, I'm on the list, you know.

So, yeah, so we actually did have two homicidal maniacs in the last roughly seven months come to aspirationally try to kill me and a bunch of other people. So it's not just me. And there wasn't like an actual issue that they articulated. They were just in the homicidal maniac career. So.

And so I do need to be kind of careful.

It does.

Which is not great because it like forces me to be a bit more sort of standoffish, you know, and you know, like, so now like, I've like stopped signing things because just, you know, something. Because like, that would be a sure way to like homicidal maniac, I'll just sign something and then shoot them type of thing. So then I, you know, I have to be more standoffish and stuff, which I prefer not to be. But it seems wise, I think, like John Lennon who was like singing about like, hey, can't we all just be nice to each other?

And then he got shot, you know, by one of his fans, like, okay, we'll try to avoid that.

So anyway, I appreciate the kind sentiments. I will endeavor to stay alive and.

All Right.

Hi Elon, I was reading an article the other day. Is Tesla still have somewhere in the product timeline a home H Vac system in the works, A Home Interact system? We do not have a Home Interact system that I'm aware of. No, no, we do not want H Vac system.

Oh, H Vac.

Oh, H Vac. Yeah, yeah, sure, yeah. In the list of things that would be like, more useful than like, it would be nice to do would be a great H Vac system that is very power efficient, you know, has like a built in HEPA filter and like take a lot of things that we've learned in the car and apply that to home H Vac.

Because I'm. I actually maybe there is a great home H Vac system out there, but I don't, I haven't found it. Which is pretty wild.

And just building in the filtration system and you know, being able to sort of maybe also control humidity somewhat and like I said, be super efficient and quiet, which are all things that we do in the car.

That would be a pretty sweet thing to do. Maybe we can combine that with the powerwall or something.

On the list of things to do that is not super high, but it would be cool to do that one day.

Hi, Elon, this is Sandeep. I just want to thank you for making the Cybertruck.

You're welcome.

Yeah, kids love it. Really? My four and six year old, they love it.

Exactly. That's what I say, kids. It's like amazing.

It's like I went to Disney last week. Nobody looks at Disney. Everybody looks at Sabbath.

Yeah.

There's a funny story, if you may allow. So we were preparing my daughter, we went to Destin and then from there we were driving to Disney and then we were preparing my daughter to. She's four years. To just get used to the lines and you know, you have to have a lot of patience to get in the line. And she's like, dada. Why don't we just say we'll show you our cyber truck and then we'll cut the lines. I mean, she got used to it because we were driving superchargers and then everybody was coming over taking pictures.

Sure.

It's a funny story.

Yeah, yeah.

And then I have a request. Would you allow 6 year old to tour the factory? Maybe someday in the future. Right now it's capped at 12 years. Okay. Even though we have referral rewards, it's capped at 12 years. Would you allow the kids to tour the factory? Maybe they can see the robots.

Sure, sure.

We make an event or something like that.

Yeah.

Maybe we could have like a worse like worried about kid safety and stuff, you know, in case kid like goes running off into like the robot cell or something. But, but yeah, we could I think lower this, maybe have like a more limited tour that's know enables younger kids to go. Sure, we'll do that.

All right.

See.

Okay.

Donald Trump has been a big critic of electric vehicles.

But last week he surprised us by saying he's a big fan of Tesla and a big fan of you.

What did you tell him?

Well, you know, I can be persuasive.

So.

Yeah, actually I don't exactly know why he. This is a good question. I mean, you know, I have had some conversations with him and he does, he does Call me out of the blue for no reason. I don't know why, but he does. And it's, and it's like he's very nice when he calls and you know, and I was like, you know, you know, electric cars I think are pretty good for the future. America's the leader in electric cars, you know, you know, buy America and stuff.

And, and I think he actually, a lot of his friends now have Teslas and they, they will love it. And he's a huge fan of the cybertruck, so I think those may be contributing factors.

Hey, Elon, on your right.

Yeah.

My name is Keanu, longtime fan of your brand and company and everything. I'm wondering, when you take the driver out of the seat for the driverless car, how do you plan to handle interventions which I'm sure will happen at point some, some rate, even if they're rare?

Yeah, so there's, I mean, clearly we want to make sure that if there's no driver in the car, the probability of a severe intervention is extremely low. You know, we want to make sure that the car is extremely cautious around vulnerable road users like cyclists or pedestrians.

So really we'd want the probability of, of a, of, of hurting a pedestrian or cyclist to be incredibly low. Like suit way lower than someone driving a car. Now then there are, there are interventions that happen where the car is simply like, I don't know, gets stuck down a one way street whereas with a like crazy construction situation that, but in which case it'll, it'll basically phone home and say, help, I need, I'm stuck.

And then I think we could probably solve a lot of these things just by, you know, remoting into the car and kind of like steering it like a giant RC car.

That's what I'm guessing. That's what I think we'll do.

Hey, Elon Anwarbek from Kazakhstan. Thanks for installing superchargers. I just want to acknowledge Tesla, Boomer Mama and Sawyer Merritt for all the efforts. They did organize the.

There's a after party to celebrate at Luster Pearl east if you guys want to come and celebrate all the efforts for the community. Congratulations, Elon. Thank you. Yep.

Thanks for hosting today, Elon. It's about four years since you did battery day and grew up there was very ambitious targets of getting to a terawatt hour of batteries. And earlier in your presentation, you alluded to the great difficulty of scaling up this new technology. I think as shareholders, we're most interested. Could you tell us a little bit more? Because in terms of affordability, which you also mentioned as this hu huge burden.

These 4680s get manufacturing credits and also on the consumer side get large credits as well. If you could scale that so that we could produce full level of cybertrucks

and also start to put them back

in the Y and the three, potentially it would greatly improve affordability and be great in terms of scale.

Sure.

So achieving affordability is actually at this point more about the rest of the car than it is actually about the battery pack. So the price per kilowatt hour of cells from suppliers has dropped dramatically. So there was a massive spike for a few years in the cost of lithium ion cells from suppliers because every car manufacturer put in gigantic orders for battery packs. Like giant, giant orders. So the price of like, of lithium and like all the raw materials and the battery cells went, went crazy.

It doubled like the price of cells doubled. Price of lithium I think went up at almost like five, five fold or like 500%. Like went up crazy numbers. So since then the demand like suppliers, battery cell suppliers have increased their supply and the orders from other car manufacturers have declined.

So I mean it is worth noting that it's, it's like, it's tough sledding out there for EVs. You know, GM announced I think a 50% reduction in recently in their EVs. Ford announced reductions. I think almost every manufacturer has announced reductions in their electric vehicle plans. Now Tesla we're going to increase, but it, but it is not an easy market.

And but to get the cost of car down, it's really the battery pack is a piece of the puzzle, but it's really everything else too. So I really would describe like we're making a car more affordable. It's like Game of Thrones, but pennies. Game of pennies. So there's quote to first approximation, 10,000 unique items in a car. And if each one costs five bucks, then you've got a $50,000 car. So to actually get the cost of car down, you've got to chisel away pennies to get the 10,000 items to be $3 or something instead of $5.

That's kind of what it amounts to.

Now with respect to our own cell production, we do see a path to cost parity by the end of this year.

A very difficult path to cost parity. Currently our 4600 cost more than our suppliers. Now they cost more than the suppliers today, but they cost less than the suppliers a year ago. So you know, there's a bit of a beast famine Thing with battery cell supply, it's kind of like for DRAM chips, you know, like the DRAM industry goes from like, you know, oversupply to under supply. The price of DRAM changes like crazy. It's kind of like that for sales.

So. But we expect to achieve cost parity with even the much lower supplier sell price today by the end of the year.

Hi, Mr. Musk. I've been a long term investor since 2012. Great. And I voted for your compensation package in 2018. Again, thank you this year. And shout out to Tesla Boomer Mama. Alexandra,

I had a privilege to talk

to you in 2013 briefly. And you spoke about the truck which became the Cybertruck. You're a man of reward. I have friends all over the world in Asia, Canada and Europe. And I know the cybertruck's been on tour. How's been the reception for the cybertruck and when will it be or is it possible to sell and when will it be sold internationally? Yeah, so I think probably, you know, we might be able to certify it for other markets sometime next year.

But like for sure this year it's just North America. We did design the car to North American requirements because if you start going with the superset of all international requirements, it forces a lot of constraints on the, on, on cybertruck that would make the product frankly worse.

So I think we'll need to make a special version that is for example, China compliant or Europe compliant.

But it really doesn't make sense to add that complexity until we've achieved higher volume production on Cybertruck. There's still a lot of work to do on cost, cost down for the Cybertruck. You know, I generally would say like the, the level of difficulty of going from, say, prototype, you're from prototype to production, it's 100 times harder to go from, from the, you know, from prototype to production is 100 times harder.

And then once you reach production, to improve the price by 20%. Like the cost of goods by 20% is harder than reaching production in the first place. Like it's mega pain. It's like, you know, and to be frank, it's not the funnest job in the world. You know, like chiseling pennies is not. It's like washing the dishes, frankly, you know, like wash a lot of dishes here. Like it's super fun to make prototypes. That's really fun.

And it's, it's kind of cool to get, it's pretty cool to get the production going. Then the cost grind is a grind that is like, you know, it's hard work, you know, there's some reward to it. Like there's sort of a reward to washing the dishes. I guess you have clean dishes.

But it's, that is, that is tough.

So that's real hard work. I just have to say that there's been a really big reception. Like I have friends that's, you know, message me and said they love the cybertruck design and it's, it's definitely a head turner, so. Oh yeah, I think from a demand standpoint, we would have plenty of demand in other parts of the world for cybertruck, no question. Like wherever it goes anywhere in the world, it's like it gets a massive crowd.

So, so it's not a demand thing. It's. It's really, we've got to reach volume production, go through the cost grind. I can't. It would be difficult for me to overstate the difficulty in the cost grind. It's really intense. And then we've got to recertify the car which includes making some design changes for to meet to be compliant in other markets.

All right.

Good afternoon, Elon. I like your shoes. I don't know if everybody else can see them. Yeah, out there. Cybertruck.

Yeah.

My question was in regards to the foundation series, do you guys have a certain number that you are planning on making or timeline of when the non foundation series are going to come out? Yeah, we'll start making the, we'll end the foundation series actually pretty soon and the non foundation series sometime next quarter.

Hi Elon, thank you for delivering shareholder value and congrats on your pay package. And I'm Michael. I ask a lot of questions on the quarterly investor calls and I wanted to ask you, can you give me a little bit more color on the neural net scaling laws that you're seeing as you're training fsd? Like how much data, how much more inference compute. How much more training compute do you need for each successive? 9 in reliability of the full self driving driving software.

And how much data are you currently collecting? Thank you. Yeah, that's a great question.

I mean that's something that we go through really every week at Tesla. I spend a lot of time with the Tesla AI team and I think probably like, you know, like a huge portion of, I mean it's really like a big part of my job is like, okay, we need to get the car to drive itself, you know, and yeah, you need to solve things on a chip level for inference, for training, for software. There's also massive data storage challenges because you have petabytes of video coming in.

And even for saying for a big Nvidia cluster, it's actually really hard to make a big Nvidia cluster actually work in training. It doesn't just open the box and plug it in. It's not like that at all. It's like massive effort required to get a big training cluster to operate with, to operate at all and then operate with high uptime.

So I mean with respect to like the various versions that we're seeing, like you go from version 12, 3, just ignore the 12 and think of it like 3, 4, 5, 6. That's actually the way to think about it, ignore the 12 part and just call it like version 3, version 4, version 5, that's version 6. That's how to think about it. Between each one there's anywhere from, I don't know, a minimum of a doubling of improvement to in some cases, I think a five to tenfold improvement.

You know, especially as we eliminate sort of say, I think we're close to eliminating almost all static object collisions. So you know, like curves and medians and we're pretty close to just having it to have almost zero static object collisions. So once you get to zero static now you have a gigantic improvement in miles between intervention.

Then you get into classes of dynamic objects that are different classes and different things can get obviously very complicated. If you've got a complex intersection with traffic lights all over the place and like somebody bumped the one traffic light and now it's pointing in the wrong direction or the traffic light's broken and there's someone who is not obeying the traffic laws because they're just running a red or they got confused about the traffic light.

So you get these very complex situations that are much harder to solve.

So and then like as I was saying earlier, it actually gets, as the system gets better, it gets harder to figure out which AI model is better. Like, like, because now, you know, like, okay, it's thousands of miles between intervention.

How do we as quickly as possible figure out which AI model is better? And when, when you make these different AI models, they're, they're obviously not like super deterministic. So because something a new model that does that eliminates one problem, but creates another problem. So.

We're trying to solve this by a combination of simulation, uploading models, having them run in shadow mode. So it's actually kind of helpful that not everyone has full self driving because we can see we can run it in shadow mode and See, what would this new model have done compared to what the user did? So since we've got, you know, millions of cars that we can do this with, that gives us a delta between what the AI model predicted would do and the user would do.

And if you kind of sum up the errors between them, you can see, oh, there was a bigger error stack from this model versus that model when you uploaded them into. We each uploaded them into 100,000 cars.

But that's the biggest limiter right now. It's not training, it's not data. It's actually testing the AI models and then figuring out clever ways to figure out if a new model is better or not. Like we know like there are sort of particular intersections that are difficult. And so we, if we just like all the models will be great at like driving on a, on a, on a well painted road with no intersection, they'll all be great at that.

But, but will they be great at the complicate. So, so we've identified several thousand intersections in the United States where we'll, we'll actually take the new model and try the car in those, those very difficult intersections. Like you know, the sort of the famous, like Chuck's left turn.

Yeah,

so if you imagine Chuck's left turn. Hey Chuck.

I wonder what the neighbors think.

But there's many such complicated traffic situations in the country. So we've identified several thousand which we then send cars to those locations to compare the models.

And then as I mentioned, we have yet to uncork any of the ability of hardware 4, which we'll do later this year.

So I think the exponential rate of improvement of miles between intervention will continue and will accelerate.

Elon, I was wondering if, as you've

called a Tesla, basically a robot with wheels. And so I'm wondering kind of as that, as that logic transfers over to Optimus, what kind of will Tesla use

any large language models on top of

the current Tesla robotics capabilities? And also if so, will Optimus have any kind of personality and would that be something that the end user might be able to have control over? Yes to all of those questions.

Yes to all.

So like we're headed for a wild future. I mean wild, wild, wild.

You know, cars will drive themselves, people will have super helpful humanoid helper bots.

And yeah, you'll be able to customize the personality, customize the voice and, and really kind of the robot will kind of get to know you as well and know your preferences. You won't even have to say things because we'll just kind of understand you. I think it will get to this point where it's like it'll be odd not to have a robot buddy, like you know, sort of not, not to have your C3PO and you know, R2D2.

So yeah.

Hi Ellen, Max here, huge fan, thank you for hosting us. I wanted to ask you, how do you, how much integration work do you think will be needed to deploy Optimus into different types of factories, industries like unique technicians or just have them go watch? And how do you foresee that? Well, I think if you fast forward a few years it really will be quite a generalized robot. Like say two years from now, like next year it'll like this year it's still quite specific, you know, task specific training.

So to do a particular task we have to kind of do fair bit of training to solve that task.