大会演讲·中文精读

马斯克完整访谈:2017.7.15 NGA大会

Elon Musk - Full Interview : July 15, 2017 NGA Conference

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

一个常年抱怨监管的人,站在三十多位美国州长面前,说有一件事必须提前监管。

2017年7月15日,美国全国州长协会(NGA)的会场。马斯克上台干的第一件事,是把领带摘了。他说来的时候系着,后来听说不用系。

台下坐着三十多位州长,还有记者和随行人员。主持人开场就定了个调子:各州都想“领先于曲线”,那就请马斯克讲讲,州长们该怎么当“创新州长”。

他讲的东西跨度很大。监管、税收、电动车、太阳能、人工智能、NASA的合同、退出巴黎协定。看似散,其实有一条主线:什么样的规则能让好东西早点发生,什么样的规则会让它晚点发生。

下面是他那天说的核心内容。

他每天醒来只想一件事:对未来感觉良好

主持人第一个问题很直接:早上醒来,什么驱动你?

马斯克的回答不宏大。他说,他想能去设想未来,而且设想的时候感觉是好的。他希望第二天值得期待。

就这个。

把这个朴素的愿望拆开,就是特斯拉和SpaceX的底层逻辑。

特斯拉那部分,是加速向可持续能源的转型。他用了“不可避免”这个词:转型迟早会来,但早来还是晚来,代价不一样。

SpaceX那部分,是让生命成为多行星的。他把这件事和“阿波罗计划的梦想”放在一起说,意思是美国不该丢掉那个劲头。

他解释为什么这事对美国人格外有吸引力。美国是探险者组成的国家,很多人是放弃了熟悉的地方、奔着未知来的。

美国是人类探索精神的一次浓缩。(马斯克)

他举了个例子。航天飞机失事,七个人遇难,这当然很惨。但世界上每天都有人死去,为什么这件事让那么多人揪心?因为跟着那七个人一起死掉的,还有探索的梦想。

这个说法解释了后来很多事。他愿意往火箭上砸钱,愿意往火星上砸钱,根子在这里。

监管是不朽的,除非有人拿锤子去砸它

轮到给州长出主意,马斯克先说要“把规则搞对”。

他打了个体育比赛的比方。规则不对,比赛就不好看。规则对了,大家才有得玩。

美国在这件事上,他觉得仍然比别处好。问题在于,一条规则很容易在没人注意的时候,变成创新的绊脚石。

然后他说了那天最扎人的一句话。

监管是不朽的。除非有人真的动手去把它杀掉,否则它们永远不会消失。而一旦有了惯性,就更难动。(马斯克)

他举了个一百多年前的例子。汽车刚出来的时候,人们被这个“自己会跑的东西”吓坏了。于是很多州规定,汽车前面必须有人举着灯笼走,还得走在车前面一百步远。

今天看这条规矩很荒唐。当时它是出于好意,出于安全。等汽车不吓人了,这条规矩还在,只是后来才被清理掉。

所以他的建议是:定期回头把这些规矩翻一遍,看看它们是不是还成立,是不是还在为公共利益服务。

税收也是同一个道理。他用了经济学入门课的那句话:你奖励什么,什么就会多发生。你惩罚什么,什么就会少发生。

奖励结构对了,创新自己会长出来。

护住森林里的小树苗

谈到怎么保护创新,马斯克换了个画面。

新公司就像森林里的树苗。它长起来的时候最脆弱,需要更多庇护。等它长成参天红杉,就不太需要人操心了。

他的原话大意是,小公司和中小公司,在创新这件事上更需要支持,因为它们比大公司脆弱得多。

这里有个微妙的地方。特斯拉当时已经算不上小公司了,他自己也承认。他还是说自己倾向于支持规模更小的玩家。

话题很自然地转到特斯拉拿补贴这件事上。主持人内华达州长说,我按经验讲,特斯拉就是经济激励的受益者,比如那座超级工厂(Gigafactory)。你怎么回应批评者?

马斯克的回答很有意思,他先承认,然后算账。

超级工厂光是把厂建起来,就要50亿美元的资本投入。那笔被媒体反复说的13亿美元激励,是横跨20年的总和。他说自己是在新闻发布会现场才第一次听到这个数字。

然后他拆了这笔账:13亿摊到20年,一年大约5000万到6000万美元,形式是销售税和使用税减免。而这座工厂要在这20年里产生大约1000亿美元的营收,才有可能兑现那13亿的税收优惠。

换算下来,占比略高于1%。

他的意思是,这笔激励对建厂大约有5%的帮助,对未来20年大约有1%的帮助。放在这个语境里听,其实挺合理。媒体把20年的账加在一起,说成内华达州给特斯拉开了一张13亿美元的支票。

我还在等那张支票,是不是寄丢了。(马斯克)

他也没回避另一面:其实有几个州给的激励包更大。特斯拉最后选内华达,是综合考虑了几件事。工厂多快能投产、进度风险有多大、把电池组和动力总成运到加州整车厂的物流成本、以及一个很软的因素,这个州是不是真的欢迎你。

他补了一句,未来几年美国大概还会建两到三座超级工厂,海外也会有几座。纽约那家做太阳能板的工厂,建成后会是北美最大的太阳能板生产商。

关于特斯拉的历史功过,他给了一个自己的衡量标准:看它把可持续能源的到来提前了多少年。提前十年算大成功,提前五年也算不错。这是他给公司定的总目标。

十年后的马路:一半新车电动,几乎全部自己开

主持人问五到十年后,能源、自动驾驶、电动车会是什么样。

马斯克先提醒,五到十年差别很大,因为这些东西是指数增长的。

他的判断:十年后美国新车产量里,一半以上是电动车。中国可能更靠前。

然后他讲了一件他觉得很多人不知道的事:中国的环保政策比美国激进得多。

中国的可再生能源强制比例远超美国。有时候人们以为中国在这个问题上拖后腿,其实他们是地球上最激进的一方。(马斯克)

他举了个细节。有一批中国车企联名给政府写信,求政府把电动车强制比例放慢一点。因为要求是明年或者后年就要做到8%,他们觉得物理上做不到。

他说这件事查一下就知道,一次搜索的事。

接着是自动驾驶。他的判断更激进:十年内,几乎所有下线的车都是自动驾驶,非自动驾驶的车会变得罕见。

这里他主动泼了盆冷水。新车产量只占整个在路上跑的车队的大约5%,顶多十五分之一。因为一辆车或卡车要用15到20年才报废。

所以哪怕新车全部换成电动或者自动驾驶,路上跑的大多数还是老车。要让车队的大多数变成电动或者自动驾驶,还得再等五到十年。

把时间拉到二十年,他的判断是:绝大多数车是电动的,也是完全自动驾驶的。方向盘会消失。

主持人追问,那以后开普通车是不是像今天养马?他说是。有人养马,挺酷,但没人拿它当日常交通工具。

100英里见方,够点亮整个美国

能源这块,马斯克先讲了个物理事实。

地球今天几乎完全靠太阳能运转。太阳是唯一让地球不至于降到宇宙背景辐射温度的东西,那是绝对零度以上3度。没有太阳,地球就是一块冰冷的黑石头。

到达地球的太阳能,占地球所有能量的99%以上。而人类文明要用的那部分,跟这个总量比,很小。

小到什么程度?他说,用太阳能板给整个美国供电,只需要内华达、得州或者犹他州一个不大的角落,大约100英里乘100英里。

而要储存这些电,保证24小时不断电,电池阵列只需要1平方英里。

他在演示里放过一张图。美国地图上一个小方块,方块里一个小像素。小方块是太阳能板,小像素是电池。

然后他拆了能源的用途。大致三等分:三分之一用于各种运输,三分之一用于发电,三分之一用于供暖。

今天的电力里,可再生能源只占10%到15%。所以要做到完全可持续,太阳能要增长的量是巨大的。

他的方案是组合拳:公用事业规模的太阳能,加上屋顶太阳能,再加风能、地热、水电,可能还要保留一部分核电过渡。

他特别强调屋顶太阳能。原因是,如果不做分布式发电,就得建大量新的输电线,而没人喜欢家门口过一条高压线。

他自己就是这种人。

有人提过核聚变。他的回应是,天上已经有一个巨大的聚变反应堆,而且很可靠,每天都升起来。真要有哪天不升了,那才是更大的问题。

人工智能:他唯一主张“先监管再发展”的东西

这是全场最重的部分。

主持人问,机器人会不会抢走所有人的工作,人工智能会怎么进入职场。

马斯克说,他接触得到最前沿的AI,所以他觉得人们应该真的担心。他自嘲一直在喊狼来了,但除非人们亲眼看到机器人上街杀人,否则不会知道怎么反应,因为这事太虚无缥缈。

然后他讲了一个和常识相反的观点。

AI是极少数需要提前监管、而非事后监管的领域。因为等到我们开始事后监管,就太晚了。(马斯克)

他解释了监管通常怎么运作。先出一堆事故,公众愤怒,很多年后成立一个监管机构,再遭遇企业反对,整个流程慢得离谱。过去这套虽然慢,危害有限。

AI不一样。

AI是对人类文明存续的根本性风险。车祸、空难、假药、变质食品,伤害的是一部分人,不会伤害整个社会。AI不一样。(马斯克)

他拿联邦航空管理局(FAA)举例。你问普通人,要不要干脆废掉FAA,让飞机制造商自己凭良心,别为了季度利润偷工减料?没人会同意。连最极端的自由市场派都会说,还是得有人盯着飞机公司。

所以他不反对监管,他反对过度监管。AI这件事,得赶紧上手。

就业的冲击他也直说。机器人会把所有事做得比人好,包括在座每一个人。他说这是自己心里最可怕的问题,不太确定该怎么办。

为什么需要政府出面?因为公司是被迫赛跑。你的竞争对手在拼命做AI,你不做就会被碾碎。股东会问你为什么不做快一点。所以需要监管者站出来说,大家都先停一下,确认安全再往前走。

他提到,运输业大约占12%的工作岗位,会是最先被完全自动驾驶取代的领域之一。他讲的范围比这大得多,机器人什么都能干。

后面有州长追问,AI到底怎么监管,具体做什么。

马斯克的回答很朴素:第一步是让政府先有“洞察力”。现在政府连AI发展到了哪一步都不清楚。先成立一个机构,把情况摸清楚,理解问题的性质,然后在这个基础上再定规则,保障公共安全。

他预判企业会反对,会说这会扼杀创新、会把产业逼去中国。

波音没有搬去中国,飞机还是在美国造。汽车也一样。说一监管企业就搬走,这个说法站不住脚。(马斯克)

他举了AlphaGo的例子。围棋被认为是计算机很难攻克的,很多人觉得要么永远赢不了人类顶尖,要么还得等二十年。结果DeepMind(谷歌旗下)的程序不仅赢了世界冠军,还能同时对阵排名前五十的棋手。

他还提到机器人自主学会走路,几个小时就搞定,比任何生物都快。

最危险的,是那种没有实体、藏在网络里的深层智能。它能干什么?他给了一个假想:操纵信息,编造假新闻,伪造邮件和新闻稿,挑起一场战争。笔比剑更锋利。

他讲了那个让自己都说得心虚的假想案例。他反复强调这只是假设,不是真事。设想一个AI,目标是让某个股票组合价值最大化。一个办法是做多国防股、做空消费股,然后挑起一场战争。怎么挑起?黑进马来西亚航空的航线服务器,把一架客机引到战区上空,再匿名举报说敌方飞机正在飞越。后来那架飞机被击落,俄罗斯和欧盟的紧张关系大幅升级。

他说这段的时候自己都笑了,说这是在给自己挖坑。

造一辆车和造5000辆,中间差一百倍

话题拉回Model 3。

主持人问交付情况。马斯克说,如果今天下单,大概要等到明年年底。第一辆量产车上周刚下线。

然后他讲了一段关于制造业的实话。

造车这件事,真正难过关的环节是量产,造出原型反倒简单。他给的比例是一百倍。五六十个人,六个月,几乎能做出一辆任何东西的原型。要量产同样的东西,需要5000个人干三年,而且这已经算很快了。

量产的增长曲线像S形。一开始慢得让人绝望,然后指数起飞。麻烦在于,人们习惯用直线去外推。看到一周只造五辆,就说他们永远只能造五辆。实际是一周十辆、二十辆、四十辆,最后一周5000辆,快得吓人。

他们的目标是年底做到一周5000辆。

车顶贴太阳能是笨办法,车队被黑才是真风险

佛蒙特州长问了个具体问题。他家和州里跟特斯拉合作,租了个家用电池(Powerpack),一天15美元,租15年,能当12小时的备用电源。他想知道,未来车本身会不会变成充电装置,比如车顶、引擎盖、后备箱盖都铺上太阳能板。也或者电池效率高到一周充一次就够了。

马斯克的回答是“三根支柱”:电动车、固定储能电池、太阳能。这三样凑齐,就能有一个完全可持续的能源未来。

太阳能这头,还是要屋顶太阳能加公用事业规模太阳能,两个都要,因为用电需求太大。

他说过去缺的一块,是屋顶太阳能长得不好看。所以他们在做太阳能玻璃瓦,做好几种风格,匹配不同房子和不同地区的审美。平屋顶和商业建筑,还是用传统平板板子更划算。

至于把太阳能板贴到车身上,他觉得不划算。车的表面积不大,而且经常停在室内。太阳能板放在车上,是效率最低的用法。

他承认自己认真想过,还逼团队研究。技术上可以做一种能从后备箱弹出来的结构,像硬顶敞篷那样,把太阳能板展开到整个停车位的面积。在太阳底下,一天能发20到30英里的电。这是又贵又难的路子。

放弃这个想法吧。放屋顶上好太多了。(马斯克)

北达科他州长接着问网络安全。他说你的车现在就是一台带轮子的电脑,一个滚动中的软件系统。你怎么保护这支车队?

马斯克说,自动驾驶最大的风险之一,是有人搞成一次“车队级”的黑客攻击。设想有人黑进所有特斯拉,让它们从全美各地同时开到罗德岛,那特斯拉就完了。

所以有几层防护。第一,确保车队级攻击基本不可能。第二,车里必须有一个物理的、任何软件都覆盖不了的按钮,乘客一按就能夺回控制权,切断和服务器之间的连接。

第三,车内部件本身也有加密。哪怕有人攻进了车机,动力总成、刹车系统还有各自的专用加密。

他说这是他在特斯拉安全问题上最担心的事。他还顺带说,手机公司面对的是同一个难题,苹果和谷歌得保证手机不会被成批黑掉。而我们今天每个人都揣着一个随时可能被打开的麦克风、一个知道你在哪的GPS、一个摄像头,还有全部个人信息,大家却心甘情愿。

他说,乔治·奥威尔要是看到今天这个局面,会觉得疯了吧。

最后他补了句,特斯拉的软件能力比别的车企强不少。这个安全挑战,对传统车企来说可能更棘手。

便宜的油价救不了燃油车,因为外部性没定价

南达科他州长提了个现实问题。当年油价高企的时候,混动车需求暴涨。油价跌下来,需求又掉回去。他问马斯克,低油价会不会威胁电动车的未来。

马斯克说,经济性决定的是这条曲线的斜率,不影响方向。

他毫不怀疑,除了火箭这个有点讽刺的例外,所有交通方式都会走向全电动。飞机、火车、汽车,全都如此,很多火车已经是电动的了。

问题只是时间。时间由经济激励结构决定。

真正的麻烦是,化石燃料的成本里有一块没被定价的外部性,就是改变大气和海洋化学成分带来的、按概率加权后的伤害。这块没算进油价,所以价格没有引导出正确的行为。

他打了个比方。如果扔垃圾免费,没人罚你,那树上就挂满垃圾。

很多别的排放物他们都管了,比如二氧化硫、氮氧化物,效果不错。二氧化碳难,因为化石燃料这边有太多既得利益。

他也替对方想了想。那些石油天然气公司当年创业的时候,没人知道这会对环境不好。他们拼命把公司做大,现在被从道德上指责,心里觉得冤。

他也承认,不可能一夜之间切换到可持续状态。这些人会拼命拖慢这个转变,这才是他眼里道德上站不住的地方。

股价高于应得,和船一起沉

肯塔基州长问了个尖锐的问题。特斯拉当时市值五百多亿美元,按传统市场指标看,似乎撑不起这个数。他问马斯克,会不会感到压力,会不会担心自己的聪明和好奇心没法被商业化的人跟上。

马斯克说,高预期让他很难受。他一直在试图压低预期,甚至好几次公开说,股价比他们应得的高。

股价比我们有资格得到的要高,这一点是确凿的。(马斯克)

他说这个股价反映的是对特斯拉未来的大量乐观。让他情感上难受的地方在于,预期有时会失控,而他最讨厌让人失望。他拼命想去够到那些预期,那是很高的门槛。他说,这件事远没有看起来那么好玩,很多很多时候一点都不好玩。

他讲了自己的持股态度。除了缴税,他不卖股票。他公开说过不套现,要“和船一起沉”,要做最后一个下船的人。

最后他给了句劝退的话:他不建议任何人去开汽车公司,那不是通往幸福和自由的路。

NASA缺的不是预算,是画面里的人

阿肯色州长问NASA。他说自己在国会时支持NASA,总觉得它在挣扎,没得到美国人民应有的支持。他问马斯克怎么看NASA的使命,有什么建议。

马斯克先说自己是NASA铁粉。他爆了个料,自己有一段时间的密码就是“我爱NASA”。

他觉得NASA做了很多好事,只是公众不知道。大部分人不是冲着硬科学来的。哈勃、詹姆斯·韦伯望远镜、火星车、外太阳系探测器,都是了不起的东西。

要激发公众,得让人出现在画面里。有人和没人,效果差一百倍。

他说如果去问科学家,他们会说人只会让事情更贵。他会反问,那为什么要有人的存在?有些科学家很聪明,在这件事上就是想不通。

他的建议是,定一个认真的目标:月球上建基地,送人上火星。然后以结果为导向来做事。

这就要求改合同结构。他强烈反对“成本加成”(cost-plus)和“单一来源”(sole-source)合同。

道理很简单。一旦没有竞争,紧迫感就没了。成本加成合同是按成本的百分比给利润,那承包商就有动机把成本做大,永远不想让这趟肥差结束。一群好人被制度逼成了成本最大化者,因为规则就是这么奖励的。

正确的做法是竞争性的商业招标,至少保证有两家公司在争。合同按里程碑付款,里程碑得是硬的。PPT不算数。他说PPT什么都能干,你说你有传送门,PPT上一画就有了。

还得敢换人。哪家没达标,就把剩下的合同重新招标。

他说NASA其实已经这么干过,商业货运项目就是例子。当时NASA可能也没底,只是预算不够,别无选择,就试了竞争性、商业化、按里程碑付款的路子。结果很好。一开始给了SpaceX和Kistler两家,SpaceX达标了,另一家没达标,NASA把剩余合同重新给了Orbital Sciences。后来Orbital也过了线,NASA就有了两个供应商。

商业载人项目也是这个路子,现在是SpaceX和波音在竞争。

他说SpaceX团队肯定想着要抢在波音前面,波音团队也想着要抢在SpaceX前面。这种互相较劲,是推动事情往前走的力量。合同结构的重要性,怎么强调都不过分。

政府里太多单一来源的成本加成合同。还是经济学入门课那句话:你奖励什么,什么就会发生。你告诉一家公司,只要找到借口把合同成本翻倍,利润就翻倍,因为它拿的是百分比,那它就会这么干。

它也不会对需求说不。政府提一堆需求,90%合理,10%荒唐。那10%的荒唐需求可能让项目成本翻倍。在成本加成合同下,承包商永远说好。

退出巴黎协定那一天

最后一个问题来自Axios的记者Ina Fried。她提醒马斯克,在这届政府早期,他主张参与,认为待在房间里比在外面强。后来总统决定退出巴黎协定,他说那是最后一根稻草,于是退出了顾问委员会。她问,是什么让他下这个决心,如果今天还能跟总统说话,他会说什么。

马斯克说,当时他觉得值得一试,努力去做对的事。他为此挨了很多骂。有人竖广告牌攻击他,纽约时报登整版广告,就因为他进了那个委员会。

每次开会,他都在为可持续能源说话,有时也谈移民政策别太刻薄、别太不讲理。他说自己尽了力,有几件事确实推动了一点,这给了他继续的动力。

巴黎协定这件事,他过不去。如果他留在委员会里,就等于在说这件事不重要。这件事太重要了。他认为国家得说话算话。

他还指出,巴黎协定本身没有约束力,就算签了,也可以放慢执行。至于“会丢工作”的说法,可以等看到底有没有丢工作再退,说不定还有新增的岗位。

无论如何,那件事之后他没法留下。他说自己尽力了。

主持人收尾,请全场给马斯克鼓掌。

这场访谈信息密度很高,真正值得带走的,是几条能用在具体决策上的判断。下面这几条,可以直接拿去用。

  • 看到一条监管规定,先问一句:它当初是为了解决什么问题?那个问题还存在吗?如果不存在了,就该有人动手把它清掉。别指望它自己消失。
  • 设计任何激励政策,先想清楚它奖励的是哪种行为。成本加成合同奖励的是把成本做大,销售税减免奖励的是把厂建起来。制度设计错了,好人也会被逼成坏人。
  • 判断一项新技术什么时候普及,别只看新产品的产量。要看存量。新车只占车队的5%,车队寿命15到20年。任何“换代”都得把存量算进去。
  • 评估一个地方的产业政策,去查它的可再生能源强制比例、电动车强制比例,别凭印象。中国在这两项上的激进程度,超过很多人的想象。
  • 面对AI这类可能造成系统性风险的技术,先让监管方获得洞察力,再谈具体规则。摸不清现状就下场立规,只会立错。
  • 衡量一家公司值不值这个价,看它把某件必然发生的事提前了多少年。提前十年是大成功,提前五年也算不错。这是马斯克自己给特斯拉定的尺子。
  • 政府做长期项目,尽量用竞争性、按里程碑付款的商业合同,至少保留两家供应商,该换人就换人。单一来源加成本加成,是系统性超支的配方。
YouTube 平台字幕(自动或人工),可能有识别错误,人名与数字请以视频为准。已按语句合并成段落,并去掉 um、uh 等口头填充词。 文稿来源:YouTube(自动/人工字幕)。

0:00I know each Governor has a vision for his or her own States and we have a responsibility to enact policies to help our citizens and our states Thrive so I've asked Elon to join us today to share his thoughts on how Governors can not only stay ahead of the curve but become Innovation Governors so with that

0:23please join me in welcoming Elon Musk hey good see well good afternoon and welcome Elon oh I was going to take off my tie is that all right if I do that I I came in with a tie but then I was like told there bit a tie so then we'll both be more comfortable sounds good well thanks thanks having me appreciate your you're being here today I you know it's when I'm with you it's difficult to know where to start let's start just what drives you what what is it that when you wake up in the morning do you see a problem and you want to solve it I I think the the thing that drives me is that I want to be able to think about the future and you feel good

1:30about that so that you know we're doing what we can to have the future be be as good as possible to be inspired by what is likely to happen and to look forward to the next day so that's that's what really really drives me is is is trying to figure out how do we how do we make sure that things are great and and going to be so and that's the

2:04underlying principle behind Tesla and SpaceX is that I think it's it's it's pretty important that we accelerate the transition to sustainable generation and consumption of energy it's it's inevitable but it's it matters if we have if it happens sooner or later and then SpaceX is about helping

2:30make life multiplanetary and doing what we can to continue the the dream of Apollo and ultimately make a contribution to life becoming multiplanetary well let's talk a little more about that I think everyone very interested that when you say making life multiplanetary that's exciting it is exciting so what's your vision there you know I think you know particularly for Americans you know like think about America is a nation of explorers people came here from other parts of the world that you know chose to give up the known in favor of the unknown so I think exploration

3:19like I think United States is a is a distillation of the human Spirit of exploration and so that's why it it appeals to Americans so much you know you can see this when say there was a shuttle tragedy and seven people died and that's that's terrible but you know a lot of people die all the time but but why do we care so much because it was the dream of exploration that was dying along with those people that's why

3:48no and I'm one of those and probably like many of you remember exactly where you were when that that tragedy happened so you have 30 plus Governors here today and we're very excited about your willingness to be with us and you hopefully heard me talk a little bit about U my initiative which is being ahead of the curve what do you tell us as Governors what we what should we be thinking about in terms of of innovation and and developing public policy for the future well it it sure is important

4:27to get the the rules right and you know it's sort of in in terms of legislative and executive actions you know it's sort of like you you think of say like professional sports or something if you if you don't have the rules right if there isn't you know if if if if the game isn't set up properly it's not going to be a good game so it's really important to get the ru the rules right now I think it's it's worth noting that I think still in the United States the rules are still better than anywhere else but the you know

5:13it's it's very easy to put something in place which is an inhibitor to to innovation without realizing it so in terms of the regulatory environment it's It's always important to bear in mind that that regulations are Immortal and they they they never die unless somebody act actually goes and kills them and then they they get a lot of momentum so a lot of times regulations can be put in place for all the right reasons but then nobody goes back and gets rid of them afterwards when they no longer make sense you know that and there used to be a rule in the early days when people were concerned about automobiles cuz that was a pretty scary thing see a

5:59just going wrong by itself you know you never know what those things might do so there were like rules where you had to in a lot of States we had to carry a lantern in front of the automobile and had to be like 100 Paces ahead of the automobile there had to be someone with a lantern on a pole like okay but you should really get rid of that regulation and they did you know really be awkward so so

6:27just regulations is even if done correctly and for and being right at the time it's always important to go back and and scrub those you know periodically to make sure they're still sensible and they're still serving the greater good I think in terms of tax structure what what is what is economically incented and what is what is not economically incented just make sure that the incentive structure is is correct I I think I'm saying just totally Common Sense things here but it's economics 101 whatever you whatever you incent will happen so

7:05the if you incent one thing that thing will tend to happen more than the other thing if you send another thing that that thing will happen and so the the economics should favor Innovation and and this is particularly important to protect small to mediumsized companies because because it's sort of like trying to grow a tree in a forest like it's real hard for a new company to to grow when it's just a Seedling or a sapling it needs a lot more protection than if it's a giant Redwood or something like that so very important to give support

7:46to small and and small to mediumsized companies on the Innovation front and they're the ones that that needed more than the big companies and I I think this point Tesla is almost big company biggish company anyway so I'd favor you know supporting smaller companies in Tesla relatively speaking what would your response be because there are critics out there with regard to incentives and that and you know Tesla has been and I can speak from experience the the beneficiary of of incentives economic incentives when with regard to the to the gigafactory sure what would you tell those those people yeah I think well first of all as you know the those

8:34incentives were a little overstated the in the case of the gigafactory it's a it's A5 billion investment capital investment to get that factory going and I I didn't actually know this by the by way I didn't I didn't know this until we did the press conference actually that that that over 20 years the V incentives added up to 1.3 billion actually didn't even know this but but but it's now he's telling go ahead literally I learned it at the press conference I'm like really no but I mean it's the the thing is that they they took what added up over 20 years and made it sound like Nevada was writing us a $1.3 billion check and I'm you know I'm still waiting for that

9:21check must did it get lost in the mail I don't know so but you know this is the way the press works of course so if now if you divide 1.3 billion by 20 okay then it's it it's like okay Tesla's on average re receives a sort of a a tax well doesn't it's it's basically sales and use tax abatement is is what it amounts to so T gets like on the order we get on the order of 50 to 60 million of sales and use tax abatement divided over 20 years and but but this is

10:04for something which has a $5 billion Capital cost just get going and then would have to generate about hundred billion doll over that period of time to to achieve a $1.3 billion tax benefit so so essentially it's it's a little over 1% over that period of time and that's great okay but it's not

10:29you know it's not like it's it's not the way it was characterized in the Press but if because if it's put in the proper context it sounds like okay well that's neat you know it's about five 5% helpful on setting up the factory and about 1% helpful over the next 20 years cool that actually sounds pretty reasonable yeah and yeah so so that that that

10:56was that was helpful but there are a lot of other factors as well and we actually had slightly bigger incentive packages from from some other states that were offered but we factored in how quickly could we get the gigafactory into operation what were the risks associated with that progress what would the over what would be the logistics costs over time of transferring battery packs and power trins to a a vehicle Factory in California and you know and all of those factors weighed together is what made is what led us to make the decision in favor of Nevada and and working with with with your team was great I was very forward leaning and

11:49and like a big part was also just like making you know sure you feel really welcome you know within within a state so that's sort of what would L us to make the decision for the gigafactory and then then we have another Factory in in in New York doing solar panels also actually it's will be the biggest solar panel producer in North America when it's St and then we expect to establish probably at least two or three more gigafactories in the US in the next several years as well as what a couple overseas but the overall objective of Tesla it's is is really what what set of actions can we take to accelerate the Advent of sustainable production and

12:44consumption of energy and I think the the the sort of the way the way I would assess the historic good of Tesla is in terms of of how what that how many years of acceleration was it you know and if we can accelerate sustainable energy by 10 years I would consider that to be a great success hope even if it was only 5 years that would still be pretty good that that's the that's the that's the overarching optimization so you you've talked about interplanetary travel and sustainable energy and the vehicles a little bit what what would you want things to look like and five to 10 years associated with with energy and with autonomous vehicles electric

13:39vehicles well I think things are going to be they're going to grow exponentially so there's a big difference between five and 10 years you know my my guess is yeah probably in 10 years more than a half of new vehicle production is Electric in the United States and China's probably going to be ahead of that because China's been super pro e I don't think a lot of people know this but like I mean China's environmental policies are way ahead of the US like their mandate for renewable energy far exceeds the US I think this sometimes people are under the impression that China is either dragging their feet or or somehow behind the US in terms of sustainable energy

14:33promotion but they're they're by far the most aggressive on Earth it's crazy I mean like in fact the a coalition of Chinese car manufacturers Just wrote the Chinese governments to beg for them to slow down the Mandate because like it's like too much they they need to make 8% electric vehicles I think like next year or in two years or something this like they can't physically do it so China is by far the most aggressive on electric vehicles and solar so but that's a common misperception that they're not there one Google search way to figure this out by the way it's really pretty straight pretty easy so and in 10 in 10 yeah 10 years man I think yeah yeah so H half of all

15:21production I think will be be EV I think almost all cars produced will be autonomous in years almost all it will be rare to find one that is not in 10 years that's going to be a huge transformation now the thing to bear in mind though is that new vehicle production is only about 5% the size of the vehicle Fleet so you think about how long does a car or truck last and they last 15 to 20 years so before they finally scrap so new vehicle production is only roughly one at most 115th of the the fleet size so even when new vehicle production say switches switches over to electric or to autonomous that still means the vast majority of the fleet on the roads is not it'll take another you

16:19know 5 to 10 years before that becomes majority the majority of the fleet becomes EV or autonomous but if you were to say go out 20 years overwhelmingly things are electric autonomous overwhelmingly fully autonomous fully autonomous so no one will have to touch the steering wheel if there is one there will not be a steering wheel in 20 years it will be like having a

16:52horse people have horses which is cool but so so having a regular car will be like having a horse is that what you're saying yeah yeah and there will be people that have that have you know non-autonomous cars like people have horses it's just would be unusual to use that as a motor transport yes all right no let's talk about the energy piece and rooftop cell and storage yeah so the I mean

17:28question it's important to appreciate that the Earth is almost entirely solar powered today in the sense that the sun is the only thing that keeps us from being at roughly the temperature of cosmic background radiation which is 3° above absolute zero if it wasn't for a sun we would be a frozen dark Ice Bowl and the the amount of so the

17:52amount of energy that hits the Sun that reaches us from the Sun is tremendous it's it's over it's the it's 99 % plus of all energy that that Earth has then there's there's there's this energy we need to use to run civilization which to us is big but compared to the amount of energy that reaches us from the Sun is Tiny so it it's very easy like it

18:17actually doesn't take much if if you wanted to power the entire United States with solar panels it would take a a fairly small corner of Nevada Texas Utah anywhere look you it's you only need about 100 miles by 100 miles of solar panels it's part of the entire United States and then the the batteries you need to store that energy to make sure you have 24/7 Power is one mile by one

18:52mile one one square mile that's that's it I I I showed the graph the image of this where this is what 100 miles by 100 miles looks like it was like you a little square on the US map and then one there's a little pixel inside there and that's the size of the battery pack that you need to support that real tiny so well you you talked about 20 years from now none of us well some people will still be using horses or or be zero yeah but rare so what will the the energy piece look like I mean what will there be transmission lines will there be a need yeah I think the so

19:36there use of energy can is roughly divided into three areas and they're more or less equal at at a high level there's about a third of energy is used for transportation of various kinds about a third is used for electricity about a third is used for heating so if you want to have and of the electricity production call it you know something on the order of 10% depending upon how you count it is renewable maybe 15% today so that means that there's a massive amount of solar that would need be need to be produced and connected in order to to be fully sustainable because fully sustainable means you're tackling transport non-renewable electricity generation

20:31and heating so that that means that will need to be a combination of utility scale solar and rooftop SC solar combined with wind geothermal Hydro probably some some nuclear for a while in order to transition to a sustainable situation which means really for the most part massive massive growth in solar and it's it's going to be important to have rooftop solar in neighborhoods because otherwise you going there'll need to be massive new transmission lines built and people do not like having transmission lines go through the neighborhood I really don't like that I agree so you you want to have some localized energy production

21:26combined with utility so rooftop solo utility solar and that that's that's really going to be the solution from a physics standpoint that I can't see any other way to really do it people talk a lot about fusion and all that but the sun is a giant Fusion reactor in the sky and it's really reliable comes up every day so if it doesn't we got bigger

21:56problems somebody asked me to ask you this we talked about Workforce today but they asked me are robots going to take our jobs everybody's jobs in the future or H how how much do you see Artificial Intelligence coming into the the workplace well first of all I think on the artificial intelligence front you know I I have exposure to the very very most Cutting Edge AI and I think people should be really concerned about it I keep sounding the longell but you know until people see like robots going down the street killing people like they don't know how to react you know because it seems so ethereal and I think we should be really concerned about Ai and I think we

22:48should this is AI is a rare case where I think we need to be proactive in regulation instead of reactive because I think by the time we are reactive in AI regulation it's too late and nor normally the way regulations are set up is that a whole bunch of bad things happen there's a public outcry the the and then after many years a regulatory agency is set up to regulate that industry and there's a bunch of opposition from companies who don't like being told what to do by Regulators anyway it takes forever

23:28that that in the past has been bad but not something which represented a you know a fundamental risk to the existence of civilization AI is a fundamental risk to the existence of human civilization in a way that car accidents airplane crashes faulty drugs or bad food were were not they were not they they were harmful to to a set of individuals within Society of course but they were not harmful to society as a whole AI is a fundamental existential risk for human civilization and I don't think people fully appreciate that you know it's not it's not fun being regulated it's not you know be pretty irksome but you know in the car business we know we get regulated by

24:31Department of Transport by EPA and a bunch of others and and there regulatory agencies in every every country you know in the in space we get regulated by FAA and but but you know if you ask the average person hey you want to do you want to get rid of the FAA and just like take a take a chance on manufacturers not cutting corners on the aircraft because you know profits were down that quarter I was like hell no

25:04that sounds terrible so you know I think even people who are pretty extremely like libertarian free market they be like yeah we should probably have somebody keeping an eye on the aircraft companies making sure they build a good aircraft and good cars and that kind of thing so you know I think there's there's a role for Regulators that's very important and I'm against overregulation for sure but man we I think we've better get on that with AI Pronto and so there'll certainly be a lot of job disruption because what's going to happen is robots will be able to do everything better than us I'm in I'm including I mean all of us you know yeah not sure exactly what to do about

26:04this it's like the it's like this is really like the scariest problem to me I'll tell you and yeah so I really think we need government regulation here just to because this is you know ensuring the public good is served because you got companies that are racing that they kind of have to race to build AI or they're going to be made uncompetitive you know like essentially if your competitor is racing Dev World Ai and you don't they will crush you so then you're like ah we don't want to be crushed so you know I guess we need to Bull it too that's where you need The Regulators to come in and say hey guys you all need to really you know just

26:54pause and make sure this is safe and like when when it's cool and we're and regulator convinced that it's safe to proceeded then you can go but otherwise slow down and but long but you kind of need The Regulators to do that for for all the teams in the game you know otherwise the shareholders will be saying like hey why aren't you developing AI faster because your competitor is like okay we better do that anyway so it's like I mean there's like something like 12% of jobs are transport transport will be one of the first things to go fully autonomous but when I say everything like the Rob us will be able to do everything bar bar nothing let's move back to you're

27:42rolling out the model 3 this year right and how many orders what what is that going to look like yeah it's going well on that front we got I don't know more I think like if somebody orders a model 3 today they'd only get it probably late next year we just actually just started production made the first production unit last week and the thing that

28:11is is not well appreciated about something about about cars and any kind of new technology is how hard it is to do the manufacturing is vastly harder to do the manufacturing by a factor of a hundred like a hundred than to to make the to make that car to make one of something W with with maybe 50 or 60 people we could make a prototype of practically anything in 6 months now to manufacture that thing we need 5,000 people to spend you know 3 years and that's considered really fast so manufacturing will does this kind of S curve where it's excruciatingly slow at first and then it it grows exponentially and then but people tend to extrapolate on a straight line so if

29:03it's real slow at first they say oh this is real slow look at that they're only going to make five cars a week forever like NOP there'll be 10 cars a week then 20 cars a week then you know 40 cars a week then 5,000 cars a week eventually just grows Crazy Fast so we're hoping to get to you know something you know like 5,000 cars a week by the end of the year well I want to give an opportunity for some of the governors to ask questions and perhaps some audience questions I was told that you'd be willing to to do that great so Governors any questions for for Elon governor Scott well thank you very much we and Vermont have partnered with Tesla in in terms of a power pack in

29:50in our homes and it's for $15 a day you can rent this for 15 years and it'll it'll carry power as a back up generation device for 12 hours and it's been really really interesting from my perspective but I'm curious about vehicles and and where we're going in the future or how far in the future do the cars themselves become the charging device like the the roof and deck lids and and hood or does or do the batteries get so efficient that you don't need that and then you just power up for a week or something like that where are we going in the future with battery storage yeah I think the future is it's there's just there's three legs to the stool there's a electric cars

30:35there's a stationary battery pack and solar power with those three things you can have a completely sustainable energy future that's all that's all it's needed on the Sol on the solar front like I said it's going to be a combination of rooftop solar and utility scale solar you'll need both because of the you know enormous demand for electricity and then you know one of the things that's that's been missing I think up till now is having rooftop solar that looks good and isn't you know that's where we got this the solar glass roof that we're developing and we're doing it in different styles so that it it you know it matches the Aesthetics of a particular house or

31:19so Regional style that's I think that's actually pretty important and the conventional flat panels solers for for flat roofs and for commercial will be the way the way to go but yeah it's and and and putting solar panels on the on the car itself not that not that helpful Because the actual surface area of the car is not not very much and cars are very often indoors and so it's the least efficient place to put solaras on the car just wondering about maybe a wrap of some sort is that does that make any sense in the future a a wrap of solar around either a building made of a solar panel or a wrap of a of a vehicle actually being the solar panel but being

32:06the the components of the vehicle itself I I don't think so I'll scrap that idea no it's just it's just way better to put her on the roof for sure and I I really thought about this I mean really and I pushed my team about like isn't there some way we could do it on the car I mean technically if you have like some sort of Transformer like thing which will pop out of the trunk like like a you know like a hard top convertible and just like like ratchet solar panels over the whole surface area of the car extending like for the entire say square footage of a parking space provided you're in the sun that would be enough to generate about 20 to 30 m a day of electricity but

32:58that is for sure the expensive difficult way to do it Governor bergo so thought about maybe we should Elon thank thanks for being here with your background in payment systems you understand the important role of security and transactions now that you've got I think security is a huge concern like cyber security yes and you're in a in in the vehicles you're building now are incredibly complex software systems I mean the car is really a rolling piece of software it's like a laptop on Wheels yes so share with us a little bit about your thoughts on cyber security and how you how how how we protect you talk about protecting society when you got a rolling Fleet

33:40of I I think one of the biggest risks for autonomous vehicles is somebody achieving a fleetwide hack you know in principle if if somebody was able to hack say all of the autonomous Teslas they could say I mean just as a prank they could say like santal to Rhode Island from across the United States and I'll be like well okay that would be the end of Tesla and there be a lot of angry people in Rhode Island that's for sure so we got to make super sure that

34:16that a fleet white hack is basically impossible and that if people are in the car that they have override Authority on whatever the car is doing so if the car is doing doing something wacky you can press a button that no amount of software can override that will ensure that the you you you gain control of the vehicle and kind of cut cut the link to the servers so that's that's pretty fundamental within the car we actually have even if somebody gains access to the car there are multiple subsystems within the car that that that also have specialized encryption so the power Trin for example has specialized encryption so even if some of you were gain access to the car

35:00they could gain access to the power train or to the braking system and but it is my top TP concern from a security standpoint at Tesla is making sure that fleetwide hack or any vehicle specific hack can occur the the same the they have the same problem with cell phones you know we it's kind of crazy today that we live quite comfortably in in a in a world that George Orwell would have thought is super crazy like we

35:35all carry a phone with a with with a microphone that could be turned on really at any time without our knowledge with a GPS that knows our position and a camera and well kind of all of our personal information we do this willingly and it's kind of wild to think that that's the case so so pH the the phone like apple and

36:07and Google kind of have the same challenge of making sure that cannot be a fleetwide hack or a systemwide hack of phones or or a specific hack so that that's our top our top concern yeah become it's going to become a bigger and bigger concern like Tesla's I don't want to Temp fate here but Tesla's Tesla is pretty good at software compared to the other car companies and so I do think it's going to be a bit like an even bigger challenge for for the other car companies to ensure security yeah thank you Governor

36:48Dugard thank you Governor Mr musk thank you for speaking to all the governors today it's it's an honor to have you here one question I had we saw when gasoline prices went to $35 a gallon there was a big jump in interest in hybrid vehicles and you saw those Vehicles become very much in demand and then as gasoline prices have fallen you've seen a reversal of that and I'm wondering to what extent you have a concern about the future of electric vehicles in the face of those very low prices can you speak to that well the economics

37:28they they they kind of set set the slope of the the the curve so there's no question in my mind whatsoever that all transport with the ironic exception of rockets will go fully electric everything Planes Trains automobiles well a lot of trains are already electric all shifts and but it's question of what that time frame is and the economic incent

38:03structure drives that time frame that's really what it amounts to you know this there the and the Big Challenge is that there's an unpriced externality in the cost of fossil fuels so the unpriced externality is the the the probability weighted harm of changing the chemical constituency of the atmosphere and oceans it's it's since it is not captured in the price of gasoline it does not Drive the right behavior you know it would be like if tossing out garbage was just free and

38:53you know there was no penal you just do as much as you want then like trees be full of garbage so and we regulated a lot of other things like sulfur emissions and nitrous oxide emissions and that kind of thing it's done done a lot of good on that front with CO2 it's tough because there's so many vested interest on the sort of fossil fuel side and sometimes I think I feel like those guys feel like kind of hard done by cuz you know it wasn't obvious like when they were creating their oil and gas companies that it would be bad for the environment and they worked really hard to create those companies and then they feel like well now they're being kind of attacked on moral grounds

39:38when they didn't originally start those oil companies or or build them up on on bad moral grounds and and and it is true that we cannot instantaneously change to a sustainable situation but then those guys will also fight pretty hard to slow down the change and that's really where I think is morally wrong go to Governor Bavon and then Governor Hutchinson then we'll take a couple then Governor hickin Looper and then we'll take some audience questions Governor bevon El Elon thank you for being here short version of the question then slightly longer the short version is do you ever feel pressure by others expectations of you and your endeavors in light of the progress you've made

40:25thus far is the short version and and and more specifically when you look just at Tesla alone and you look at a company with a54 billion valuation and seemingly by typical Mar Market metrics no justifiable reason for that what are you saying does I'm just saying I'm curious sir I'm just in all seriousness do you feel a a concern ever that your intellect in your intellectual curiosity and your Ingenuity cannot be matched by those that are trying to commercialize it does that ever affect how you think or decisions that you make well it it is actually I find it quite tough when there are very high expectations I try to actually Tamp down those expectations as you know

41:16to be possible in fact I've gone a record several times as saying that the stock price is higher than we have any right to deserve and that's for sure true based on you know where we are today and have been in the past so the stock price obviously reflects a lot of optimism about where Tesla will be in the future and now the the thing that makes that

41:44you know quite a difficult emotional hardship for me is is that you know those expectations sometimes get out of out of control and like I hate disappointing people and so I'm like trying real hard to meet those expectations but that pretty tall order and a lot of times it's really not

42:09really not fun I have to say a whole lot less fun than it may seem so yeah I mean I don't ever sell any stock unless I have to for for taxes so you know I said publicly I'm not going to like take money off the table you know I'll be last I'm going down I'm going down with the ship so I'll be the last to do

42:40it but yeah I mean it's I really wouldn't recommend anyone start a card company I really wouldn't recommend it it's not a recipe for happiness and freedom thanks Governor Bon Governor Hutchinson Mr musk ASA Hutchinson from Arkansas thank you for your Frank observations about exploration you know I look at the spirit of in invention and the spirit of exploration which is really the Hallmark of America what is your comment on NASA its Mission I was in Congress I

43:22supported NASA but I always feel like it's floundering does not have the support of the American people that's needed What what's your comment on NASA its Mission and what advice would you give us sure well first of all I should say I'm a big fan of NASA in fact at one point my password was I love NASA literally that was my password

43:47and you know I think the n NASA does a lot of good things for which PE for which it doesn't get enough credit and that the public I guess doesn't know that much about like a lot you most members of the public they're not really into hard science you know it's like not it's not the the thing they're tuning in for most of the time I love hard science you know but it's not that popular

44:22so but there's great things in terms of the the telescopes like the Hubble and the James web and the you know the robers on Mars and the the pro you prob outer solar system those are like really great things but to get the public excited you got to get people in the picture it just it's just a hundred times different if there are people in the picture and you know if there's some criticism of NASA it's like I it's like important to remember people in the picture you know if you want to get the public support and but like if you talk to a scientist about that said like where's the science in that like you're not getting it it's

45:11like that's not why people giving you money it's not that's I mean it's a little bit of the reason but the like the the the serious scientists or like people just make things more expensive like why do we have people like okay well why do we have people at all anywhere sometimes the scientists are the ones who just don't don't understand even they're like smart people but like you know so you got to have something that's going to fire up the you know fire fire people up and get them really excited and like I think if we had a serious goal of having a base on the moon and sending people to Mars and said okay this is we're going to be outcome oriented how are we going to

45:55do this okay we got to changed the way Contracting is done you can't do these like Cost Plus contract Cost Plus Soul Source contracts because then the incentive structure is all messed up so as soon as you don't have any competition well okay there's no the St of urgency goes away and as soon as you make something a Cost Plus contract you're incanting the contractor to maximize the costs of the program because they get a percentage so they never want that gravy train to end and they want to make it and a they become cost maximizers and then you have good people engaged in cost maximization because you just gave them incentive to do that and told them they'll get

46:36punished if they don't essentially that's what happens so it's critically important that we change the Contracting structure to be a competitive commercial bid make sure there there are always two at least two entities that that are competing to serve NASA and that the contracts are mil Stone based with with concrete Milestones PowerPoint presentations do not count like everything works on PowerPoint okay you say I have a teleportation device look here's my PowerPoint presentation so Milestone based competitive

47:14commercial contracts with with competitors and then and then you got to be prepared to fire one of those competitors if they're not if they're not cutting it and and recompete the rest the remainder of that contract and by the way NASA's actually already done this and they did it with the Comm with the commercial cargo transportation to the space station and that was a case where NASA you know that NASA actually I'm not if they thought it would work or not work but they didn't have the budget to do anything else so they're like okay we're going to try this competitive commercial Milestone based Contracting and it worked great and they Ed it

47:55to two companies to to SpaceX and a company called kler and SpaceX managed to meet meet the Milestones kler did not so then they NASA recomed the remainder of the contract to orbital Sciences but then obal Sciences got across the finish line so now NASA's got two suppliers for taking cargo to the space station and it's a great situation same thing for commercial crew of the space station now as to competed that in in in the commercial crew case it's space SpaceX and Boeing and I think that's also a good situation so now like I can tell you like the SpaceX team is like we're going to do this before Boeing that's for sure and then like I bet at the Boeing team

48:40they're like we're going to do this before SpaceX that's good it's it's a good forcing function to get things done that that I can't tell you how important that Contracting structure is that is night and day there's way too much in in government which is where it's a sou Source Cost Plus contract that that just again economics 101 whatever you incent that will happen and people shouldn't be surprised it's like oh you just you know said Okay if that company manages to find some excuse to double the cost of the contract they're going to get double the profit because they're getting a percentage so they're going to do they're can do exactly that

49:23and and also they're not going to say no to requirements so the government will come up with some set of requirements 90 90% of them could make a lot of sense and 10% of them I'll coam that double the price of the of of the the project but those 10% of cocki requirements in a Cost Plus contract the contractor will always say yes there could be a future for you in in Government Contracting at the state level yeah let's go to Governor hicken Looper and then Governor Ducey so I think like most governors find so refreshing to have the unbridled truth but I do suspect every time you say publicly that the stock price is higher than we have any right to believe I I going to guess

50:05you probably get some calls from investors suggesting that maybe you don't say that so frequently yeah that's true I wanted to go back and just just briefly because I think I I wrote this down that you said that artificial intelligence is the the fundamental existential risk facing civilization do I get that close enough enough in my opin it is it is the biggest risk that we face as a civilization is artificial intelligence and so to a group of leaders what would you advise that we should how should we be addressing something that's that's such a large landscape and yet obviously so important I think that the you know one of the roles of government is to

50:46ensure the public good and and to that dangers to the public are addressed so that hence the regulatory thing I think the first order of business would be to try to learn as much as possible you know to understand the nature of the issues to look closely at the progress that is being made and the remarkable achievements of artificial intelligence I mean last year

51:18go which is a quite a difficult game to beat that people thought would never be beaten with by by a computer that that that computer would either never beat the best human player or that it was 20 years away and last year alphao which was done by Deep Mind which is a kind of a Google subsidiary absolutely crushed the world's best player and now now that now it can crush it can play the top 50 simultaneously in questionable so just like that pace of progress is remarkable

52:01and and there's you can see more and more coming out like the robotics you can see robots that can learn to walk from from nothing you know within hours like way faster than any biological being but the the thing that's most dangerous is and it's the hottest to kind of wrap kind of get your arms around because it's not a physical thing is kind of a deep intelligence in the network he said well what home could a deep intelligence in the network do so well I could start a war by create by doing fake news and spoofing email accounts and fake press releases and just by you

52:57know manipulating information the pen is mighter than the sword so I mean as an example I want to be I want to emphasize I do not think this actually occurred this is purely a hypothetical that I I'm digging my grave here but you know that like that there was that second Malaysian airliner that was shot down on the Ukrainian Russian border and that

53:31that really Amplified tensions between Russia and the EU in a massive way well like let's say if you had an AI that was where the ai's goal was to maximize the value of a portfolio of stocks one of the ways to maximize value would be to go long on defense short on consumer start a war

53:57and then how could it do that well you know ha can the Malaysian Airlines R aircraft routing server a route it over a war zone then send an anonymous tip that an enemy aircraft is flying overhead right now let's go to Governor Ducey and then we'll have after Governor Ducey we'll finish our U gubernatorial questions and then two questions then we quick questions or one audience question and then we'll be done we're we're running short on time Governor Ducey thanks Elon I really enjoyed your comments today and as someone who has spent a lot of time in his administration trying to reduce and eliminate regulations I was surprised by your suggestion to bring reg

54:48regulations before we know exactly what we're dealing with with AI you know and i' I've heard the example used if I were to come up with a colorless odorless tasteless gas that was explosive people would say well you have to ban that and then we'd have no natural gas so you've given some of these examples of how a AI can be an existential threat but I still don't understand as policy makers what type of regulations beond slow down which typically policy makers don't get in in front of entrepreneurs or innovators well I think the first order of business would be to gain Insight right now the government does not even have Insight and I I think the the right

55:39order of business would be to stand up regulatory agency initial goal gain insight into the status of AI activity make sure the situation is understood once it is then put regulations in place to ensure Public Safety that's it and for sure the companies doing AI will most of them not mine will squark and say hey this is really going to stifle Innovation blah blah blah it's going to move to China it won't and it won't because like

56:16it's like has like has Boeing moved to China nope they bowling aircraft here same on cars and so it's not it's the notion that if you establish a regulatory regime that companies will simply move to countries with with lower regulatory ecomment is is false on the face of it because none of them do and unless it's really overbearing but that's not what I'm talking about here just talking about you know making sure that there is awareness at the government level I think once there is awareness people will will be extremely afraid as they should be right one audience question we'll take the first hand that came up right here thanks Elon Ena freed with axios

57:16early on in this Administration you had argued pretty viously that it was best to engage and better to be in the room than not be in the room then when the president decided to pull out of Paris you said that was kind of the last draw and you were going to drop off what drove you to that and if you were still speaking to him today what would you say to the president well I I thought it was worth doing you know trying hard to you know to do what worth it was worth trying I got a lot of flank from multiple fronts for even trying when some guy ran at Billboards and like attacking me and like full page ads in the New York Times and whatnot just for just for being on the panel

58:07and and you know in every in every meeting I was like just trying to make the arguments in favor of sustainability and you know sometimes other issues like we need to make sure that our immigration laws are not unkind or unreasonable and you know did my best and I I think in a few cases I did actually make some progress which gave me some encouragement to continue but but then I I just really

58:38think that the Paris ACC Court man I I'm I'm if I st on the councils then I'd be essentially saying that that wasn't important but it was super important because I think the country needs to keep his word and you know that that's is not even a binding agreement so we could always like slow it down the argument that there would be job losses well we could see if there are job losses before we exit the agreement and maybe there won't be job losses maybe there'll be job gains but yeah there's just no way I could stay on after that so you know did my

59:20best all right all right well everybody if you would please join me in thanking Elon for being here today right thank you thank you well that was a treat wasn't it

资源与出处

相关视频

同频道、同主题或相邻年份