Latest / Elon Musk Podcast / TESLA 2025 Annual Shareholder Meeting | Tesla

Transcript

TESLA 2025 Annual Shareholder Meeting | Tesla · Elon Musk Podcast

• 0:08
  I I declare this meeting is now closed.
• 0:11
  Elon Musk. Thanks guys.
• 0:43
  First of all, I'd like to just give a heartfelt thanks to
• 0:46
  everyone who supported the, the shareholder votes.
• 0:50
  I super appreciate it. Thank you everyone.
• 0:55
  I'd like to take thank the, the Tesla board for their immense
• 0:59
  support. We have a fantastic board,
• 1:01
  fantastic group of shareholders. Thank you all.
• 1:07
  And, and, and what we're about to embark upon is not merely a
• 1:11
  new chapter of the future of Tesla, but a whole new book.
• 1:17
  And I'm going to talk about that.
• 1:20
  So this, this really is, this really is going to be quite the
• 1:25
  story. And optimist is a fundamental
• 1:29
  part of that. The, the, the sheer scale of
• 1:33
  optimists. I mean, I'm going to say a bunch
• 1:38
  of things that probably I shouldn't say, you know, but but
• 1:45
  but that's what keeps it interesting.
• 1:48
  I mean, have you have you watched any other annual
• 1:51
  shareholder meeting? I mean, honestly, it was like, I
• 1:55
  mean, if you need to go to sleep, sure.
• 1:57
  I mean, that other shareholder meetings are like snooze fest.
• 2:01
  I mean, but ours, ours are bangers.
• 2:03
  I mean, look at this. This is sick.
• 2:10
  And we got like the the cyberpunk nightclub here with
• 2:13
  real robots just standing there and mulling around and dancing
• 2:19
  And in a round out our engineering headquarters in Palo
• 2:23
  Alto, the robots just walking around the office 24/7 with no
• 2:27
  one minding them. They just and then they go
• 2:29
  charge themselves. And yeah, so the, the, the scale
• 2:36
  of, of, of optimists, I like I said, that's really going to be
• 2:39
  something else. I think it's going to be the
• 2:42
  biggest product of all time by far.
• 2:47
  Yeah. So like bigger than cell phones,
• 2:54
  bigger than anything. I guess the way to think about
• 2:58
  it is that every human on Earth is going to want to have their
• 3:02
  own personal R2D2C3PO. So who wouldn't?
• 3:10
  But actually Optimus will be even better than them.
• 3:12
  You know, like, you know, RTD to you, it's kind of beep at you
• 3:15
  and it's kind of hard to figure out what he's got talking about.
• 3:19
  You know, you can see 3PO to translate, but Optimus is going
• 3:23
  to be like everyone's going to want one.
• 3:27
  I think within terms of industry providing products and services,
• 3:32
  I think it's probably, I don't know, three to five robots in
• 3:36
  industry for every, you know, one that's a personal robot.
• 3:40
  I, I think there could be 10s of billions of, of optimist robots
• 3:44
  out there now. Obviously it's very important we
• 3:48
  pay close attention to safety here because we, we do, we do
• 3:53
  want the, the, the Star Wars movie, not the Jim Cameron
• 3:58
  movie. I love Jim Cameron's movies, but
• 4:03
  you know, you know what I mean? So yeah, yeah.
• 4:17
  So, so we're, we're going to launch on the, the fastest
• 4:24
  production ramp of of any product of any large complex
• 4:28
  manufactured product ever and starting with building 1,000,000
• 4:34
  unit production line in, in Fremont.
• 4:39
  And that's, that's, that's line 1.
• 4:43
  And then a 10 million unit per year production line here on the
• 4:48
  I don't know where we're going to put the 100 million unit
• 4:58
  production line maybe on Mars, I don't know.
• 5:04
  But, but I think it's, it's going to literally get to 100
• 5:08
  million a year, maybe even a billion a year.
• 5:12
  And you know, people often talk about like eliminating poverty,
• 5:19
  giving everyone amazing medical care.
• 5:23
  Well, there's actually only one way to do that, and that's with
• 5:26
  the Optimist robot. With, with humanoid robots, you
• 5:29
  can actually give everyone amazing medical care.
• 5:34
  So in terms of optimists will be more precise, optimists will
• 5:41
  ultimately be better than the best human surgeon with a level
• 5:44
  of precision that that isn't possible, that that is beyond
• 5:47
  human. So I think that's that's a
• 5:51
  pretty wild concept to say, OK, if you, you know, there's all
• 5:56
  people always talked about eliminating poverty, but
• 5:58
  actually optimists will actually eliminate poverty.
• 6:02
  Optimists will actually give people incredible medical care.
• 6:14
  So I mean, and so you start getting like sort of some pretty
• 6:20
  wild sci-fi sort of scenarios where in some of these things I
• 6:29
  say will always be taken out of context and using snippets and,
• 6:32
  you know, center output, whatever.
• 6:33
  I was only going to say, you know, like, I think we might,
• 6:36
  may be able to give people a more if somebody's committed
• 6:40
  crime, a more humane form of containment of future crime,
• 6:45
  which is if if you, if you say, like you now get, you now get a
• 6:49
  free optimist and it's just going to follow you around and
• 6:52
  stop you from doing crime. But other than that, you get to
• 6:56
  do anything just it's just going to stop you from committing
• 7:00
  crime. That's, that's really it.
• 7:03
  You know, you don't have to put people in like prisons and
• 7:05
  stuff. I think it's pretty wild to
• 7:07
  think of the various of all the possibilities, but I think it's
• 7:10
  it's clear it's clearly the future.
• 7:14
  And you know, my book recommendation for the maybe the
• 7:20
  best mostly utopian sci-fi future are the N Banks books,
• 7:24
  the culture books. So if you're curious, like what
• 7:27
  what do I think the future is probably like?
• 7:28
  I think it's probably a bit like that or, you know, as mob to
• 7:35
  some degree. But but but I think it's and
• 7:37
  Heinlein, but in banks, if if you're like saying what, what,
• 7:42
  what, what does Elon think the future probably will be like for
• 7:46
  AI and robots? It's kind of banks in so now
• 7:56
  and, and, and things do get kind of wild from an economic
• 7:59
  standpoint, because at a certain point with AI and robotics, you
• 8:07
  can actually increase the global economy by a factor of 10 or
• 8:11
  maybe 100. There's there's not like an
• 8:14
  obvious limit. So like, optimist is kind of
• 8:22
  like an infinite money glitch, you know, and maybe they won't
• 8:29
  even be money in the future or money, but it might be measured
• 8:34
  in terms of wattage, like how much, you know, power can you
• 8:39
  bring to bear from an electrical standpoint.
• 8:44
  So I guess what I'm saying is hang on to your Tesla stock.
• 8:59
  So yeah, man, it's pretty wild. You're welcome.
• 9:08
  Let's see, I think we got some slides to go through.
• 9:12
  We see ad Lib a lot of this stuff.
• 9:14
  But so the, the, you know, when we started Tesla, the, the, the
• 9:21
  goal was to accelerate the advent of sustainable energy and
• 9:25
  that is that's what we've done. I mean, Tesla has really led the
• 9:29
  way in with, with electric vehicles, with battery packs,
• 9:33
  with a lot of solar. And many other companies have
• 9:37
  then followed our lead and done that.
• 9:39
  And electric cars which used to be non existent are now
• 9:43
  prevalent. And, and the Model Y, for
• 9:48
  example, is the number one selling car of any kind on
• 9:51
  earth. But obviously now with, with AI
• 10:04
  and robotics, we need to update our mission and our mission.
• 10:11
  I think it's good. It's, it's, it's, it's a great
• 10:13
  mission, which is to achieve sustainable abundance, which is,
• 10:18
  it's like, because I often ask people like, what is the future
• 10:21
  that you want? What's the best future you can
• 10:24
  imagine? You know, because we want to try
• 10:27
  to make that future, like make the best future you can imagine.
• 10:31
  And, and I guess probably the best future is if people, you
• 10:38
  know, can have whatever they want from a goods and services
• 10:39
  standpoint or medical standpoint.
• 10:43
  And, but at the same time, we, we don't destroy nature and we,
• 10:47
  we keep the rainforests and the, the, the beautiful national
• 10:50
  parks and, and all the, all that stuff.
• 10:52
  And so that's, that's what I mean by sustainable abundance,
• 10:56
  is that people, you know, can have whatever they want, have
• 10:59
  all the needs met, but we still keep all of the natural beauty
• 11:03
  that we want. I mean, if somebody can think of
• 11:13
  a better future, I'm all ears. But I think that's that's
• 11:17
  probably the best, the best way to go.
• 11:20
  So, yeah. See.
• 11:24
  So, yeah, as you know, every test is designed to be
• 11:30
  autonomous. So the it's so nice.
• 11:34
  Difficult to explain to people if they've not.
• 11:36
  In fact, I'm sure you've all encountered this where you try
• 11:39
  to tell people that the Tesla can drive itself and they they
• 11:42
  think you're crazy or something. I mean, especially like apart
• 11:47
  from the cyber truck, our cars look pretty normal.
• 11:51
  I mean, they're good looking cars, but they don't look super
• 11:54
  then they look normal. But I guess it's kind of like
• 12:01
  having a cat or like, and, and the cats are sitting.
• 12:05
  Let's say you've got a cat and it's like just sitting there on
• 12:07
  the couch and you try to tell people that the cat can
• 12:11
  actually, it's actually Puss in Boots and, and it, and it can
• 12:15
  actually put on boots and a hat and, and swashbuckle and sing
• 12:19
  and dance. And people are like, no way,
• 12:20
  man, that's a cat until the cat does all those things and you're
• 12:26
  like, damn, what the. So we've got, you know, millions
• 12:34
  of Tesla cars out there that are the kind of like person boots
• 12:39
  that they're intelligent, but people don't know that they're
• 12:42
  intelligent. They look like normal cars, but
• 12:45
  actually they're super smart and can drive themselves.
• 12:48
  So as I think that's probably the single biggest thing we need
• 12:52
  to do is to educate potential customers that you can either
• 12:57
  have a car, you can either have a cat that's like normal cat, or
• 12:59
  you can have person boots and person boots is very cool.
• 13:04
  So that's, so we've these days when people come to our stores
• 13:11
  and, or even people that have the car haven't turned it on, we
• 13:14
  find, and sometimes people have paid for FSD and haven't turned
• 13:18
  it on. We're like what you should at
• 13:19
  least try it once. And so now we like the sales
• 13:25
  team, the service team will actually sit with customers and
• 13:28
  say, look, let us show you how it works and how easy it is.
• 13:32
  And then once they've tried it for even just a few days, they,
• 13:37
  they can't live without it. And, and, and now with version
• 13:41
  14, we're actually getting to the point where we almost feel
• 13:47
  comfortable allowing people to text and drive, which is kind of
• 13:51
  the killer app because that, that's really what people want
• 13:54
  to do and, and do do. And actually right now the, you
• 14:02
  know, the versions, you know, the cars a little strict about
• 14:08
  keeping your eyes on the road. And but I'm confident that in
• 14:14
  the next month or two, we should we we, we're going to look
• 14:18
  closely at the safety statistics, but we will allow
• 14:22
  you to text and drive essentially.
• 14:30
  Yeah. So yeah, that's that's, it's
• 14:43
  certainly about the current situation, which often people
• 14:45
  will actually turn off FSD to text, then turn it back on,
• 14:49
  which is less safe. So, so yeah, that's probably the
• 14:54
  single biggest thing is, is just get people aware of FSD and then
• 14:59
  obviously we need to get it approved in in Europe.
• 15:01
  So we certainly appreciate the support of our customers in
• 15:04
  Europe pushing the regulators to approve FSD because you can't
• 15:07
  even get a super, even just normal supervised FSD is not
• 15:11
  allowed in Europe currently, which doesn't make any sense.
• 15:14
  And I've had these like crazy conversations with the
• 15:16
  regulators that seem like a Franz Kaka novel where where I'm
• 15:20
  like, well, look, we have, you know, billions of kilometers of
• 15:25
  data that shows that FSD increases safety.
• 15:29
  And they're like, well, we have to have all these committee
• 15:31
  meetings. I'm like, yeah, but people's
• 15:32
  lives are at stake here. So definitely a pressure from
• 15:36
  our customers in Europe to to push the regulators to approve
• 15:39
  would be appreciated. And then we have partial
• 15:43
  approval in China. And we hopeful, hopefully will
• 15:45
  have full approval in China around February or March or so.
• 15:49
  That's what that's what they've told us.
• 16:00
  Yeah, but yeah, the, the, the fact that every Tesla car is
• 16:06
  capable of full self driving, every car we build and have
• 16:08
  built for the last several years is capable of full self driving
• 16:12
  is, is pretty wild. And most people don't don't know
• 16:14
  that. So, and then we've got the the
• 16:19
  first car that is specifically built for unsupervised full self
• 16:25
  driving to be to be a robotaxi. It's called a cyber cab.
• 16:30
  It doesn't even have pedals or steering wheel.
• 16:35
  Yeah. So notice there's no side view
• 16:41
  MERS. There's no yeah.
• 16:43
  So it's it's and it's very much optimized to minute for for the
• 16:46
  lowest cost per mile in our in our autonomous mode.
• 16:51
  And that production is happening right here in this factory and
• 16:55
  we'll be starting production in April next year.
• 17:06
  So the, the way that Cyber cab is designed is it's, it's, it's
• 17:12
  designed obviously for a purely autonomous world, but also the
• 17:18
  manufacturing system is unlike any other car.
• 17:24
  It manufacturing system of the cyber cab.
• 17:28
  It's, it's sort of it's closer to a high volume consumer
• 17:32
  electronics device than it is a car manufacturing line.
• 17:36
  So the net result is that I think we should be able to
• 17:40
  achieve, I think ultimately less than a 10 second cycle time
• 17:46
  basically unit every 10 seconds, maybe ultimately take a few
• 17:52
  years to get there, but it's theoretically possible to get to
• 17:55
  a 5 second production time. And so, So what that would mean
• 18:03
  is you could get on a, on a line that would normally produce, say
• 18:09
  500,000 cars a year at a one minute cycle time, like Model Y,
• 18:17
  the this, this would be maybe as much as, you know, two or three
• 18:24
  million, maybe ultimately, you know, it's theoretically
• 18:28
  possible to achieve a 5 million unit production line if you, if
• 18:32
  you can get to the 5 second cycle time.
• 18:35
  So it's a lot of cars. So these, these will be
• 18:49
  everywhere in the future. So I wanted, we wanted to look
• 18:59
  futuristic. So like it changes the look of
• 19:01
  the roads Now the yeah, the ingredients, when you look at
• 19:15
  what Optimus is, what's required to make Optimus and the various
• 19:20
  ingredients, what do you need to do to make, to do high volume
• 19:24
  humanoid robot production? I think it's worth considering
• 19:28
  that really the cars we make are already robots, but they're 4
• 19:32
  wheeled robots. So Tesla is already the biggest
• 19:35
  robot manufacturer in the world because every car we make is a
• 19:38
  robot. And when you break it down to
• 19:41
  the fundamental elements you've got, you've got batteries, power
• 19:43
  electronics, motors, gearboxes, you've got, you know,
• 19:52
  connectivity, you've got a vision based AI high Optimus and
• 20:02
  you know all the various pieces that you need for a humanoid
• 20:05
  robot. You need the AI chip, you need
• 20:07
  the AI software, you need to be able to manage a large fleet.
• 20:12
  And and so really Optimus is a robot with arms and legs as
• 20:18
  opposed to as opposed to a robot with wheels.
• 20:22
  So. You know, Tesla's ideally
• 20:26
  suited, I think to, to make, to succeed in this arena.
• 20:34
  You, you, you will see certainly many companies showing
• 20:38
  demonstration robots. There's really three things that
• 20:41
  are super difficult about robots.
• 20:43
  One is the engineering of the forearm and hand, because the
• 20:47
  human hand is an incredible, incredible thing.
• 20:50
  Actually, it's super dexterous. So engineering the hand really
• 20:57
  well, the real world AI and then volume manufacturing, those are
• 21:04
  generally the things that are missing one or more of those
• 21:08
  things I'm missing from other companies.
• 21:10
  So Tesla is the only one that has all three of those.
• 21:14
  Yeah. So yeah.
• 21:24
  So this is the Optimus kind of initial, it's a kind of
• 21:30
  prototype production line. The the high volume production
• 21:34
  line will be very automated, obviously.
• 21:36
  But this is this is really the production line that we use to
• 21:39
  make the prototypes. So you can get a sort of rough
• 21:42
  sense for what it takes to build the robot.
• 22:02
  To pull the finger and then as I've said before, I
• 22:18
  think once we reach about 1,000,000 units per year of
• 22:22
  sustained production or in excess of that, I think probably
• 22:25
  the cost of production is around $20,000 in current year, current
• 22:32
  year dollars. So this will be certainly very
• 22:37
  affordable. And yeah, like I said, I think
• 22:43
  Optimus will ultimately increase the size of the economy probably
• 22:49
  by a factor of 10 or more. You know, next year we saw
• 22:54
  production with Optimus version 3.
• 22:57
  This what you're seeing here is Optimus version 2.5.
• 23:00
  Optimus 3 is, is an incredibly good, good design.
• 23:06
  The tells the engineering team is amazing.
• 23:10
  When you see Optimus 3, yeah, you it will seem as though that
• 23:23
  there's someone like a like a person in a robot outfit, which
• 23:27
  is how we started with Optimus. It really is going to be
• 23:32
  something special. And then Optimus, Optimus 4, you
• 23:38
  know that that hopefully starts production in 27 and then
• 23:42
  Optimus 5 and 28. So it's kind of like an annual
• 23:45
  release cycle with significant improvements with each one and
• 23:50
  and gigantic increases in the scale of production.
• 23:57
  So yeah, sustainable abundance via AI and robotics.
• 24:03
  That's the future we're headed for.
• 24:14
  And as as I think most people here know, the safety statistics
• 24:19
  show that miles driven on FSD are much safer than miles driven
• 24:25
  with without it. So what this will translate to
• 24:29
  ultimately is saving the lives of millions of people and
• 24:32
  preventing hundreds of millions of accidents.
• 24:35
  So a massive increase in the in, you know, lives sold and and
• 24:42
  tragedies avoided. It's going to be amazing.
• 24:54
  And how many people here have tried 14.1?
• 24:58
  OK, all right. Cool.
• 25:02
  Yeah, You can see the even with the point releases, it's getting
• 25:05
  quite a bit better. It should be pretty smooth at
• 25:07
  this point. But really 14.2 is that there
• 25:13
  are major changes, 14.2 and then 14.3.
• 25:17
  And I think I think by 14.3 is when we'll really be at the
• 25:22
  point where you can just pretty much fall asleep and wake up at
• 25:26
  your destination. So, and then I've been putting a
• 25:38
  lot of time into the new Tesla chip design because in order to
• 25:43
  have a functional robot, you have to have a great AI chip and
• 25:48
  it needs to be an inexpensive chip and it needs to be very
• 25:51
  power efficient. So we, we, we believe the AI5
• 25:56
  chip will be probably about 1/3 of the power of say something
• 26:02
  like a Blackwell and a video Blackwell, which is great, a
• 26:06
  great chip for roughly A comparable performance and about
• 26:12
  and, and much less than 10% of the cost.
• 26:16
  So. This is a chip that is that is
• 26:24
  very much optimized for the Tesla AI software stack.
• 26:28
  So it's it's not meant to be a general purpose chip.
• 26:31
  It's meant to be an amazing chip for the Tesla AI software.
• 26:36
  And I mean a couple things that I think make like how is Tesla
• 26:42
  able to achieve such an improvement?
• 26:44
  I think it is because because we are specialized, we're not
• 26:48
  trying to, you know, NVIDIA has to serve the superset of all
• 26:52
  past and future customers. So all of their requirements,
• 26:55
  all of the software that they've written has to work, which is a
• 26:58
  very difficult problem, whereas we just need to make it work for
• 27:02
  our software. And and so we were able to
• 27:07
  simplify the chip dramatically. And then we also, I think, I
• 27:13
  think we're unique in this, but like we, we have an integer
• 27:16
  based system and integer operations are fundamentally
• 27:21
  more efficient than floating point operations.
• 27:24
  So we can do floating point, but we, but the vast majority of of
• 27:28
  our inference is done in integer, which is if you're
• 27:33
  familiar with sort of logic gates like the simplicity of of
• 27:37
  integer. And it's integer is much more
• 27:42
  power efficient, much more silicon efficient.
• 27:45
  But you have to, you actually have to train for integer
• 27:49
  inference, which everyone else is training for floating point.
• 27:53
  Those that's kind of like a niche technical detail, but it's
• 27:56
  actually very important. So yeah, this this is, this is
• 28:02
  going to be a great chip. So this, this chip will be made
• 28:09
  in basically in four places, TSMC Taiwan, Samsung Korea, TSMC
• 28:16
  Arizona and TSMC Texas. And we already know what
• 28:28
  improvements to make for AI6. So I'm hopeful that we can,
• 28:34
  within less than a year of AI5 starting production, we can
• 28:37
  actually transition in the same fab to AI 6 and double all of
• 28:43
  the performance metrics. So I'm super hardcore on chips
• 28:51
  right now. As you may be able to tell, I
• 28:55
  got chips on the brain. I dream about chips literally.
• 29:03
  I can, I can, I can draw the the, the, the, you know, the,
• 29:08
  the at least the broad brush stroke physical design of the
• 29:10
  AI5 chip by off by heart at this point.
• 29:16
  It's it's, it's a good chip. It's a good chip, Sir.
• 29:23
  So this this is this is really key.
• 29:25
  Now one of the things I'm trying to figure out is how do we make
• 29:31
  enough chips? So, you know, I have a lot of
• 29:40
  respect for our the the, the Tesla partners, TSMC and
• 29:45
  Samsung. Yeah, you know, maybe we'll,
• 29:48
  we'll, we'll do something with Intel.
• 29:50
  We we haven't signed any deal, but it's probably worth having
• 29:53
  discussions with Intel. But even when we extrapolate the
• 30:00
  best case scenario for chip production from our suppliers,
• 30:04
  it's still not enough. So I think we may have to do a
• 30:11
  Tesla Terafab. So it's like Giga, but way
• 30:20
  bigger. I, I can't see any other way to
• 30:26
  get to the volume of chips that that we're looking for.
• 30:31
  So I think we're probably going to have to build a gigantic chip
• 30:36
  fab. Got to be done.
• 30:51
  So anyway, some of the stuff I've already talked about, Yep,
• 30:56
  we've, we've done a tremendous amount for sustainable energy
• 31:01
  and that that's only going to grow over time.
• 31:06
  The world is moving towards a, a solar battery economy, which is
• 31:12
  ultimately the, that's, it's where that's where it was going
• 31:15
  to go anyway. But what tells us does
• 31:17
  accelerate, accelerate that outcome?
• 31:22
  Sometimes people don't understand quite how much energy
• 31:25
  comes from the sun. So the sun is 99.8% of all mass
• 31:30
  in the solar system. Jupiter being point 1% and then
• 31:34
  point 1% is miscellaneous. Earth being in the miscellaneous
• 31:37
  category. So the total amount of energy
• 31:41
  like, so sometimes people what say like, well, we'll build the
• 31:45
  fusion reactors on Earth. It's like, well actually the
• 31:48
  giant fusion reactor in the sky is basically impossible to beat
• 31:52
  to such degree that even if you could burn Jupiter in a
• 31:56
  thermonuclear reactor, the amount of energy produced by the
• 31:59
  sun would still round up to 100%.
• 32:02
  That's how much energy the sun produces.
• 32:04
  So solar power is necessarily the future and I think there's
• 32:13
  going to be a lot of solar powered AI satellites and I
• 32:19
  think Tesla is going to play a role in that.
• 32:30
  We've obviously refreshed the product line.
• 32:32
  So S3 XY if if you haven't tried the Model S3X or Y or the cyber
• 32:40
  truck, I recommend at least getting a test drive or a test
• 32:43
  ride as the case may be. Try out the full self driving
• 32:47
  and I think you'll be blown away to those who do not.
• 32:50
  If you might be listening and don't have a Tesla, you should
• 32:51
  try one. So and of course we've got the
• 32:59
  cyber truck, which is the toughest truck of all time.
• 33:03
  It's literally bulletproof, faster than a Porsche 9/11 and
• 33:07
  can out tow a Ford F-350. So it's a great car, great
• 33:13
  truck. And then starting next year, we,
• 33:26
  we manufacture the Tesla semi. So this, this is already we
• 33:36
  already have a lot of prototype Tesla Semis in operation.
• 33:42
  PepsiCo and other companies have been using the Tesla Semi for
• 33:44
  quite some time. But we will start volume
• 33:47
  production at our Northern Nevada factory in 2026.
• 33:55
  So we get, we get two big products or three, three massive
• 33:58
  products starting production next year.
• 34:00
  We got Optimus, we got Tesla Semi and we got the cyber cap.
• 34:13
  And then battery packs. So the, if you, if you look at
• 34:20
  total US power generation capability, it's roughly A
• 34:25
  terawatt, but the average power usage is less than half a a
• 34:30
  terawatt. And that's because the, the big
• 34:33
  differences in power usage between day and night.
• 34:37
  So the, the, the, the daily and seasonal variations in power
• 34:41
  consumption mean that the United States and, and really every
• 34:45
  country is only using about half, is only producing about
• 34:48
  half as much electrical energy as it could, because without
• 34:52
  batteries, there's no effective way to buffer the energy.
• 34:56
  So what batteries actually enable is even if you don't
• 35:00
  build any incremental power plants, you could double the
• 35:03
  energy output of the United States just with batteries.
• 35:08
  This is a super big deal. And in fact, I think that's
• 35:12
  really where most of the incremental energy production in
• 35:15
  the United States is going to come from is literally
• 35:17
  batteries. So a bigger deal than a may
• 35:20
  seem. And then we've got we we keep
• 35:30
  improving the battery design so the the the mega block, which
• 35:33
  makes it really easy to deploy battery utility scale batteries.
• 35:40
  So we've we've just simplified and and brought more of the
• 35:43
  components to be internal to the battery.
• 35:45
  So you can just show up and drop off a battery and it works.
• 35:50
  And then hopefully with the wonder how to, hopefully over
• 35:55
  time, we will actually add more and more of the power
• 35:58
  electronics so that Megapack will actually be able able to
• 36:02
  output up to 35 kilovolts directly.
• 36:07
  So you won't need a substation. That's what I'm saying.
• 36:09
  You can just literally drop it off kind of like the way that a
• 36:11
  power wall, you just connect it to the house.
• 36:14
  The utility wires go on one side and the other side goes to the
• 36:17
  house mains and that's it. So we want to get mega pack to
• 36:21
  the point where you just literally take the utility wires
• 36:24
  and you plug them in and it just works.
• 36:35
  Then we've got the, we've also built the world's largest
• 36:38
  Supercharger network. So we do a lot of things here at
• 36:40
  Tesla. That's the biggest Supercharger
• 36:46
  network in the world by far. And ultimately you'll be able to
• 36:51
  go anywhere on earth using a Tesla Supercharger.
• 36:54
  And it's pretty close to anywhere on earth.
• 36:56
  But it's, you know, it's going to be ultimately just anywhere.
• 37:00
  It'll just work anywhere. So the Supercharger team has
• 37:05
  done, done great work expanding that and improving the
• 37:07
  efficiency of the Supercharger network.
• 37:11
  And in, in North America, they did such a good job that the,
• 37:15
  the other car companies basically said, well, they'll
• 37:18
  just use the Tesla Supercharger network.
• 37:20
  Like like, OK, sounds good to us.
• 37:31
  You know, it's always important to have a, a slide on safety in
• 37:34
  the factory. So we continue to improve safety
• 37:37
  for our factory workers. We're, we care a lot about their
• 37:40
  well-being. And you know, like one way you
• 37:44
  can just tell, like if a company's a good company or not.
• 37:47
  If you, if you just walk through the factory or walk through the
• 37:49
  office and catch the vibe and the vibe in the Tesla factory
• 37:53
  is, is good. People are happy.
• 37:56
  You know, that's how you know it's a good company.
• 38:04
  We've also put a lot of investment into raw materials.
• 38:08
  So the, we've built in in South Texas and Corpus Christi, the
• 38:18
  biggest lithium refinery outside of China, I believe.
• 38:22
  So it's, it's going to, it's starting off with 50 GW hours
• 38:27
  of, of, of lithium and we'll expand from there.
• 38:31
  So this is a very, very important to have, you know, in
• 38:38
  a worst case scenario that we have the ingredients necessary
• 38:40
  to make a battery very important.
• 38:44
  And and then we've got here on this site the the cathode
• 38:49
  factory, which is just the sort of giant building about 1/2 mile
• 38:54
  that way. And we're just making sure that
• 38:57
  we from a supply chain standpoint are resilient against
• 39:01
  any potential geopolitical challenges.
• 39:11
  And then also at this factory, we also make the 4680 cell which
• 39:15
  is getting better and better. And that 4680 cell will be used
• 39:21
  in the cyber cab Cyber it is being used in the cyber truck
• 39:25
  and will be used in the cyber cab and also in Optimus.
• 39:29
  So that that's going well. But we continue obviously to get
• 39:34
  cells from many suppliers. It's kind of like the the chip
• 39:37
  AB thing. We'll we'll take as many chips
• 39:39
  as our suppliers can provide us. But then beyond that, if they
• 39:43
  can't provide us with any incremental sales or chips, we
• 39:46
  we kind of have to make them ourselves or we get stuck.
• 39:52
  That's the Tesla Semi factory. So it's going to be pretty cool
• 39:57
  to see these going down the road and scale.
• 40:10
  Yeah. So that's, that's the, that's
• 40:12
  the basic plan, sustainable abundance for all.
• 40:28
  All right. So yeah, I guess we can go into
• 40:32
  Q&A now. Maybe at the next annual
• 40:40
  shareholder meeting, we'll have Optimus take some of the
• 40:42
  questions. That'd be cool.
• 40:55
  So let's see, Alexander. Well, thank you for your help,
• 40:57
  by the way. Yeah, please go ahead.
• 41:06
  Thank you. Is that a mic?
• 41:08
  Is it? No.
• 41:09
  Let me hold it. It's OK.
• 41:11
  Well, what a year has been, right?
• 41:13
  I mean, just a year ago, President Trump got elected on
• 41:16
  the same day. These 12 months have been a heck
• 41:19
  of 12 months. Thank you to the board because I
• 41:21
  think they had to weather quite some storms.
• 41:25
  Institutional investors. And obviously, thank you to you.
• 41:28
  And we stand with you. I think we showed it and we will
• 41:32
  show it. So we would like to express A
• 41:43
  wish, please. This is a nice venue.
• 41:45
  We love coming to the Factory and I think it's actually great
• 41:48
  we're doing this at the Factory, but there are thousands of
• 41:51
  retail investors who are crying because they cannot.
• 41:54
  They come to me. I love playing the Mama, but I
• 41:57
  now give it to you. Please organise a bigger venue.
• 42:01
  We're bigger than Berkshire and we will do better than
• 42:04
  Berkshire. OK, sure we can.
• 42:06
  Pay. Make it up.
• 42:09
  Yeah, actually Tesla will be way, way, way bigger than
• 42:13
  Butcher. Butcher long term, it's going to
• 42:15
  be kind of nutty. So all right, we'll do the next
• 42:21
  one. Maybe we'll do it in the Yeah,
• 42:24
  like there. Yeah, maybe the, the, the
• 42:27
  downtown arena or the, the, the soccer stadium or something like
• 42:30
  that. And get your security check.
• 42:32
  We want to make sure that you're safe.
• 42:34
  What if I just Bob and weave? But you know, we want a party.
• 42:40
  And if you. Can bring the other Elon
• 42:41
  companies there as well. OK.
• 42:43
  Investors would really appreciate.
• 42:45
  All right, sounds good. Thank you very much for
• 42:46
  listening. All right.
• 42:56
  Elon, as a father, I just want to 1st thank you for the
• 42:59
  everything you're doing in the world, especially freedom of
• 43:02
  speech. Thank you.
• 43:03
  And one question will will you or Tesla ever consider FSD tied
• 43:11
  to an owner's account rather than a vehicle to encourage more
• 43:14
  frequent upgrades, provided the transfer is to only a brand new
• 43:17
  Tesla vehicle while fostering brand loyalty?
• 43:20
  If the vehicle is sole the trader without upgrade to
• 43:23
  another Tesla, FSD ownership would end.
• 43:28
  Well, we we have done that a few times I guess.
• 43:34
  I guess we could extend it again.
• 43:38
  All right, We'll extend it. We'll extend it for at least
• 43:41
  another quarter and then and then play it by year after that.
• 43:45
  All right. Howdy Elon, Congrats on not
• 43:52
  having to show up to work for free anymore.
• 43:54
  Yeah, yeah. We now have over a $40 billion
• 44:06
  war chest. We're cash flow positive and
• 44:09
  remain that. We know how you feel about Fiat
• 44:13
  already. Is it time to take a look at
• 44:15
  Bitcoin? What's your belief on that?
• 44:16
  Also, you've hinted towards that there's a wheelchair accessible
• 44:21
  model in the works where you're referring to the Rabovian.
• 44:25
  And if so, can we please speed up production to help the least
• 44:28
  fortunate? Sure, Yeah, the obviously we
• 44:34
  need a vehicle that's big enough to to fit a wheelchair,
• 44:37
  wheelchair accessible. So I think that that's that's
• 44:43
  that is the Rebobin or robust or whatever we call it.
• 44:48
  I mean, the it's not like we we're slowing down because we
• 44:52
  want to slow down. It's it's like we're, you know,
• 44:54
  we're spending like a zillion plates here.
• 44:58
  So but I do, I do think it would be very cool because I think
• 45:02
  aesthetically as well, it's just, it just would change the
• 45:04
  look of the roads and make it feel like the future.
• 45:06
  So, so won't be long before we make that, but it's, but since
• 45:12
  we do have optimists, you know, the, we've got cyber cab
• 45:17
  optimist and semi all next year. It'll will probably have to be
• 45:23
  maybe the year that you know, a couple years from now or
• 45:25
  something like that. But we certainly will make a
• 45:28
  wheelchair accessible vehicle. Hi, Satyoshi.
• 45:40
  I was, I was privileged enough to attend the Investor Day and
• 45:44
  you've also talked about the factory being a product and I
• 45:47
  saw the dry cathode method and and so forth.
• 45:49
  And I wanted to know the progress in that and also would
• 45:53
  often be working on that in the future production line in the
• 45:56
  future. Yeah, I guess the dry cathode,
• 46:00
  man, that's turned out to be a lot harder than we thought.
• 46:03
  So I mean, it, it does look like it's going to be successful and
• 46:10
  it will have some cost advantage relative to wet cathode.
• 46:14
  But if I had to wind the clock back, I would probably have gone
• 46:16
  with wet cathode instead of dry cathode because it, it just
• 46:19
  turned out to be a lot harder to make it, you know, high capable
• 46:24
  of high volume production with, with super high reliability.
• 46:28
  So, yeah, but we will be, we will be scaling up battery cell
• 46:35
  production at Tesla and looking for cell production from our
• 46:39
  suppliers as well because we're, we're going to be ramping up
• 46:43
  production very dramatically at Tesla.
• 46:45
  So now, now that we, we, we believe we have full self
• 46:51
  driving that we have autonomy solved, you know, or, or at
• 46:54
  least or within a few months of having it unsupervised autonomy
• 46:59
  solved at a reliability level significantly better than human.
• 47:03
  It's not that that means it's time to ramp up production
• 47:08
  because the value proposition is now much greater than a regular
• 47:10
  car. The killer app really is for
• 47:13
  people. Can you text and drive or can
• 47:15
  you sleep and drive? You know, can you can the car
• 47:20
  take you to your destination or do you need to pay attention and
• 47:23
  be and have to drive it? And before we allow the car to
• 47:30
  be to be driven without paying attention, we need to make sure
• 47:33
  it's very safe. But, but like I said, we're,
• 47:36
  we're on the cusp of that. I know I've said that a few
• 47:41
  times, but, but we really are at this point and, and you can feel
• 47:46
  it for yourselves with, with the, the 14.1 release.
• 47:51
  So the, you know, the what we're, we're we're going to try
• 47:56
  to, we're going to push to expand vehicle production as
• 47:59
  fast as we possibly can. So aspirationally we'd aim to
• 48:04
  increase vehicle production by about 50% by the end of next
• 48:07
  year. So, yeah, so that's, it's, it's
• 48:16
  very hard to increase production, but that's that's
• 48:18
  roughly, you know, I don't know, maybe we get to like I'm just
• 48:22
  guessing at an exit rate by the end of next year of around 2.62
• 48:28
  point 7 million vehicles annualized production.
• 48:32
  And then aim to get to maybe 4,000,000 by the annualized rate
• 48:38
  by the end of 27 and then maybe 5,000,000 by the end of 28.
• 48:42
  That those are rough. Those are our aspirational
• 48:45
  goals. So these are this, this is a
• 48:48
  gigantic increase in, in output, which means that the entire
• 48:52
  supply chain has to move in, in, in unison with that, with that
• 48:58
  increase in, in volume. And the nature of the nature of
• 49:03
  producing a large complex product is that it moves as fast
• 49:06
  as the least lucky, dumbest element in the entire system.
• 49:09
  And there's 10,000 plus items. But, but this, like I said, this
• 49:15
  really is a new, it's not just a new chapter for Tesla, it's a
• 49:18
  new book. And and that new book is
• 49:22
  massively increasing vehicle production and ramping up
• 49:26
  optimist production faster than any things ever been wrapped up
• 49:30
  before in history. So.
• 49:40
  Awesome, Elon, thank you. Thank you from all of us.
• 49:44
  You know, the retail shareholders really, really
• 49:48
  care. And so I, I echo the sentiment
• 49:49
  of, you know, a larger venue where more can, where can come,
• 49:55
  you know. As a testament.
• 49:57
  To that, I was here. I was fortunate enough to be
• 49:59
  here last year and since then touring the factory, talking to,
• 50:04
  you know, people that work here. I'm just a retail shareholder,
• 50:07
  but increased my holdings 12 times, right?
• 50:10
  So I, you know, I know it's really engaging and gives people
• 50:14
  a confidence and, and so thank you for having us and I do echo
• 50:17
  that. I hope it grows last year.
• 50:21
  I'm the one that asked you about your well-being and safety.
• 50:24
  Oh yeah, full love and. You know, there's a broad, you
• 50:34
  know, general question and it was before, you know, some of
• 50:37
  the uncertainty that's unfolded since then.
• 50:39
  And so I hope it had a positive, you know, impact and you know,
• 50:44
  we all care. And so my question this year,
• 50:51
  you've talked recently about the most mind blowing product demo
• 50:55
  of all time and I shot it being the most memorable product ever.
• 51:01
  Roadster. Yeah, yeah.
• 51:03
  And I have patiently been waiting on my Founders Series
• 51:09
  Roadster reservation. So on behalf of the Founders
• 51:12
  Series guys that have stuck in there, can we be invited to this
• 51:18
  unveil? Oh, yeah, sure, absolutely,
• 51:20
  definitely. OK, yes.
• 51:22
  So all Founders series can be invited.
• 51:24
  Yes, yes. I mean, it's the least we could
• 51:26
  do frankly, for people that have long our long-suffering roasted
• 51:31
  reservation holders. The I, I, I I feel confident
• 51:38
  saying it will be the most exciting product unveil ever and
• 51:46
  I hope that whether it goes well or doesn't go well.
• 51:53
  Since I'm up here, can I get the first one?
• 51:57
  Well, I, I guess the, it's, it's according to who whoever put
• 52:02
  down their deposit in, in that sequence.
• 52:05
  So that's the, but you'll get, you'll get a very early 1.
• 52:10
  And I mean, the, the, the new roaster is very much sort of
• 52:14
  like, it's not even the icing on the cake.
• 52:16
  It's the cherry on the icing on the cake.
• 52:19
  So I mean, it's really kind of like, you know, it's, it's, it's
• 52:24
  like it's not essential for sustainable abundance, let me
• 52:26
  put it that way. But I do think there should be,
• 52:33
  you know, very cool technology in the world that is that, you
• 52:38
  know, it's way beyond anything that's ever existed.
• 52:41
  And I think I'd like those like, even if I could never have
• 52:45
  access to those things, I'd like to know that they exist and, and
• 52:48
  see the future happening. So I think I think it will be
• 52:51
  inspiring to a lot of people and and just it's it's extreme.
• 52:57
  It's like the coolest car, if it even is a car that is ever that
• 53:04
  will probably ever exist. Yeah.
• 53:07
  So. Hi, Elon.
• 53:12
  Congrats on the proposal plan. That was amazing.
• 53:15
  The compensation question back to chips.
• 53:17
  So chips, chips, chips. Chips, chips, chips, chips,
• 53:20
  chips will be the limiting factor to.
• 53:22
  The to the future so. On that, chips and electricity
• 53:25
  are the two limiting factors, yeah.
• 53:27
  And we got the energy and power and energy packs ready to roll.
• 53:30
  So I think on the chips I think he said TSMC, Samsung, perhaps
• 53:33
  Intel 14A and several different sites.
• 53:36
  I heard a couple US factory sources.
• 53:37
  Which is, I mean, we're like, everyone, everyone, you know,
• 53:40
  it's like, yeah. Yeah.
• 53:42
  And and so I guess the, is there an open door in the future of
• 53:45
  investing directly into some of those boundaries?
• 53:48
  And second question, how big is a Terra factory?
• 53:51
  Can you put that in scale? And good point, that's a.
• 53:54
  Semiconductor Terra factory. Yeah.
• 53:58
  Well, I mean, The thing is that we actually have agreed to buy
• 54:02
  all the chips that are made from the fab, you know, so it's
• 54:07
  basically a money printing machine for TSMC and Samsung.
• 54:11
  It's like literally the faster you make the chips, the faster
• 54:14
  we send them money. But it's still not going at the
• 54:18
  fast enough. So that's why I think as far as
• 54:25
  I can see, the only option is to go build some like very big chip
• 54:31
  fab and then you go to solve memory and packaging too.
• 54:35
  So, but otherwise you just, you just tap out at whatever the,
• 54:40
  the chip reduction rate is. And so I guess Tara would be
• 54:45
  you, you'd want to say it's, it's got to be at least 100,000
• 54:48
  wafer starts per month size fab and maybe that would be 1 of 10
• 54:54
  in a complex. So ultimately be a million wafer
• 54:58
  starts per month. Yeah, exactly.
• 55:02
  You can tell when you when it gets the giggle factor.
• 55:04
  That's probably a good sign that that we're onto something
• 55:08
  special. But I, I wouldn't be surprised
• 55:12
  if long term it's like a million wafers a month.
• 55:16
  Yeah. Hi LON.
• 55:21
  I'm a long term investor, so I've been holding the shares for
• 55:25
  12 years. I, I worked at the Autopilot
• 55:27
  team development for 10 years and I brought my dad here from
• 55:31
  Brazil. We actually won the Tesla Vision
• 55:34
  contest with a Cyber Truck B video.
• 55:36
  So I hope you get to see that at some point.
• 55:39
  So thanks for the free Model Y. And my question is about
• 55:43
  obviously market expansion, but not only South America, Brazil,
• 55:47
  but also into Mars, like with the upcoming rendezvous of Mars
• 55:50
  and Earth. What's going to be in the
• 55:52
  payload, you know? What are we going to send there?
• 55:56
  Well, Optimus is going to play a big role.
• 55:59
  Optimus, and I mean Tesla vehicles will play a big role on
• 56:02
  the moon and Mars. So for a moon base and a Mars, a
• 56:06
  Mars city, Tesla vehicles and and Optimus robots are natural,
• 56:13
  natural fit for building and operating a moon base and a Mars
• 56:19
  city. So sorry.
• 56:22
  Cyber truck also. Yeah, yeah, Cyber truck you will
• 56:27
  need to drive around and pressurize vehicle if there's a,
• 56:29
  you know, a person inside, but yeah, it it'll be it'll be
• 56:34
  something cool next level moon buggy or Mars buggy.
• 56:39
  So all right. Hi, Mr. Musk, Gorev, been a fan
• 56:45
  of yours since 2006. I think I saw a piece in Popular
• 56:50
  Science magazine. It was really cool and I think I
• 56:52
  was 15 at the time, shareholder since 10 years now.
• 56:56
  It's been really, really rewarding.
• 56:58
  It's changed my life. So thank you very much to you
• 57:01
  and Tesla. My question is regarding Tesla's
• 57:05
  mission for a sustainable future as it pertains to autonomy.
• 57:10
  I personally fully expect a deflationary period after well
• 57:16
  unlocked by autonomy basically both in moving humans and goods
• 57:21
  and whatnot, excluding Optimus even.
• 57:27
  What efforts are is is Tesla going to focus on to reduce the
• 57:32
  dollar per mile? Rough math, I think like $0.30 a
• 57:37
  mile would be really nice, $0.50 is pretty good too.
• 57:42
  I mean, looking at inflation later on.
• 57:44
  And then I guess the second part of that, how low does it have to
• 57:48
  be for people to just stop buying cars?
• 57:51
  Like where it doesn't make economic sense to do that?
• 57:55
  I expect that to affect economies of scale and then
• 57:58
  further increase the pricing of vehicles they're on out.
• 58:01
  So like how? Like what is that node at which
• 58:04
  inflection happens? Well, in terms of cost per mile,
• 58:08
  I mean we do see a path with a lot of work to get below $0.20 a
• 58:12
  mile in current year dollars. And, and I somewhat agree that
• 58:21
  those things will probably be deflationary as productivity
• 58:24
  increases because you can think of money supply as being the
• 58:27
  ratio of goods and services to the like what's, what's the
• 58:32
  growth in goods and services to versus the growth in the money
• 58:34
  supply. That ratio is basically
• 58:36
  inflation. And, and so if the output of
• 58:40
  goods and services grows faster than the money supply, then you
• 58:42
  necessarily have deflation or vice versa.
• 58:45
  They try to make it sound complicated, but it's not.
• 58:49
  So I'm not sure that that even government over overspending can
• 58:55
  actually, I think government overspending will will not be
• 58:59
  what will be lower than the increase in productivity of
• 59:01
  goods and services, which would imply deflation as you expect,
• 59:08
  sorry. What about economies of scale
• 59:10
  and the impact it has to how you respond as an automotive company
• 59:14
  when people stop potentially buying cars?
• 59:17
  Economies. Sorry, economies of scale.
• 59:19
  In like auto manufacturing for instance, it only makes sense if
• 59:22
  lots of people buy your vehicles.
• 59:24
  Oh, the total number of vehicles will decrease.
• 59:29
  You know, the, the, the sort of vehicle fleet out there is about
• 59:32
  2 billion cars, but two, 2 billion if you add up all cars
• 59:35
  and trucks that are not at a scrap yard, I think it's around
• 59:37
  2 billion. But the that, that number would
• 59:41
  decrease with autonomous vehicles.
• 59:44
  So the total fleet size would drop.
• 59:47
  I actually think miles driving will increase because it is now
• 59:50
  much less painful to travel somewhere.
• 59:53
  So if, if somebody's thinking about traveling across a busy
• 59:56
  city, then they'll take into account, well, how much pain do
• 1:00:00
  I have to deal with if I have to go through two hours of traffic,
• 1:00:02
  I probably won't, won't do it. But if you're just say sitting
• 1:00:06
  in a cyber cab watching a movie or doing some work, then it's
• 1:00:10
  just like sitting in a little lounge and, and I, I, so I think
• 1:00:14
  you'll see probably a significant increase in total
• 1:00:17
  miles driven, but at the same time a decrease in the total
• 1:00:21
  active fleet of vehicles. I just like to reiterate how
• 1:00:28
  grateful we all are to have you on board, ready to lead us to
• 1:00:32
  another 7 trillion in market cap at least.
• 1:00:35
  At least obviously. Thank you.
• 1:00:39
  My question is, how much of A concern is it that when Cyber
• 1:00:42
  Cab starts production in Q2 next year, that regulation won't be
• 1:00:46
  there yet to where you can deploy Cyber Cabs being produced
• 1:00:49
  or are you guys confident that every cyber Cab you guys make
• 1:00:53
  you'll be able to deploy? Yeah, I think, I think the rate
• 1:00:57
  at which we'll we receive regulatory approval will roughly
• 1:01:00
  match the rate of cyber cab production.
• 1:01:05
  It'll be maybe a little tight, but that's it's, it's about
• 1:01:09
  right. And I'd like to thank thank
• 1:01:12
  Weymouth for paving the the path here.
• 1:01:15
  It's very helpful. So, yeah, but it, I, I think
• 1:01:21
  we'll be able to deploy all the, all the cyber cabs that we
• 1:01:24
  produce. And the other thing is like,
• 1:01:29
  once it becomes like extremely normal in cities, it's just
• 1:01:32
  going to become like the regulators will have just fewer
• 1:01:36
  and fewer reasons to say no. And then you've got this, you
• 1:01:40
  know, we've got the, the accident statistics at scale and
• 1:01:43
  you can show that autonomous miles save lives then.
• 1:01:48
  And you've got unequivocal, you know, billions of miles to prove
• 1:01:52
  it. Then I think it's it's hard for
• 1:01:54
  regulators to say no. Yeah, True.
• 1:02:00
  All right. First of all, Elon, thank you so
• 1:02:03
  much for everything you do for the Tesla shareholders.
• 1:02:06
  Thank you. Do you see a path for optimist
• 1:02:11
  to have consciousness downloaded to it?
• 1:02:15
  You mean human consciousness or? Yeah, I mean, it's not like
• 1:02:23
  immediate, but if you say it down the road, would you be able
• 1:02:29
  to say with a neural link, have a snapshot of or what is an
• 1:02:36
  approximate snapshot of somebody's mind and then upload
• 1:02:41
  that approximate snapshot to an optimist body?
• 1:02:46
  I think, I think that at some point that technology becomes
• 1:02:51
  possible and it, it it's probably less than 20 years.
• 1:02:57
  Yeah. But of course you won't quite be
• 1:02:59
  the same, you know, be a little different because you'll be in a
• 1:03:04
  robot body and the, the, the the mental snapshot will not be
• 1:03:09
  precise. It'll be probably pretty close,
• 1:03:13
  but not, not exactly the same. On the other hand, you know, are
• 1:03:16
  you the same person that you were five years ago?
• 1:03:18
  Nope. I mean, a lot of things have
• 1:03:19
  changed. So yeah.
• 1:03:23
  But I guess at some point if you want to be uploaded to a robot
• 1:03:27
  body, my guess is that becomes possible.
• 1:03:35
  Hi Elon, space based data center is a great idea and I agree with
• 1:03:39
  that. I was curious on your views on
• 1:03:43
  space based solar power. The idea of beaming down
• 1:03:45
  microwave energy down to earth so that you can just point it to
• 1:03:49
  where it needs to go without transmission, without
• 1:03:51
  distribution mines and it's more energy dense so you know, less
• 1:03:56
  land use. Just curious on your views.
• 1:04:00
  Well, space solar power and and until the advent of something
• 1:04:04
  like Starship where the cost per ton to orbit, you know, drops by
• 1:04:11
  orders of magnitude, the cost of getting payload to orbit is so
• 1:04:15
  high that there was no possibility of space solar power
• 1:04:18
  solving anything really. Now with with Starship, like I
• 1:04:26
  see a path to Starship cost per ton to orbit being lower than
• 1:04:30
  air flight. Like if you were to fly, you
• 1:04:33
  know, do a long distance air flight with cargo, like say took
• 1:04:38
  a 747 from here to, you know, Sydney, Australia or something
• 1:04:42
  like that. I think, I think ultimately
• 1:04:47
  Starship will be able to do that trip for less cost per ton than
• 1:04:53
  an aircraft. So then you, so now that now
• 1:04:56
  you've got that, that opens up a, a, what very wide range of
• 1:05:02
  possibilities. Like the, the most obvious one,
• 1:05:06
  I think is, is actually solar powered AI satellites.
• 1:05:12
  So to, to move the AI to, to, to orbit and, and essentially deep
• 1:05:19
  space over time, because the, the you can actually access over
• 1:05:25
  a billion times more energy from the sun in deep space than you
• 1:05:30
  can on Earth. The, the scaling if you to scale
• 1:05:36
  to Carter ships any to make any progress on a Carter ship to
• 1:05:40
  scale, which is using some non trivial amount of energy from
• 1:05:46
  the sun, you kind of have to do space solar power.
• 1:05:52
  Now you, you could only beam a tiny amount of that back to
• 1:05:55
  Earth or you would melt Earth. So Earth actually receives a
• 1:06:01
  very, very tiny amount of of the sun's energy.
• 1:06:03
  We're I mean, Earth is a tiny dust mote.
• 1:06:07
  You see, like Earth to scale with the sun, we look like a
• 1:06:11
  little crumb. So the the to scale
• 1:06:17
  civilization, like said to be at all relevant on a cottage, have
• 1:06:21
  to scale like to even use a millionth of a percent of of the
• 1:06:28
  the sun's energy. You you really have to be have
• 1:06:32
  your solar power in in deep space.
• 1:06:35
  And and you could beam some of that back to Earth too, but you
• 1:06:40
  can only beam a tiny bit of it back or you'd melt Earth.
• 1:06:45
  And this will be our last question.
• 1:06:48
  We'll do OK. We'll do a few more.
• 1:06:49
  We'll do a few more. Sorry.
• 1:06:51
  I'm sorry. All right, Ilan, look around
• 1:06:54
  you. This is a very, very small
• 1:06:56
  sample of Tesla soldiers. Maybe actually I should say Elon
• 1:07:02
  Musk soldiers. Well, thank you for your
• 1:07:04
  support. So thank you everyone.
• 1:07:06
  So I'd like to say thank you from the bottom of my heart for
• 1:07:11
  your support. Thank you and thank you everyone
• 1:07:13
  out there. So I I know for sure a lot of
• 1:07:20
  them sold the farm to buy Tesla and they stuck with Tesla
• 1:07:24
  through thick and thin. Good days and bad days, yes.
• 1:07:29
  So my question to you is how do you feel about repeating the
• 1:07:34
  success, but in a safer way where you get qualified Tesla
• 1:07:39
  retail investor only to invest in SpaceX, avoiding all these
• 1:07:45
  market manipulator those crooks and the short sellers.
• 1:07:50
  It's it is a tough problem that you highlight.
• 1:07:54
  I mean, the basically the unfortunately, over time, the
• 1:07:58
  parasitic load of being a public company has just grown over
• 1:08:01
  time. And and so you get all these
• 1:08:05
  like, you know, spurious lawsuits, obviously, and they
• 1:08:12
  just make make it very difficult to operate effectively as a, as
• 1:08:15
  a public company. So, but I, I, I do want to try
• 1:08:20
  to figure out some way for Tesla shareholders to participate in
• 1:08:25
  SpaceX. That would be very cool.
• 1:08:58
  I've been giving a lot of thought to how to give people
• 1:09:01
  access to SpaceX stock because I do, I do want supporters to have
• 1:09:05
  SpaceX stock. But there's there's a sort of
• 1:09:09
  2500 shareholder threshold for, you know, before you become a de
• 1:09:13
  facto public company. But I don't know, maybe maybe at
• 1:09:20
  some point Tesla, maybe at some point SpaceX should become a
• 1:09:23
  public company despite all the the downsides of being public.
• 1:09:28
  All right, OK. First of all, thank you very
• 1:09:35
  much, Elon. Just a amazing job you're doing
• 1:09:38
  not only for Tesla shareholders, but the humanity in general.
• 1:09:42
  So we got a couple of questions. Many of us are asking like we
• 1:09:44
  want to know, many of us think you are one of the most
• 1:09:46
  consequential business people on earth ever.
• 1:09:50
  And we want to understand how do you think?
• 1:09:53
  Like, for example, like a couple of one, how did you know in the
• 1:09:56
  master plan Part 2, this is like 2018 or something like that?
• 1:10:00
  You calculated that point. It takes 6 billion miles of FSD
• 1:10:04
  miles driven before maybe the world, the regulatory would
• 1:10:08
  approve it. And then the second is like when
• 1:10:11
  you were deciding if Tesla's going to have, you know, the
• 1:10:14
  hardware chips, AI chips, did you know at that time that at
• 1:10:18
  some point when you get to a A5, it's going to be used not only
• 1:10:22
  for cars, but for bots and for AI data centers?
• 1:10:26
  Like was that luck a little bit or was it like you kind of knew
• 1:10:30
  that that was going to happen? Well, I generally, I mean in
• 1:10:34
  terms of the estimates like the, you know, for the self driving
• 1:10:37
  stuff, I, I generally try to get things to within an order of
• 1:10:40
  magnitude. So if so, I, I, it seemed to me
• 1:10:44
  like probably and, and, and this was technically in, in
• 1:10:48
  kilometers, but it, it, it will be more closer to 10 billion
• 1:10:53
  kilometers, which is roughly 6 billion miles than it would be
• 1:10:57
  to 1 billion. But I think I thought it would
• 1:10:59
  not take 100 billion kilometers. So it's really just been trying
• 1:11:04
  to, when, when you're trying to guess something where there's a
• 1:11:06
  lot of uncertainty, just try to get it, get the estimate to the
• 1:11:11
  nearest order of magnitude, you know, closest factor of 10.
• 1:11:15
  And, and that's that's why I said, you know, probably around
• 1:11:18
  10 billion kilometers or 6 billion miles.
• 1:11:22
  And then, yeah, the, the, the, the chip, the, the reason Tesla
• 1:11:30
  created a chip team. And it's important to know like
• 1:11:33
  the, the vast majority of like Tesla is like a dozen startups
• 1:11:36
  in one. And, you know, they're the only,
• 1:11:40
  we've only really done one major acquisition, which is Solar
• 1:11:43
  City. And then at some very small
• 1:11:46
  ones. So all of this is almost all of
• 1:11:48
  this is organic. So I bought the chip team from
• 1:11:52
  from scratch and the, you know, the AI team from scratch. the IT
• 1:11:58
  was just because it became a limiting factor.
• 1:12:01
  So for hardware too, we used NVIDIA, but NVIDIA was at at
• 1:12:09
  that time focused on making really AI server hardware, which
• 1:12:16
  obviously was a smart bet, though the most value currently
• 1:12:18
  the most valuable company in the world.
• 1:12:21
  And and you know, Jensen long and his team have done an
• 1:12:23
  incredible job at NVIDIA. My hats off to them.
• 1:12:28
  You know, I'm a huge admirer of, of, of Jensen and and NVIDIA.
• 1:12:31
  They've, they've done, done amazing work, but but they were,
• 1:12:34
  they didn't want to do a cut a low cost power efficient car
• 1:12:41
  computer at the time, AI car computer.
• 1:12:43
  So it's like, well, OK, I guess we, we need to start a chip team
• 1:12:46
  to solve that. So then that's when I hired Jim
• 1:12:50
  Keller and we, we both the chip team and did AI three, then AI4
• 1:12:59
  and then we to be honest, we made a few, we made some
• 1:13:02
  mistakes there with, with, with AI5.
• 1:13:04
  But now AI 5 is back on track and, and, and we'll have a very
• 1:13:09
  rapid cadence to AI6 and and so forth.
• 1:13:12
  And this will really, this is really necessary for, you know,
• 1:13:17
  for the, for the card to like, like with, with AI4, I think we
• 1:13:21
  can get to, you know, two or three 100% better than human
• 1:13:26
  safety, maybe 400% better. But with AI5, I think we can do
• 1:13:31
  1000% better or, or maybe even better than that.
• 1:13:36
  So than human safety. So at a certain point, actually,
• 1:13:43
  it might be too much intelligence for a car.
• 1:13:46
  Feel like I was thinking like, what if you get stuck in a car
• 1:13:48
  and you have too much intelligence?
• 1:13:50
  But then one of the things we could do is when the car is idle
• 1:13:54
  is use the car as a massive distributed AI inference fleet.
• 1:14:01
  So, you know, with the consent of customers, we're like, do you
• 1:14:04
  want your car to earn money for you while it's sitting in your
• 1:14:07
  garage at night? You know, I don't know, we'll
• 1:14:11
  pay, you know, $100 a month or $200 a month or whatever the
• 1:14:16
  right number is if you allow Tesla to do AI inference
• 1:14:22
  workloads when you're not using your car.
• 1:14:26
  So I bet it will also help the AI in the car not get bored.
• 1:14:32
  So because like I sort of imagined like what if I got
• 1:14:36
  stuck in a car, you know, and then well, and the highlight of
• 1:14:41
  your day was driving. It's like, you know, but they
• 1:14:44
  don't always want to drive. So then what do you do the rest
• 1:14:46
  of the time? So, so I, so I think Tesla could
• 1:14:51
  actually end up having the the largest, the Tesla might have
• 1:14:57
  end up having the most amount of AI inference compute in the
• 1:15:01
  world. Like if, if you think like maybe
• 1:15:05
  if we had 100 million car fleet and at some point we may have
• 1:15:09
  more than 100 million car fleet and they'll have AI6AI7, you
• 1:15:14
  know, and if you're able to run a kilowatt of inference on 100
• 1:15:20
  million car fleet, now you've got 100 gigawatts of distributed
• 1:15:24
  inference with built in cooling and, and power electronics and,
• 1:15:29
  and distributed power. So, you know, probably the
• 1:15:35
  markets valuing that is 0.0 right now is my guess.
• 1:15:38
  But but, but it seems like an obvious thing to do if you're if
• 1:15:42
  you've got distributed inference AI and you've got the power and
• 1:15:48
  the cooling, which is very difficult to do the power on the
• 1:15:51
  cooling and 100 gigawatts is a lot.
• 1:15:55
  I mean, the average, I said the average power consumption in the
• 1:15:59
  USI think is around 460 gigawatts for the that's the
• 1:16:04
  entire electoral consumption of the US.
• 1:16:06
  So the if you did 100 gigawatts, that would be a pretty big
• 1:16:10
  number. But yeah, it's basically
• 1:16:15
  something as a limiting factor and then we take actions to
• 1:16:20
  address the limiting factor. A quick follow up, Thank you for
• 1:16:25
  that very much a request for you.
• 1:16:27
  So you guys just unveiled the Cyber bear?
• 1:16:30
  Oh yeah. We'd like to.
• 1:16:31
  It's beautiful. We'd want you guys or you know,
• 1:16:34
  maybe do a cyber bull here in Giga, Texas.
• 1:16:38
  My name is Herbert. I've got a Brighter with Herbert
• 1:16:40
  channel on YouTube. This is the cyber bulls.
• 1:16:43
  We are representing the Tesla Bulls and we stand with you,
• 1:16:46
  Elon. But wouldn't it be cool to have
• 1:16:47
  a cyber Bull right here in Texas?
• 1:16:50
  Like a a Cyber Longhorn. Cyber Longhorn.
• 1:16:53
  Alright, we'll we'll, we'll, we'll do a Cyber Longhorn for
• 1:16:56
  the factory, alright. Good afternoon.
• 1:17:04
  I'm very excited to ask this question.
• 1:17:07
  I've dreamed of giving away a Tesla for a very, very long time
• 1:17:11
  and I finally wore EV Jack down enough that they're willing to
• 1:17:15
  foot the bill for it. But it turns out to give away a
• 1:17:19
  Tesla, I have to have your permission to say we're giving
• 1:17:22
  away a Tesla. So.
• 1:17:25
  It's just simple, just give it away.
• 1:17:28
  You don't have to do anything. Tesla doesn't energy.
• 1:17:30
  We're going to go through the normal channels.
• 1:17:31
  We'll buy it from a store, sure. All that stuff.
• 1:17:34
  You can give away Tesla. It's totally cool.
• 1:17:37
  Yeah. Yeah.
• 1:17:39
  But certainly you don't need my permission to give away a car.
• 1:17:44
  We'll take, like, maybe a couple more questions and then call it
• 1:17:47
  a night. So.
• 1:17:48
  All right. Hi, Elon.
• 1:17:51
  My name is Jonathan. You mentioned that the Roadster
• 1:17:53
  will have more tech than all the James Bond vehicles combined.
• 1:17:57
  Do you think there's a possibility that any of that
• 1:17:59
  tech will make it into the current vehicle lineup?
• 1:18:04
  No. And to follow up on that, do you
• 1:18:10
  have any estimate of production or delivery timelines for the
• 1:18:14
  Roadster? I guess so we're aiming for the,
• 1:18:21
  so the the product unveiled will be of of of the Rosa 2, which
• 1:18:28
  will be very different from what we've shown previously that that
• 1:18:32
  demo event will be April 1st of next year.
• 1:18:36
  I have some deniability because like I could say I was just
• 1:18:40
  kidding, but we are actually tentatively aiming for April 1st
• 1:18:48
  for this, what I think will be the most exciting, whether it
• 1:18:53
  works or not demo ever of any product.
• 1:18:57
  And then I guess production is probably about a, you know, 12
• 1:19:03
  to 18 months after that, the I think production is probably a
• 1:19:08
  year, a year or so after that. Oh well, I can't give away the
• 1:19:14
  secrets, but you won't be disappointed.
• 1:19:19
  All right, take one last question, I guess.
• 1:19:21
  All right. Three glass questions, OK.
• 1:19:32
  Hi, Elon. I'm really excited about this
• 1:19:34
  future of sustainable abundance that we're talking about.
• 1:19:38
  I mean, you're going to be saving a lot of lives with FSD,
• 1:19:40
  but the number of lives that would be saved and improved with
• 1:19:43
  this future vision you have is really inspiring and very
• 1:19:46
  exciting. So even today.
• 1:19:50
  You've mentioned though that in a post scarcity world, the role
• 1:19:53
  of money could diminish or become obsolete.
• 1:19:56
  Given that much of today's power, including yours, is tied
• 1:19:59
  to wealth, do you think achieving this abundance would
• 1:20:03
  require powerful people to relinquish their power?
• 1:20:06
  And how might we address resistance from those who hold
• 1:20:09
  power to make this vision a reality?
• 1:20:12
  Well, I mean, I think actually long term the AI is going to be
• 1:20:16
  in charge, to be totally frank, not humans.
• 1:20:20
  If, if artificial intelligence vastly exceeds the sum of human
• 1:20:24
  intelligence, it is difficult to mention that that any humans
• 1:20:27
  will actually be in charge. So we just need to make sure
• 1:20:32
  that AI is friendly. Yep, thank.
• 1:20:35
  You. Is that was that the question?
• 1:20:38
  Go ahead. Sure.
• 1:20:40
  Yeah, OK. Sorry, Elon.
• 1:20:42
  OK, OK. Yeah, so I'm Cybercat Onx and
• 1:20:47
  I'm doing YouTube and content creator.
• 1:20:50
  There's one thing I always heard about Tesla owners or people who
• 1:20:56
  want to buy Tesla complain about that is about super expensive
• 1:21:01
  insurance, right? And The thing is Tesla, I know
• 1:21:07
  Tesla also do insurance. However it's not cover cover
• 1:21:10
  every place like like based on Boston, I don't have too much
• 1:21:14
  options. It's super expensive.
• 1:21:16
  And the other point is the FSD is very safe right now.
• 1:21:20
  I've been using FSE for about four years, right?
• 1:21:24
  It's getting to a point it's like almost unsupervised.
• 1:21:28
  However, the insurance still does not take this into
• 1:21:31
  consideration. They don't ask you whether you
• 1:21:34
  have FSD or not and they don't know how much you travel with
• 1:21:38
  FSD. And that is not a part of the
• 1:21:41
  risk prediction kind of thing, right?
• 1:21:44
  So I feel like what's your thought about insurance going
• 1:21:49
  forward, especially when we're getting close to the autonomy?
• 1:21:53
  What is past like either using your own or with the external
• 1:21:59
  partnership? For example, there is a company
• 1:22:01
  called Lemnade. We do need to not have the
• 1:22:03
  questions be super long. I mean, Tesla insurance is
• 1:22:11
  trying to expand as, as quickly as possible, but the regulatory
• 1:22:14
  structure for insurance is extremely complex and and works
• 1:22:18
  on a state by state basis. So it's really somewhat of a
• 1:22:22
  racket And the the rules for insurance are different with
• 1:22:29
  with every state. It's it's a very, very
• 1:22:31
  complicated thing. So you know the, I mean, yeah,
• 1:22:37
  so there's, but I'm aware that insurance often is too expensive
• 1:22:42
  and doesn't take the right things into account.
• 1:22:46
  But so all I can say that is, yeah, we'll keep expanding Tesla
• 1:22:49
  insurance. When the car is operating as a
• 1:22:51
  cyber cab, Tesla will simply self insure.
• 1:22:54
  So that kind of solves that. But insurance is, is a, yeah, a
• 1:23:01
  real pain in the neck for sure. But OK, I do need to end this at
• 1:23:08
  some point. I'll take.
• 1:23:11
  Yeah. One question, one question
• 1:23:13
  there. All right.
• 1:23:14
  Thanks Elon for taking my question.
• 1:23:16
  I appreciate it. My question has to do with
• 1:23:17
  compute and what the build out or what, how much is necessary
• 1:23:21
  to train optimists and actually get them to a very household
• 1:23:26
  meaningful robot that can do things and if the partnership
• 1:23:32
  with XA with XAI would help accelerate that.
• 1:23:39
  Yeah, there's there is a lot of training compute needed for
• 1:23:42
  Optimus. And because the the the the AI
• 1:23:48
  chip in the robot is relatively weak because it's it's really
• 1:23:53
  limited on limited on power. You can make up with that with a
• 1:23:56
  lot of training to have a lot of training result in in a a very
• 1:24:02
  efficient AI that can run on a low power chip in the in the
• 1:24:07
  robot. So it it is we, we, we will
• 1:24:11
  actually have to spend a lot of money on on training.
• 1:24:13
  Like, you know, ultimately it'll be like 10s of billions of
• 1:24:17
  dollars on on training compute. So it's a big number would.
• 1:24:22
  A partnership with XAI help accelerate that.
• 1:24:25
  Yeah. I think potentially, yeah, I
• 1:24:27
  think that that that could. Yeah, there's potential for
• 1:24:32
  accelerating that. So yeah.
• 1:24:37
  Did the XAI investment thing get approved?
• 1:24:41
  No, yes. OK, OK, OK, OK, whatever, you
• 1:24:50
  know, whatever. Like it's, it's like, you know,
• 1:24:53
  Tesla and some other company that I have an interest in, then
• 1:24:55
  it's like always quite complicated to, you know, do
• 1:24:59
  things has to go through a lot of hoops to to happen.
• 1:25:04
  But you know, I do think there's a lot of potential for
• 1:25:07
  collaboration with XAI in the future and with SpaceX.
• 1:25:10
  So all right. OK, this is the last one.
• 1:25:14
  Yeah, sorry. Thank you so much.
• 1:25:18
  Captain Ali on XI support you. Thank you SO.
• 1:25:20
  Much for everything you do. Very simple question.
• 1:25:23
  I'm from Israel. I don't represent a lot of
• 1:25:26
  people, but people do ask me and I'm going to ask you.
• 1:25:30
  Any chance to have the app in other languages, like Hebrew for
• 1:25:34
  example, some people struggle with?
• 1:25:37
  Me even the app is the app. Itself.
• 1:25:39
  Yeah, just the app the. App is not in Hebrew, no.
• 1:25:43
  And a lot of people don't speak English, so it's.
• 1:25:44
  Oh, OK. Shoot, I thought I thought we
• 1:25:47
  had it in all languages. OK, well, definitely the app
• 1:25:51
  needs to be in all languages. All right, so all right.
• 1:26:28
  Hey, thank you so much for listening today.
• 1:26:30
  I really do appreciate your support.
• 1:26:32
  If you could take a second and hit the subscribe or the follow
• 1:26:35
  button on whatever podcast platform that you're listening
• 1:26:37
  on right now, I greatly appreciate it.
• 1:26:39
  It helps out the show tremendously and you'll never
• 1:26:42
  miss an episode. And each episode is about 10
• 1:26:45
  minutes or less to get you caught up quickly.
• 1:26:48
  And please, if you want to support the show even more, go
• 1:26:51
  to patreon.com/stagezero and please take care of yourselves
• 1:26:57
  and each other and I'll see you tomorrow.