https://www.youtube.com/watch?v=-uS0TG-sKi8
0:07 hi Steve welcome today I'm joined by Stephen Mark Ryan of solving the money Problem YouTube channel
0:14
um I've interviewed him in the past and it's been great getting to know him over the years I'll go ahead and link to the
0:19
video or to the interview in the video description below so in today's video I want to dive into Tesla's AI Day 2 into
0:26
Tesla's Optimus robot the future of AI Tesla stock and a lot of hot topics on
0:33
welcome you on the channel Stephen how have you been fantastic Dave great to be back and look forward to our chat yeah yeah it's a
0:40
great uh connecting so I'm curious um what was your reaction I've watched a couple of your videos on post um Artemis
0:48
um you obviously caught the bigger picture so yes help us help us see
0:54
what's going on in your mind what was kind of your reaction and yeah what's your kind of takeaway from Tesla AI day
1:01
two well I think the biggest takeaway for me is that they're going to do it
1:06
the rate of progress the seriousness the sense of extreme urgency in terms of executing on this project and their lead
1:12
I think one thing that a lot of people have just completely overlooked they're getting fixated on the body not the
1:19
brain they have they don't understand that what Tesla has taken they've effectively transplanted the brain
1:26
from their autonomous software their computer vision into the Optimus humanoid robot and they are light years
1:33
I mean half a decade or more ahead of where anyone else in the world could possibly be in terms of a brain for a
1:38
humanoid robot the body the movement you know the actuators they'll figure all that stuff out that's the easy stuff but
1:44
it's the brain that matters the most Vision perception you know planning and action in the real world and this is
1:50
something I think a lot of people are overlooking and I don't see anyone getting this thing that Tesla has
1:55
developed the brain they've got the foundational building block of AGI now in place they are very serious about executing this they have a very viable
2:02
plan to actually scale up production start with robots in their own factories they'll be almost useless in fact they'll be worse than useless initially
2:08
there'll be a drag on productivity they'll cost but over time Tesla can increase the capabilities once they're
2:14
effective and adding valuable labor to their own factories they immediately will have customers with similar needs
2:19
that they can start selling these spots to then eventually their Fleet scales much like the vehicles this thing it's
2:24
got a feedback loop gets better and better over time and Ultra long term AGI is probably as Elon mentioned an
2:31
emerging property of what they're doing and I just don't see how anyone or any
2:36
company catches up to Tesla from here now I know this sounds like a diluted Tesla Fanboy I get this a lot but I'm
2:42
looking into the future and thinking how does anyone catch Tesla how they will need to deploy Millions today that would
2:48
need to deploy millions and millions of robots whether vehicles and or humanoid in the real world to start that data
2:54
flywheel and the training effect and I don't see it happening um and I think actually you mentioned
2:59
quite recently on a video I think I was seeing a chat with Emmett peppers you can get everything that you need to
3:05
know the sum of human knowledge on the internet that's great but you also need that Real World experience I don't see
3:11
where anyone captures that real world world stuff and at this point in time I think a lot of people are getting caught
3:16
up with the window dressing and not really understanding what's happening here but I am so excited for the future
3:21
of Tesla I just I can't sleep last night I was laying awake for hours just thinking of all these possibilities
3:28
um yeah and I think it's going to take a very long time in many years maybe a decade or more before people really understand what's happening with this
3:35
thing yeah but yeah it seems like Tesla is a very clear path to AGI yeah yeah I think um yeah there's so much stuff
3:42
going on I feel like people are missing the the storyline you know of the decade
3:47
um that will impact the next you know Century I think if people would see the track record of Tesla the track record
3:53
of Elon seeing the industries he's disrupted and trust and just believe what Tesla's says that they're doing
4:00
right now you know I think at Aid that the problem was they didn't show
4:05
something that was perhaps better than what's already out there right and people are like what the heck right but
4:11
what they did show was what they're working on with the actuators of the design and all this you know the AI brains all this stuff which is uh
4:18
especially Best in Class like no one's trying to do what Tesla's doing at the scale that they're doing it and I think
4:24
that's where if you just for a minute kind of pause and hold back your cynicism your doubts and say okay if
4:32
Tesla really is building the brain and they really are building all of you know
4:38
the body and all the design for Mass manufacturing and you put these two together what's going to happen and you
4:45
can just map it out for the next five 10 years 20 years 30 years and the more you
4:50
map it out the crazier it gets you know it's like I don't even want to talk about past like 15 years because
4:58
it just you'll get lost off the internet yeah yeah exactly um
5:04
yeah so I'm curious I want to dissect a little bit um of the assumptions you're making because
5:11
there's obviously certain things you're seeing and you're assuming um that are connecting the thoughts for
5:18
you that Tesla's going to you think be successful with this bot is going to release the brains that are going to you
5:25
know basically surprise everybody but what do you think are the key assumptions you're making or the key
5:32
kind of you know points that you're believing that perhaps other people aren't seeing clearly mm-hmm I think the
5:38
biggest one as I mentioned in the intro there is that Tesla has built the foundation of AGI with Tesla vision the
5:45
the most important thing to solve for on the path to AGI is vision perception planning and action in the
5:52
real world and Tesla hasn't completely nailed that yet but they've built the seed and it seems
5:57
to me inevitable that if you run the clock forward far enough what what Tesla has today if it
6:03
continues to improve and evolve and get better and more capable and more intelligent it just continues to go from
6:09
you know basically useless to Tesla Q to below average intelligence to eventually human level and then eventually
6:16
superhuman and there doesn't seem to be any end in sight and in addition to Tesla having built the foundation
6:23
it's getting close to impossible for me to imagine a realistic scenario in which anyone can catch up to Tesla they just
6:29
like I said they'll need their own Fleet they'll need that flywheel happening so my argument is at some point AGI happens
6:35
at some point a humanoid robot that becomes more capable than a human in terms of intelligence will happen it's just a matter of when and who does it
6:41
and I don't know on the timing we're probably a couple of decades away from that point but irrespective of exactly
6:47
when it happens it seems quite clear that Tesla has an unassailable lead to
6:52
make this happen and I know it's super early to be calling that but something remarkable would need to take place for
6:57
me to to change my position on that and I'm open to changing my mind it's also I
7:02
think a point now where Elon has realized the value of AGI a switch went at some point he goes oh
7:08
wait a minute wait what we've got here with Tesla vision and he's had his AI concerns for years
7:14
AI in the long hands which I lose a lot of sleep over as well I think elon's got
7:19
to a point now where he feels an ethical and a moral obligation to make this happen under Tesla rather than you know
7:26
having Facebook attempting to do the same thing and now they're really aggressively recruiting Talent taking a
7:31
big risk as a company to show off very early technology prototypes and plans to recruit the engineers and I think
7:38
with each AI date Elon mentioned the possibility of doing a podcast with monthly updates it seems that Elon and Tesla are
7:45
absolutely hell-bent on making this happen and recruiting the talent that needs to to be there to actually execute
7:51
on this there aren't any other companies that are taking this so seriously nor that have that fundamental building block
7:57
um and I think it's got I think it's going to take a very long time for people to put the pieces of the puzzle together because until they start seeing
8:03
proof here's I'm sort of looking in the future and I can see what's inevitable but I think it's going to take many
8:08
years before other people start seeing enough evidence to to draw the same conclusions um so I think there's a huge
8:14
disconnect here and I mean EVS FSD energy nothing it's nothing to get it I
8:21
mean it's literally nothing it is a spec compared to the potential of the humanoid robot yeah the thesis
8:27
essentially is if you can scale up the economy indefinitely with useful labor things
8:32
get super ridiculous yeah yeah I want to dive into that a bit more um but before that it's it's an interesting thing to
8:39
think about how Tesla with trying to solve FSD and FSC taking so long and being basically a harder problem than
8:45
the anticipated it led them to this path where they had to develop this crazy
8:51
world-class AI team and infrastructure and compute and basically they had to
8:57
just like you know chew off a lot more than they had anticipated but by doing that it led it leads Elon down this path
9:04
where it's like wait a minute the next step is just like a humanoid and then
9:10
you've got potential ATI and it's just exactly what he had
9:16
um hoped for in in a sense in the past with open AI where he wanted to Steward kind of you know the direction of Asia
9:23
in the right way but now he gets that opportunity in Tesla and it's fascinating how this stuff works out you
9:29
know you you couldn't have predicted this probably like you know yeah six or seven years ago it was a bit hard
9:35
I absolutely agree I think at some point like I said a light's going off in elon's head he's like oh okay this is
9:41
perfect let's go yeah um so why do you think Tesla would be in
9:47
the lead for potentially solving AGI compared to let's say Google like what's your take
9:53
on you know that Tesla versus Google race or open AI
10:00
the real world application so as I mentioned you can crawl the internet there's a lot of knowledge available
10:06
there that can train on images and recognition and categorization and blah blah blah blah and rules based but when
10:12
it comes to interacting in the real world I don't see anyone like Google Amazon pick your large tech company that has
10:18
anything at the scale that needs to be out there to actually collect enough data to have a meaningful impact so if
10:24
Google wanted to try and attempt to catch up to Tesla they'd probably need to build out their own Fleet of autonomous vehicles and I mean millions
10:30
of them to start getting some data in that form or if they really wanted to they could just start with a humanoid robot but I think they'd be at a
10:37
disadvantage because there's a lot of valuable useful information and and they could really develop great computer
10:42
vision just through vehicles Tesla's so far ahead so many years ahead already and it's just a feedback loop
10:49
the more time they have the more data the more efficient their neural networks get the better and more capable I find
10:54
it very difficult for to imagine a company like Google or another big tech company having a legitimate chance of running Tesla down and let's go you know
11:00
what if it put 100 million billion dollars into this project ASAP immediately let's just do whatever it
11:06
takes to start trying to run down Tesla and I think the amount of Courage that would be required to do that as a
11:11
company you imagine how Google investors would feel if they said you were going to deploy a hundred billion dollars just this year to start our own flywheel to
11:18
try and catch Tesla on this crazy humanoid robot that most people don't think will happen I think Tesla the agility of the company
11:25
the DNA of the company almost allows them to take these kind of risks without anywhere near the same penalty as
11:31
another company would have they still have that startup culture and I think most of the big techs are
11:36
behemoths now they're not really at that point where they're willing to to bet the company on something or take these huge risks and make the Investments you
11:43
know yeah you always got some great AI stuff but I just they've got tiny little elements they've got great language and and search and and stuff like that but
11:50
I just don't see any other big tech companies that have anything comparable to what Tesla's built with their Vision system what are your thoughts on that do
11:56
you see yeah how do you say it yeah I mean I was having lunch um uh AI Dave with some uh subscribers followers and
12:04
um someone was commenting that Apple would never do what Tesla's doing with public FSD beta like it would be just
12:11
okay to enjoy their image everything right yeah Kim cook would be vomiting
12:17
violently just thinking about an idea like that exactly and and that like kind of you know exemplifies the difference
12:24
between Tesla and you know an apple where elon's like hey I tried the latest
12:29
you know craziest beta version that might kill me and he's due to himself and it's such a such a contrast to to
12:37
running a huge you know typical let's say multi 100 billion dollar company
12:42
um yeah I guess the question is you know with Google open AI others will
12:50
there be an AGI like on the computer that's different than an AGI in real
12:55
life or do you think they'll just intersect in and just be one do you think that eventually Tesla
13:03
can compete and overtake you know the online AI experts in those companies and
13:10
what they're doing by somehow their real world AI develops so fast that it just
13:16
engulfs you know the online you know presence too or you know how do you kind
13:21
of foresee this playing out you know next five or ten years I definitely think there will be a
13:27
convergence of these capabilities I don't know exactly how it plays out but I do have a general hypothesis that at
13:33
some point what Tesla or anyone else is building in terms of intelligence becomes better at
13:39
Learning it's not just learning but it becomes better and more efficient at learning more generalized a lot more streamlined
13:45
like human beings are a great example of you know millions of years of biological evolution our neural networks are
13:51
relatively efficient you know a baby only needs to take the same action so many times to realize you know and they
13:57
can move through the world in a certain way don't do this don't do that and humans are the same we can watch somebody perform an action and then a
14:02
few attempts you know when the this thing at the AI day that little love I tried to do the the hands right I've never tried to do that before and I'm
14:08
like you know like what what it took me three or four attempts to get the rough shape and then about five ten attempts before
14:14
I could just slam it out so we have relatively efficient neural Nets in terms of learning I see it getting to a
14:19
point with um you know the AI that Tesla's developing that it'll be able to to pick up and learn things so much more
14:25
efficiently and quickly that we're just we're we're very early days now training isn't as efficient I
14:30
think as it will be ultimately at that point in time I think whatever the gaps are in your capabilities will be very quick to we have a very quick to sort of
14:37
fill those in yeah so I think that this stuff will converge I think it's definitely a possibility that maybe there's like an online AGI that's super
14:45
capable and maybe Tesla's humanoid robots much more capable in the real world but I see those abilities
14:50
converging over time and you know it seems to me like you can effectively take your your networks and
14:55
kind of layer them in it's just like the brain that's evolved over time we have our basic brain and then things have got
15:01
more complicated and sort of lay it on and lay it on later on I think that's probably how AGI will go as well start
15:07
with you know some very basic capabilities and stack things on and on and on so I don't know yeah yeah I think
15:12
um I think uh Eid the Q a Elon says something about I think having the data
15:18
in the compute um you know it's like they're they're in
15:23
the lead at least with real world AI but when you think about it there's potential that the amount of data you
15:29
accumulate in the real world um it's just could be so massive compared to if you're just isolated in
15:37
an online environment because like for example with open ai's gpt3 like sure they you know scrape all of reddit's you
15:44
know whatever Pages all that but that's something Tesla could do too you know it's like it's not mutual exclusive per
15:50
se but to be like to have let's say potentially hundreds of millions of
15:56
robots interacting with daily tasks and communicating and doing the actual labor
16:03
that humans do and actually almost acting like a human too right like that
16:09
amount of data I don't think you can compare with an online only type of data
16:15
set I agree it's just like I feel like it's like online is almost like a two-dimensional world yeah versus the
16:21
three-dimensional world you know there's so much that is just missing from the online World it seems like the online
16:27
AIS seem to be pretty good at mastering certain tasks with just brute for course a lot of information it seems fairly
16:33
simple you can just brute force your way and it's very quick for AI there's a little
16:39
AI somebody sent me recently where it can transcribe a video and then try and figure out what the message and the meaning of the video gives you a tiny
16:44
little summary and it writes this video is about blah blah blah and I ran it on a few of my videos I'm like holy crap this is incredible in a matter of like
16:51
two or three seconds it analyzes an entire video spits out a response but that's very basic compared to
16:57
traversing the real world which is complicated and chaotic and uncertain and there's infinite variety there you
17:04
know there's only so many correct answers to what's one plus one it's two right whereas it get into the real world
17:09
and things just take on an entirely new dimension yeah and yeah I think I think you're right on that point yeah I mean
17:16
it's also you learn things in the real physical world in a different more in different way than you do just online
17:23
like let's say someone says something to you and they give you a weird face and there's there's more context and nuance
17:29
and more information to take in than just you know you say something funny online and you get a a Emoji reaction or
17:37
something you know that amount of data is so limited compared to what you intake in the physical world I think
17:43
there's a there's a um you could make a hypothesis or
17:49
argument that you need to have real world AI to get to AGI like the amount of data you need you have to get that in
17:56
the physical world because that's where the bulk of the data will lie and that if you just put yourself in the computer
18:02
like you know a lot of these other companies are you're going to hit a local maximum you're not going to have
18:07
the data that a real world AI company like Tesla has if that's the case then
18:13
it's like oh my gosh it's like Tesla's AI feature is going to be insane
18:18
um yeah yeah I tend to lean toward that thesis too you know yeah so I'm so
18:24
excited yeah um so economic potential of the humanoid robot so
18:31
um where where does your brain go in terms of how big this could be well honestly the answer to that
18:37
question is when we start butting up against the laws of physics and the amount of available matter and energy in our solar system I know that sounds
18:43
ridiculous but that's where my brain goes first and okay there's that's where the real physical limitation actually is
18:49
we've got to take a step back from there and is there an arbitrary limitation I don't know I generally don't have an
18:56
answer to that my suspicion given the history what's happened with world GDP and Technology sort of building on the
19:02
back of previous Innovations it seems probable that there isn't a a reasonable limitation to the size of the
19:08
economy if we can create useful labor we can create useful things I don't see an arbitrary limitation and
19:16
that's a pretty mind-blowing and or disturbing sort of reality to contemplate because if you
19:23
think about the potential what if we could increase the size of the global economy by 10x or 100x or a thousand X
19:28
or 10 000 X or more which I do believe is it's possible and feasible just a matter
19:34
of when it just gets absurd I mean it truly gets just absolutely absurd
19:39
um yeah mind-blowing yeah do you have any thoughts to a limitation in terms of
19:45
like what could actually happen because I'm done yeah yeah I mean so on a tangent here I think
19:52
um with AI and robots there's this bias people have as humans and it's I think
19:58
it's a natural kind of almost self-preservation bias where it's hard to imagine robots being smarter than you
20:05
you know you think you are the top of the you know structural Universe
20:10
um species that you are the smartest and to think that you know a robot or AI or computer can
20:16
be smarter than you I think that's like so unnatural it's like almost death you know to the person and
20:24
it's easier for people to think that robots can be stronger physically than them it it's like you could say okay you
20:31
look at their metal body and you're like oh my gosh you know yeah but to be smarter that's where I think
20:37
um most of the I would say pessimism but the um not tapping into what I think the
20:43
potential of robots will be an AI if if you can't see that they'll be smarter
20:49
than you you're not going to see the potential of the future you know but if you kind of lay that down and say okay
20:54
what if these neural Nets advance so much that yeah robots are smarter AI is
21:00
smarter than me and most almost everything I still think there's gonna be certain things that humans do that
21:06
are so almost I would say I would say mystical or Supernatural but are Beyond
21:13
perhaps reason or rationale that will be uniquely human I still think that that's going happen but in terms of like sheer
21:21
intelligence and capability yeah you know it's very possible that
21:26
these neural Nets once they Eclipse human you know intelligence then it just exponentially goes up where you can't
21:32
even compare you know their intelligence to human intelligence and then you get into the scenario where they can make
21:39
more robots themselves they can create their own factors they can design things that we can't even understand what
21:44
they're designing um that's right and yeah there is this basically self-reproducing you know
21:50
species almost and that's where you're like oh yeah there is no end to that
21:55
potentially um and that's where when we talk about the potential of the economy I'm like
22:00
yeah what's the potential of a self-reproducing robot that designs the
22:05
designs are getting better and better than you you don't even understand the designs as humans yeah right yeah I mean
22:12
of course that that poses some potential dangers and all that stuff but in terms of pure you know potential economically
22:19
we're talking about something that has never happened in the history of the world you can even I think compare it
22:25
almost you know to anything that's happened it's just something so crazy um I was thinking about this um maybe
22:31
you'll have some some thoughts on this but I was thinking um this past week humans have this
22:37
limitation of reproduction where it's extremely difficult not just to reproduce but to care for your Offspring
22:44
there's so much time and energy there's like health and safety and all these
22:49
issues um that you can't just reproduce a thousand offsprings you know then have
22:55
them do a thousand each right like that's just not how humans work um yeah but that limitation doesn't
23:01
apply to robots you know because you don't have to clean up for them you don't have to you know breastfeed them
23:08
you have to do anything they're just there and as the brains get exponentially smarter and better and
23:13
whatever and as they are able to you know create their own robots then you've got this crazy Limitless reproducing
23:21
situation it's just yeah mind-boggling yep I I I'm not sure if you've mentioned
23:27
this but I certainly have have come to the conclusion that the Tesla humanoid robot is the final product
23:33
yeah because as you mentioned being able to create better versions of yourself not needing to be trained not being an
23:39
infant just out of the box capable yeah just get the implications are
23:45
really ridiculous yeah yeah I call it the the product that ends all products it's like the end yeah it is the end it
23:52
really is um what's interesting also is like Elon is chooses not to produce like a dog or
23:59
some other types of robots he's going for the human robot like humanoid robot right off the bat
24:06
um and I think part of it is his Focus where he's like you know we have a shot
24:13
at this we're gonna go for it he understands this mostly about the brains um another part of it is I think he
24:19
understands once they put it out and they can get a lead and get millions of things these things out how can you like
24:26
who can catch up you know it's like whoever's first and why start with the dog when the competition is going to
24:33
come out through a humanoid robot you know it's like let's play this like go with the real thing get it out to millions and then it's like I would say
24:40
it's game over but it's like yeah your lead grow and grow it seems like yeah I think everyone said something in
24:47
the AI Day 2 presentation about the fastest possible path to a useful
24:53
robot whatever that is so he's had that thought why start with anything other than the final product let's just get
24:59
something that's useful as fast as possible then we can iterate and improve and yeah the sense of extreme urgency
25:05
that I really got from AI day 2. it seems like Elon is like okay this is the
25:11
absolute priority we've got to get this done as fast as possible and he senses this first mover Advantage there's a
25:17
point where it's unassailable if you get enough of an advantage you just don't get caught
25:22
um you know it just it's just like a slug trying to out compete a human you know once you get to a certain level of capabilities
25:28
why even bother yeah yeah I totally got that AI there too is Elon didn't see it explicitly but it was
25:36
implied that the robot the humanoid robot is is way way on a different level
25:42
than anything else you know like obviously all that stuff is just preparation yeah you know Stepping Stones to get to this um to the robot
25:49
and it's it's interesting how how clear Elon is thinking on this you know
25:55
um and to me it's it's it's it's obvious it's clear but it's just most CEOs wouldn't I think think so clearly on it
26:02
and make the next logical step which is to bet the whole company or you know your everything on it and it just seems
26:09
like he sees it clearly and he does what logically he needs to do um it makes so much sense right but it's
26:16
like so abnormal it seems like in you know the business or or Tech World um
26:21
yeah um how about timeline what are your expectations on you know when Tesla can
26:27
start using this in the factory in a useful way but not just that what are the Stepping Stones like what
26:34
are the next steps do you think that it'll just be factories and then you know selling to businesses or do you
26:39
think they'll use it for like last foot delivery train the you know the robots to do that before getting to the home or
26:46
yeah I mean do you think three to five years like Elon said is a realistic timeline to get in the home like what
26:52
are your thoughts on timeline well in terms of timeline I do actually my test evaluation model has the first
26:58
Tesla Bots being sold to customers hypothetically in the best case scenario in three years and I don't expect any
27:05
meaningful volume for about five years and this was a fairly bullish estimate so the timelines match fairly closely I
27:11
was also expecting the hardware about 25 000 so Elon and I are kind of guessing
27:17
around the same timing but of course there's so many unknown unknowns it's very difficult to know for sure when
27:22
this will be commercially viable I think the strategy that Tesla will deploy obviously their own factories is the first place to to nail useful productive
27:29
labor but I think if Elon and the team at Tesla can identify any other areas where they can become useful quickly
27:36
maybe lifestyle delivery is actually not as hard as you might think they may actually go after these opportunities as
27:41
well I think Tesla's probably going to want once they've got useful humanoid robots in their own factories they really will be desperate to get them
27:48
into other applications not just other factories but where's the next thing because once you're in a factory doing
27:53
useful labor you can get better and better and better you don't need to be you can now start focusing on what else can we add new capabilities that this
28:00
thing currently can't do at all and and be very aggressive in those so I'm not sure exactly in terms of I'd guess
28:06
probably late this decade there'll be some sales to other commercial customers initially and I think it's really the
28:11
2030s and Beyond where we start to see the Tesla Bots being used in a much broader range of applications and
28:17
becoming increasingly more capable finding its way into homes probably 2040s is where things become super
28:24
insane where it has a bot can be useful in at home caring elderly everywhere what do you mean so if if I can get a
28:31
Tesla bite in my house in let's say five years what is it going to be able to do then do you think not a lot I think it'll be able to do
28:38
some useful stuff yeah I think don't get me wrong here like it'll be useful right but I don't expect that in terms of
28:45
being super it's kind of gonna be like having your kids help out around the home right they can be useful but they
28:51
can also not be super helpful I think if Tesla can get a few basic tasks down that'll be great but my expectation
28:57
maybe I'm a little bit bearish on this I don't think that there's going to be major utility with a Tesla body in the
29:03
home probably until late this decade maybe around the turn of the decade in the best case scenario that's not to say
29:09
that it may not be useful for some people if you just want a robot that can take out your trash you might be able to have that a little
29:15
bit sooner but the the number of different tasks that you'd want a robot in the home to be capable of doing
29:21
before you really think of it useful from a consumer point of view there's a broad range of things you know doing your laundry loading unloading
29:28
folding clothes these kind of things I think it's going to be a few years before Tesla's really useful in that regard because there's a
29:35
there's a point where if you have a robot and it's not able to fold your clothes properly it's making a mess and
29:40
it's unloading the dishwasher but it's dropping Forks all over the place I think there's going to be a few years where with Tesla's kind of at that level
29:46
especially with the manipulation people don't I think appreciate how incredible a human hand is the sensitivity the way
29:53
that we can pick up an object just look know exactly how how firmly to grasp and stuff that stuff that Tesla doesn't
29:59
really have any progress yet like it's really early days they're just figuring out the sort of
30:05
you know very basic movements I think it is going to take a while to iron that stuff out but the more Bots that are
30:11
deployed the more the flywheel continues I think the capabilities will grow very fast and surprise people I think it'll
30:17
disappoint people in the first few years but then out of nowhere much like we've seen breakthroughs with AI you know
30:22
image categorization and then with chesco these kind of things the capabilities are decent and then out of
30:27
nowhere you just have this this vertical moment where it just becomes almost superhuman what are your thoughts on timing I'd love to know yeah you know it
30:34
I actually I resonate with some of that um because some of the the in-home tasks
30:42
are super hard like cooking you know chopping stuff and like yeah just and to
30:48
be useful to be really useful in the house you need to do like that type of stuff you know you can't just vacuum you
30:54
can have a Roomba for that or you know like um so the high value stuff is super hard
30:59
my kind of question question mark and I was looking into some of the the new AI
31:07
occupancy networks and they call them call them Nerfs the neural radiant Fields but they're basically rendering
31:16
3D spaces super efficiently now and I wonder if in the next three or four
31:23
years there will be huge advances in this field of basically
31:30
3D you know modeling rendering rendering you know control all that stuff inside that environment to the point where what
31:38
we have now is just like Kids Place like almost nothing you know and that type of
31:43
breakthrough will enable the robot to be pretty much superhuman in terms of its
31:50
ability to move around um in an environment like right now it's hard to hard to imagine that a robot
31:56
superhuman and disability move around because we just don't have something to compare it with right um but if you track the progress and the
32:04
speed of how fast AI is improving and how this field is just like breaking
32:09
through with a lot of things with especially with real world AI it's just there's gonna be a time I don't know
32:15
when that is whether it's in four three or four years or five or six years it doesn't seem like it would take longer
32:21
than Phi though um but I think that could be one of the breakthroughs there might be another
32:26
breakthrough that's needed to actually you know function in a superhuman dexterous way to be super you know
32:34
helpful but it seems like there's there's a we're not too far away
32:39
um but part of my brain you know it's still like comparing it to now so I'm like wait a minute that
32:45
doesn't seem realistic um but I think this is going to be a Trippy next you know five or ten years
32:51
like every year we're talking like what's going on here yeah well I'm having these moments now some of the
32:57
progress with AI like Dali and Dali too just yeah my brain breaks I'm like we've gone from AI basically some cool
33:04
language stuff you know next minute here's a prompt and you create some unique and novel art and I didn't expect
33:10
that you'd be able to type a few text prompts in in 2022 and have ai render a
33:16
bunch of scenes and scenarios I just didn't expect that to happen so you're absolutely right in terms of
33:21
breakthroughs who knows exactly on the timing but there is that point with capabilities where out of nowhere it
33:26
just goes from decent to just superhuman and I think that's going to surprise a lot of people when it happens I just don't know exactly what the timing looks
33:33
like yeah yeah what do you expect like let's say next year at Tesla Aid 2023 what kind of optimists can robot would
33:40
would you expect into the demo that's a good question I think the capabilities will be vastly
33:46
improved in terms of its ability to move and manipulate objects so we'll probably see some demos where there's much finer
33:53
motor control the first version looked great but obviously needs some work but I think
33:59
we're going to see a lot better ability to move through an environment to manipulate objects and maybe some other
34:06
examples of of behavior that seems intelligent or might be surprising to a
34:11
few people manipulating things that at this point in time require an insane amount of
34:16
dexterity I think that's going to be a big tell is how much progress Tesla makes there I think the brain itself is going to continue to get better we've
34:23
seen the progress with FSD we know what's kind of happening there but it's physically moving through the world and manipulating objects that I think is
34:29
going to be a big Focus because what we saw at AI day 2022 most people are
34:34
pointing and laughing they don't really understand how quickly this project came together but I think that's going to be the massive Improvement that we see at
34:40
least that we can notice and yeah what do you what do you think what are your thoughts yeah I I think the
34:46
next few years are going to be exciting for those who see the potential but it will be a bit I would say slow but like
34:54
I'm not expecting a crazy robot next year you know it's yeah um as long as
34:59
they yeah they solve some major problems like Locomotion they can actually have a
35:04
decently moving around robot um the robot can pick up stuff you know put it in places confidently
35:12
um and do some basic tasks and it's actually they're doing their own actors
35:17
their own you know your own that's all this stuff is working then I think that's a huge success because that's the
35:24
next building block for the next year right to to iterate on that but yeah again it's like it's I think the AI role
35:31
is going to be the perfect example of like this crazy S curve where you know we see some some progress and those who
35:39
see the the future are going to be excited about it but others you know be like every AI day for the next few years
35:45
will be like um we'll be um what uh test the queue trigger yes I I 100 agree I
35:53
think that yeah the progress if you're in the know if you really know what's going on you understand you'll be blown
35:58
away but I think the general public perception for the next few AI days is like is that seriously it really I hope
36:04
you spent a whole year and that's all it can do um but yeah I think if you understand the difficulty of some of the underlying
36:10
problems I think Tesla is going to be making incredible progress um yeah yeah it's funny how a lot of
36:17
people probably who are ridiculing Tesla Optimus now who knows in 10 years
36:22
they'll probably maybe The Optimist will be making them lunch you know yeah that's right or spooning with them in
36:29
bed exactly um you said I I think it was in one of
36:34
your videos or maybe a tweet you said that Facebook and Google um them if they moved if they had a
36:41
first mover advantage in AI robots that would be like a nightmare scenario for you why why what's your kind of
36:47
reasoning behind that uh I don't see them as competent custodians of what will ultimately
36:53
become AGI um I think that both companies have horrific red flags wherever you look I'm
37:00
not disparaging them completely I mean Google makes some great products Facebook's made some cool stuff as well but
37:06
Facebook I mean I probably shouldn't even really need to explain why we don't want them developing AI I mean the
37:11
amount of mistakes it's like no offense to Mark Zuckerberg but he's not the kind of CEO I want anywhere near
37:18
that kind of Technology even if he means well the track record not so great it appears to me at this point in time that
37:24
Google is Thoroughly infected by the work mind virus so too with Facebook and I know it's very controversial to say
37:30
this but I don't think most of the people at these companies are thinking clearly I think they know what is absolutely best for the world the
37:36
Arbiters of what's true and moral and right and wrong and I don't want people
37:41
who who have that sense of we know exactly what's right for the entire world in charge of something that could
37:47
potentially you know disrupt Life as we know it there's a lot of scenarios in there which I think well-meaning
37:52
well-intentioned people in those companies could steer things in a very negative Direction and just imagine if there was
38:00
you know humanoid robots under the control of Google and you know a couple of years ago the plague next minute
38:06
Google's like yeah we can help send these robots to keep people at home you know I just there's a lot of scenarios like that where I know Elon would be
38:12
like get effed I'm not no that's wrong yeah um and so I just I feel like Elon
38:18
Musk has his heart in the right place he has good intentions and he makes good decisions and looks at as many of the
38:24
possible implications as he can reasonably consider and I've seen very many many examples where companies like
38:30
Google Facebook have not done that and yeah and I don't think they mean harm I just think that sometimes you can be a
38:37
well-meaning or you can mean well and do really bad dangerous things so yeah my worst scenario was any company
38:45
developing AGI that had any red flags whatsoever it sounds like Elon Fanboy coming out
38:52
but of all the people on the planet I think Elon is the man to be sort of the custodian of what we're developing with
38:57
AGI here yeah I've I've I used to think that no harm with AI everything's going to be fine I used to have long heated
39:03
debates with my father like 10 15 years ago I started reading into it more and completely reversed my position from
39:08
being like I'd all be fine like some of the AI people still say today to wait a minute there's a million ways this can
39:14
go wrong and I'm very cognizant now that in the wrong hands even if they're well-meaning hands things could pan out
39:21
in a way we really don't want yeah um yeah I I
39:27
it creates a lot of that um I think what most people might not I think be in
39:34
touch with and I could be wrong with this but my opinion is the typical route
39:39
that AI advances heads into the HEI all this stuff the typical route is not a
39:46
very good route like you're talking about a typical company run by a typical bureaucracy run by all this stuff like
39:53
collusion with the government and we have very little privacy very very little rights you have a scenario that's
40:00
just deteriorating but that would be kind of the normal path you know of
40:06
technology and I think what we have here is an abnormal path where somehow a guy
40:13
who is well intentioned and actually like has his brain on the right way
40:19
seeing things clearly has got in a position where he could actually influence the direction of the future of
40:27
the world in a sense and it's so abnormal and it's so optimistic for the
40:35
future right because otherwise you're looking at a pretty Grim situation and I
40:40
don't think people understand like the the two paths you know and how how abnormal you know unlikely this path
40:47
really is until just recently and here we are we're at the cusp of this
40:52
potentially happening and that's like I think one of the one of the underrated
40:58
missed stories of the entire decade you know um yeah fascinating stuff though yep I
41:05
agree and just one final point on that I think this is part of the reason why Elon has latched onto this with such a
41:11
sense of urgency because he knows he's the man to be yeah as you said he's the outlier it's like wait a minute I'm in a
41:17
position to do this thank goodness it's not zako or somebody like that all in this is the priority
41:23
because I think elon's now competing against the risk of that scenario in which this ends up in the wrong hands
41:28
even if they're well-meaning and yeah so I think that's added a lot to his sense of how quickly this needs to
41:34
happen and how holy Tesla needs to commit to making this happen yeah
41:39
um if we look back at the pandemic I mean I think it's at least for me it's
41:45
um the true colors of some of these companies came out with the occlusion of the government censorship oh yes all
41:51
this stuff where it's like stunning to see yeah I know I know your eyes are open even though you don't you know you
41:58
don't go on your rants you don't want to you know affect your entire audience but I noticed that your eyes are open and
42:04
and it it was a very disturbing time it really was I just went are you serious like is this actually happening right
42:11
now and the worst part to me is that people weren't noticing yeah it's just it's just like are you guys are you guys
42:17
seriously not seeing what's going on right now and it's not troubling you at all so yeah I mean they're never been
42:23
simultaneously gone yeah the amount of hubris from you know like some of these not like just some of the biggest tech
42:30
companies also the biggest you know you have Pharma government all coming together uh dictating you know just what
42:39
it just it just ah it's just yes and to imagine that they would be
42:46
the stewards at HEI you know and what would happen exactly yeah that's you know you're a nightmare I guess
42:52
um scenario yeah exactly right yeah yeah um and I would say one of the reasons
42:58
why you're probably and myself included by you're so excited and
43:04
um you believe in you know the potential of AI robots under Tesla under Elon just
43:10
because it's one of the few scenarios where it actually looks potentially good
43:15
um yeah yeah it's an extremely consequential situation we find ourselves in yeah
43:22
um what happens if if something happens to Elon do you think Tesla can Steward AGI potentially like in the future
43:30
ethically morally and in a good way or do you think you'll just deteriorate
43:36
I think that Tesla does have the right DNA Elon is radically transparent super
43:42
honest everyone who works at Tesla knows what the guy's about and I think the culture at Tesla is unique among any
43:48
company on Earth at least at the scale to be aware of a company at all I think that's a good sign of course who knows
43:55
over time what could happen if Elon disappeared tomorrow but I'd be willing to bet that's probably maybe a 50 50 chance that
44:02
things continue in the right path just because of the DNA of Tesla as a company I think the vast majority of Engineers
44:08
that are working at Tesla really want to have a positive impact on the world they could get more money working somewhere else it cost you a job they really care
44:14
about what they're doing and I think everything the culture at the company really comes from the top down and so I
44:21
think that the way that the culture at Tesla has been built is probably a very good sign that even if Elon did
44:26
disappear tomorrow Tes will continue to do mostly the right things becoming from a good place and again you know what 20
44:32
years ago do no evil was the Google motto and we find ourselves now in a situation
44:37
where I know I could literally get this interview deleted from YouTube by saying just a single sentence like that so
44:44
Never Say Never but um I am fairly hopeful that Tesla does have the right DNA now that even if Elon were to
44:49
disappear or step back or something happen I think that there'd be at least a 50 chance that they do the right thing over
44:55
the long term and I also have a sense of hope that they're far enough along with
45:02
both autonomy and also the humanoid robot that they still should be first to get there obviously will be a lot more
45:09
challenging so maybe I'm naive and a bit optimistic but I'm fairly hopeful now and a few years ago I really was losing
45:16
sleep like a lot of sleep thinking about where things could go so what do you think you think Tesla would be able to make it they've got the right DNA or
45:22
yeah I mean I think over time I'm probably a little bit less optimistic
45:28
um I think just being in charge of a large corporation you have money you have like you know quarterly reports you
45:35
have pressure stock all this stuff it takes a unique set of values and world
45:40
view to kind of counter that all the pressures and
45:47
yeah I think you know we're used to seeing Elon kind of counter that and he says like hey I'm
45:54
not trying to pump the stock I don't care about you know whatever and a lot of people might not believe him but over
46:00
time his message is consistent and his Focus also in what he's doing like he doesn't need to be doing what he's doing
46:06
really um all the hours and the pain you know it's just like it's crazy
46:12
um but yeah that type of individual it's like huh it's a tough one
46:18
um and we're talking about the future of humanity that is many decades also going
46:23
forward so sure let's say things are great for the next 10 or 20 years but
46:29
what's after that right um it's definitely something I'm personally concerned about I think also what's
46:36
interesting is I never would have imagined that investing in a company
46:42
would become not just an investment but it's like it's it's so tied to the future of the
46:48
world that in a sense I care more about that future of the world than I do about my investments in the company where it's
46:57
it's much more significant you know like the mission of the company and it's interesting how the mission of the
47:02
company before just transitioning the world to sustainable energy that's a huge mission in and of itself with huge
47:07
consequences but then when you add this the whole HEI AI robots like this Mission this is this is there is I
47:15
wouldn't say there is no Mission but yeah it is pretty much it really is it really is the end of missions it's like
47:22
yeah yeah and when you think of it like that it's like man there I don't think
47:28
there has been a company you know obviously this is consequential um and we are at these crazy A crazy
47:35
moment in history you know and lots of things are gonna gonna happen and we
47:40
don't know how it's going to turn out um yeah so it's um yeah definitely interesting times yeah absolutely yeah
47:50
um so um how about um I'm curious if you
47:55
um I want to hear your your thoughts on the stock market Tesla stock so we're
48:01
recording this Monday night um and today Tesla stock dropped I think over eight percent
48:07
poor delivery or delivery numbers yeah it didn't mean at least
48:14
AI disappointing exactly yeah yeah so you know over the years you've been kind
48:20
of you know taking this counter kind of intuitive approach where you're like
48:26
with a long-term perspective you know dips are opportunities you're just buying buying buying whoa what what's
48:32
your current thoughts kind of on Tesla stock are you still kind of you know buying with whatever money you can come
48:38
up with um what's your kind of perspective on that yep so um I do have some letters here
48:44
from the tax office I won't go into too much detail there but I am prioritizing Tesla stock
48:50
um at this point in times in terms of buying I'm always looking into the long-term future and at this point in
48:55
time as I've said on my channel a million times I am personally unaware of a better risk adjusted opportunity in
49:00
the stock market it isn't to say that they don't exist I just don't know of any and I'm looking into the future looking at what's happening looking at
49:07
the potential factoring in the worst case scenario the best case scenario what I think is probable
49:13
and there's nothing that compares and so if I have some money available to invest
49:19
it's all heading into Tesla at this point in time and I'm expecting over the next decade or more probably a 10x is
49:25
somewhere around the threshold give or take you know a few X but then if we add another decade onto that at another
49:32
decade onto that if you have asked me in 2016 what do I think my Tesla stock investment is going to do over the next
49:37
30 Years it would be a very different answer to today Tesla continues to come up with they just stack an S curve onto
49:43
an S curve onto an s-curve and I'm modeling Tesla privately out till 2069
49:49
and obviously the it gets very uncertain but I've got to at least do the exercise what if these Bots do scale how big does
49:56
a fleet of bots get how does the and it just gets absolutely obscene um and so I don't want to be looking
50:02
back in 2040 2050 or 2060 and thinking man I really wish I had you know invested a little bit heavier in Tesla
50:08
you know I don't personally need the capital but I do have intentions for doing some very Capital intensive philanthropy in the future and so I'm
50:15
trying to snowball my Capital as much as possible and I think that gives me a little bit more of a risk tolerance as
50:20
well because I'm not you know if I lose it okay whatever it's not going to change my lifestyle uh but yeah I just as I said it's the
50:26
best risk-adjusted opportunity that I'm aware of in the stock market and it's no contest that being said at this point in time with the turmoil in the stock
50:32
there's Bargains left right and Center you know yeah you could take a scatter approach now and just bang this grab this grab this grab this and you'll
50:38
probably be pretty happy in five years time um but Tesla's just there's been one sequence of events after the next after
50:45
the next it's just been pummeling the stock price and I personally feel that the stock is massively artificially
50:50
suppressed I think it's getting sort of spring-loaded and at some point you know when the macroeconomic environment calms
50:56
down a little bit there's a little bit less panic and fear people start looking at Tesla's earnings and looking at their growth and going wait a minute what are
51:03
you serious this is a bargain like Tesla at this point in time I feel is in Bargain Basement territory and of course
51:09
it still could go a lot lower I'm not making predictions over the short term but it's a head scratcher I mean
51:15
it really is I just every spare Cent that I can possibly accumulate is going straight in and I'm you know I'm even
51:20
considering the Big M word margin yep got it
51:27
um no yeah so considering that as an option as well but interesting I'm not in a rush yeah it's you know with Tesla
51:35
stock it's so interesting because you know whilst you start gonna give any evaluation to to Optimus if any will be
51:42
negative right um yeah I agree and you know for the next five or ten years they'll be largely you know autos and FSD
51:51
um and but when you look at the bigger picture Beyond let's say 10 years and
51:58
what Tesla's setting up with AI robots it almost makes like the next five or
52:04
ten years like stock price like irrelevant or at least if you have that long enough perspective you know what
52:09
I'm saying it's like yeah does it really matter like what the stock price is doing for the next five or ten years if
52:15
Tesla actually my thinking yeah that's literally my thinking I mean this is why I was happily buying at all-time highs I
52:22
I knew you know probably a little bit ahead of itself it'll come back but I've got a long term I don't care if I'm paying 1200 pre-split what's that 400
52:28
now you know 200 500 800 I just don't care because I'm not thinking five ten years I'm thinking 10 20 30 years into
52:35
the future and you're right I don't believe that Wall Street the stock market in general is giving any any
52:41
Credence at all they're just they're completely ignoring the possibilities with Tesla bot and AGI they just they don't exist they're not looking at far
52:47
enough into the future and and doing the exercise what if this actually plays out and so I think Tesla's just going to be
52:54
a company for a very long period of time where there's a huge gap of years between where the stock market thinks Tesla's at and what their future looks
53:00
like and the reality the massive massive disconnect and and I don't think with Tesla bot and what will
53:07
eventually become AGI that the Market's really going to start getting this for at least half a decade or more
53:12
because they're going to say hey I know we'll fill over it's not very you know it's not really going to look like a whole lot of progress and it's only I
53:19
feel like Wall Street and and most most the general retail investors even they need to see something happen today to
53:25
understand that it's going to happen tomorrow they're not able to look out far enough into the future and weight the probabilities of this happening and
53:32
think about is this possible and if not can somebody else catch them blah blah blah um and I understand why because what
53:38
Tesla's doing it needs you need a pretty broad grounding of knowledge and a lot of really nerdy topics to just to even
53:44
comprehend what's going on you see a robot and you're just a member of the general public at AI day too the
53:50
Boston Dynamics have already done that you know I don't get it but if you've got the additional layers of context and
53:56
understanding you go hang on a minute what they're doing with the brain here is just on a whole other level so I
54:01
think it's only a long time with the market not understanding and I agree this is why I just when I have money I'll buy Tesla stock I'm not trying to
54:07
time it I know I could probably sit here and wait for a big crash and load up but it's just like I'd rather just buy when I've got cash and I don't really care
54:14
what the price is you know if Tesla was at a five trillion dollar market cap I'd still be buying today
54:20
and I'd still be very excited for the prospects but it's much less than that now so even better returns yeah
54:26
interesting um do you have any updated uh thoughts on crypto I know you know you've in the
54:31
past seen the potential of crypto but you've been kind of reserved in terms of investment is there any point where
54:38
crypto comes down that you would say huh interesting uh you know I would it would
54:43
be a good investment I think we're getting close to those those points but there's no way I'm going to be investing
54:49
in crypto when Tesla's still in this these prices now some additional context some of the viewers may not know I did
54:54
actually previously invest in Bitcoin in 2013 and 14. I allocated about 15 of my
54:59
net worth I said this is high risk High reward I can see this thing potentially going to zero or a hundred thousand dollars whatever
55:06
I had you know Bitcoin in different exchanges long story short I got scammed I got hacked and I ended up losing
55:12
almost all of the Bitcoin and that was well over a hundred Bitcoin so for me
55:17
psychologically to consider re-entering at a price point instead of 200 300 for a Bitcoin or something like that I just
55:24
don't see the reward there given the uncertainty of the future I think crypto is here to stay bitcoin's probably here
55:30
to stay as well but it's so early and so uncertain and such a volatile asset class that I'm not confident enough to
55:37
say okay there's definitely something here I'm willing to allocate Capital even though I think the potential upside could be enormous
55:43
it's just something it's very still very speculative for me um and so still buying Tesla stock every
55:48
spare Cent but I'm very curious to see what's happening I've been paying attention to the nft market out of Interest I kind of I had a hypothesis at
55:54
the time it looks like most of this is going to be coming down 95 100 almost from where
56:00
we are today I don't say anything but I just I always test my hypothesis versus reality and I think we're crypto and this this
56:07
whole Space is going to go through a series where you just to the moon and it comes back to reality and again and again I don't know how many more times
56:13
this plays out but I think over the long term it is here to stay but it's really hard to pick winners and I can't make
56:19
any statements with a real high level of confidence yeah yeah yeah I agree when you compare the investment opportunity
56:24
to Tesla it's it's different you know there's so much more uncertainty I think
56:31
um where I mean Bitcoin Maximus will tell you there's 100 certainty right
56:36
here but shout out to Michael somebody yeah yeah
56:41
um this is investment advice yeah um I just yeah sell everything though exactly I just I think it's on a
56:48
different level um yeah compared to Tesla though um uh I know I I was reading a recent
56:54
thread of yours about the crater economy and you seem very bullish on kind of the
57:00
future of content creators kind of taking over more of the market as they
57:05
pretty much like you know release products that are already selling just to Their audience
57:12
um how like how big do you think this where where is
57:17
this headed the Creator economy is it Like It's Gonna Hate Everything I think I really it's going to eat everything
57:25
I mean I really genuinely believe that um one great example Mr Beast who I've
57:30
begun watching a lot of content creators on YouTube not because I actually want to watch the content but I want to see what they're doing and understand how
57:36
they're doing what they're doing Mr Beast is a great example I'm sure you've heard of him yeah biggest YouTuber basically uh and just been
57:44
reinvesting in the channel no one can really compete so Mr Beast the the brand the channel launched Beast burger and
57:51
overnight they opened a bunch of like ghost restaurants where they just using someone else's kitchen to cook their
57:56
orders and out of nowhere just simultaneously open a few hundred stores in its second year Beast Burger is going
58:02
to do over 100 million dollars of Revenue and every time Mr Beast posts a
58:08
new video he's got an audience of 50 plus million viewers it's like 10 Super Bowl ads every video
58:15
hey we've got a new product bang check it out and if you're a fan I'm already going to buy a burger why would I go to
58:20
McDonald's when I can buy a beast Burger instead and I don't think people are understanding this the the distribution
58:26
costs are effectively zero customer acquisition is almost nothing you've already got the trust here and I don't
58:31
see a way for traditional Brands to compete unless they themselves are creating content because everyone
58:37
watching this will understand there's YouTube is that they really enjoy it there's other content creators that they really enjoy watching and if this person
58:43
were to drop a product that they already buy and it's comparable in terms of quality and price would you rather a big Corporation a
58:49
faceless Corporation get that revenue or would you rather help the person that you already enjoy watching their content
58:54
uh I don't see how Legacy Brands will compete with this unless they're creating their own content you know maybe Coca-Cola has their own podcasts
59:01
and has a huge thing and people love the coke podcast and they'll still stay loyal to the brand but if not
59:06
one after the other own we're starting to see all kinds of weird convergences the YouTube Stars doing boxing right and
59:13
making more than most professional boxes would ever make in their career a bunch of amateurs punching each other in their head next minute they're taking home 20
59:19
30 40 million dollars uh I think people are massively under undervaluing how
59:25
much power I guess is probably the word an influence that creators can have and I just don't see a traditional brand
59:31
being able to compete with certain certain stuff you know if you're known for a certain thing if you're a YouTuber
59:36
and you talk about health and nutrition everyone wants to know what do you take what supplements next minute you release
59:41
your own supplement it's actually good people are going to buy it and if you've got 10 million subscribers and you're a
59:47
GP on YouTube suddenly you're no longer need to be working as a GP you can actually build out a huge supplement
59:52
company I know if David human or uh Andrew huberman or David Sinclair were
59:58
to start releasing products they would have 100 million dollar companies almost overnight because they have a wide
1:00:03
audience they're interested in health everyone's every time I mention some of the stuff that I've taken for energy everyone's begging me what product you
1:00:09
use what brand do you use and you know the potential there is just astronomical
1:00:14
I don't think people are getting it what what's your what's your game plan with YouTube
1:00:19
um like I know Finance investing Tesla is is one area one Niche are you happy
1:00:26
with that kind of Niche to stay there and to do your thing or are you tempted at all to kind of you know go into other
1:00:32
areas that might be even broader and bigger yeah um definitely in the future I'll be expanding a lot of different
1:00:39
things on topics that I am interested in passionate about that I'd like to discuss my limiting factor until fairly
1:00:45
recently has been energy since we last spoke maybe two years ago I've more than doubled my energy and the way that I
1:00:50
measure that is my work capacity and ability to recover so I can do twice as much in a day now versus two years ago
1:00:57
and I'm at a point now where I do actually have enough time and energy to start adding things on and going down
1:01:02
other rabbit holes I'm all in on this Tesla content and I'm going to be covering this for decades that's not going anywhere but I'm trying to sort of
1:01:09
hack my life and go if I can extract more energy out of the day and more productivity and do more I'm just going
1:01:16
to keep adding things on and on and on over time there's a lot of areas that I'm very passionate about some that I'm
1:01:21
kind of knowledgeable but not even close to expert level on yet but in the next few years you might see a few new channels popping up
1:01:27
and some of them may not be able to appear on YouTube but others will and yeah I think I'm just kind of in my mind
1:01:33
I won't have a positive impact on as many people as I can and you know there's some knowledge I have on Tesla
1:01:38
which can be useful to some people but there's a lot of other things that I convey a little bit on patreon but I don't really get to to sort of talk
1:01:43
about on YouTube and I know most viewers don't care so I'm going to drive people away if I start going on too many off-topic rants but I would like to
1:01:50
share that stuff in the future um so yeah there's definitely a lot more to come we're just getting started yeah
1:01:56
it's like a yeah I've watched Mr Beast his interviews for past few years I've I've watched a lot of his and
1:02:05
it bothers me in a sense because he talks about like you know the opportunity he's seizing as
1:02:13
something that's so big and it makes so much sense
1:02:19
um it makes me question sometimes like what I'm doing like I'm like am I being
1:02:25
irresponsible like should I like you know basically shoot for the stars you
1:02:31
know um I battle with this as well yeah but I have the same same thoughts yeah because
1:02:36
I'm like he's definitely tapping into something that and and other big you know content
1:02:43
creators that is on a different scale and level but you know
1:02:50
I I and I'm comfortable and I like where I'm at it's like it's it's nice I don't
1:02:56
have to stress I don't have employees or a company I just like chill and you know come up with content and it's it's nice
1:03:03
for me but I do have this question mark in the back of my head where I'm like am
1:03:08
I like I would say missing out but being irresponsible like for not going for something much much bigger and tapping
1:03:15
into something else but um yeah it's something that it bought it's it's like a constant like
1:03:22
bothering like you know like something just in the back of my head always it's just yeah so anyways I'm gonna talk
1:03:28
about content yeah when you talk about content economy all this stuff like yeah it just it resonates a lot I think
1:03:35
um yeah I'm definitely cognizant there's a because I'm the same I think you realize there is some untapped potential
1:03:41
that you have and you're very limited in focusing on a really small narrow Niche and the thing that I battle with the
1:03:48
most is I don't want to be disingenuous about anything and I can't do anything but be me 100 percent
1:03:54
and I know that there's a lot of things that are against best practices that limit growth and if I wanted to start a
1:03:59
more generic channel in the future I could easily you know try and be a Graham Stefan or meet Kevin and expand just even in the finance space quite a
1:04:06
bit and there's many levels beyond that as well and I am very confident that I could execute this strategy no matter
1:04:11
what I was focused on that could just reach a audience that's 10 100 times larger but in order to do that
1:04:18
I can't see a path without me having to conform to the algorithm or just not be
1:04:23
my full and genuine self um but it's something that I I definitely think I kind of battle with a little bit as
1:04:30
well because yeah I mean you have a gigantic brain I think don't people don't realize quite the intellectual Titan you are and I I don't say that to
1:04:37
blow smoke up you are so I generally see what you're capable of I understand your history with the App Store seeing the opportunity and executing and I know
1:04:43
that if you were to go you know what let's just 100x what we're doing now you'd do it you would execute on it and
1:04:49
and do something and it's just is it worth kind of going on a different path to to reach more people and have
1:04:55
more impact and something that I battle with because I can't not be me but I know if I was able to turn it down and
1:05:00
do things it could go on a different path and reach a much wider audience and probably have a bigger net positive impact yeah it's a tough one yeah yeah
1:05:07
definitely um I wanted to ask this I was going to ask you offline but I figure just ask you right now
1:05:13
um how is your merch sales doing like is that a decent significant size of you
1:05:18
know Revenue coming in is it worth is it worth like you know
1:05:24
something that's not time energy not well actually yes if I put the time in
1:05:29
when I was actively promoting merch last year as an experiment I just thought I'd just pump merch constantly and I don't
1:05:34
just mean insert I mean like talk about the merch uniquely in each video rather than just copying and pasting the same
1:05:40
promo it's much more effective to do that and integrate it yeah when I was doing that I was selling hundreds of dollars a merch every single day which
1:05:47
is a lot if we're posting a daily video it adds up to a lot if I took the time to do merch designs and hire artists
1:05:53
instead of whipping them together myself I could probably be making more income significantly more income from merch
1:05:58
sales alone than all other combined revenue sources by a massive margin it's just that I'm not interested in creating
1:06:05
regular new merch and promoting it constantly I'll do new things I'm more just focused on the videos and I'm leaving a lot of money on the table
1:06:11
there but I think for most creators if you haven't if you have
1:06:16
fans rather than viewers or it's a subject that people are very passionate about
1:06:21
people love merch they love to support creators and if you've got compelling designs and they're fresh and new and
1:06:26
unique people love it and you don't need to do a huge conversion to to get a reasonable you know if
1:06:32
you're posting very frequent videos and you're mentioning your merch every time it doesn't take that many people clicking through if you've got a five or
1:06:38
ten dollar profit per merch item that for it to really start adding up I think there's a lot of YouTube creators that
1:06:45
significantly more than half their revenue comes from merch even if they've got ads and sponsorships and placements
1:06:51
and stuff like that um huge untapped potential I think a lot of creators yeah and obviously I have an
1:06:57
unfair advantage in terms of the Elon and Tesla nerds because if you want a Tesla's meme shirt
1:07:02
there's not that many places you'll you'll find them or there wasn't anyway it's probably changed a lot now I've seen a lot of people copying my designs
1:07:08
and stuff like that but there was a period there where if you wanted to don't bet against Elon shirt I mean there's only one of those defined I
1:07:15
think that that unit alone has sold well in excess of 4 000 individual um items which is stunning to imagine
1:07:21
they are great conversation starters though yeah yeah yeah yeah no you make some great designs I definitely
1:07:26
appreciate it um yeah fun stuff um it's a great catching up on especially right after
1:07:32
Aid and getting your thoughts um I'm sure people will appreciate getting into
1:07:38
your head a bit um and a conversation like this just to take a step back and look at the bigger
1:07:44
picture I'll go ahead and link to your YouTube channel in the video description below any other kind of last um kind of
1:07:50
words or comments before in the end I would probably encourage everyone watching who's really not buying into
1:07:57
the thesis around the board and the AGI to try and educate yourself on as much as you can to really make a good
1:08:03
assessment of whether or not Tesla's on the path and How likely it is that they'll achieve this because even if you
1:08:09
don't buy it now just if you can put yourself in a situation where you can definitively you know what they're wrong
1:08:14
I don't disagree we have enough information to really make that judgment I think you'll be in a much better position to see the potential
1:08:19
opportunity here because as I mentioned I really do think you need a lot of Fairly specialized knowledge on a broad
1:08:25
range of really nerdy things to grasp the opportunity and as I said the bot
1:08:30
AGI dwarfs everything else in Tesla's future I'm by all multiple orders of
1:08:35
magnitude and the implications of this are ridiculous you know market cap it just it gets bizarre but I think a lot of
1:08:41
people are having a hard time making that leap because they don't have there's too many gaps in their knowledge to really kind of draw any reasonable
1:08:47
conclusions about what they're seeing uh but don't miss out on understanding the opportunity guys that's my final word all right good stuff um yeah thanks Anna
1:08:56
thanks in chatting again um loved it and we'll see you um see you later and see everyone else who's watching thanks
1:09:02
see you guys
0:07
嗨史蒂夫歡迎今天我和斯蒂芬馬克瑞恩一起解決金錢問題 YouTube 頻道
0:14
嗯，我過去採訪過他，這些年來認識他真是太好了，我會繼續前進並鏈接到
0:19
視頻或下面視頻描述中的採訪，所以在今天的視頻中，我想深入了解特斯拉的 AI Day 2
0:26
特斯拉的擎天柱機器人人工智能的未來特斯拉股票和很多熱門話題
0:33
歡迎您在頻道上斯蒂芬，您過得好嗎？戴夫，很高興回來，期待我們的聊天，是的，是的，這是一個
0:40
偉大的呃連接所以我很好奇你的反應是什麼我在帖子上看過你的幾個視頻阿爾忒彌斯
0:48
嗯，你顯然抓住了大局，所以是的，幫助我們幫助我們看看
0:54
你腦子裡在想什麼你的反應是什麼，是的，你從特斯拉人工智能日得到的收穫是什麼
1:01
兩個好，我認為對我來說最大的收穫是他們會這樣做
1:06
進度 嚴重性 在執行該項目及其領導方面的極端緊迫感
1:12
我認為很多人剛剛完全忽略了一件事，他們正專注於身體而不是
1:19
他們有大腦 他們不明白特斯拉採取了什麼他們有效地移植了大腦
1:26
從他們的自主軟件他們的計算機視覺到擎天柱人形機器人，他們是光年
1:33
我的意思是比世界上任何其他人可能在大腦方面領先五年或更長時間
1:38
人形機器人 身體 你知道的運動 執行器 他們會弄清楚所有的東西 那很容易，但是
1:44
在現實世界中，您所知道的計劃和行動中最重要的視覺感知是大腦，這是
1:50
我認為很多人都忽略了一些東西，我沒有看到有人得到特斯拉擁有的東西
1:55
開發了大腦 他們現在已經有了 AGI 的基礎構建塊 他們非常認真地執行這個 他們有一個非常可行的
2:02
計劃實際擴大生產規模 從他們自己工廠中的機器人開始，它們幾乎毫無用處，事實上它們最初會比無用更糟糕
2:08
它們會降低生產力，但隨著時間的推移，特斯拉可以提高能力，一旦他們
2:14
有效並為他們自己的工廠增加有價值的勞動力，他們將立即擁有具有相似需求的客戶
2:19
他們可以開始銷售這些地點，然後最終他們的艦隊規模就像車輛一樣
2:24
隨著時間的推移，反饋循環變得越來越好，而超長期 AGI 可能就像 Elon 提到的那樣
2:31
他們正在做的事情的新興財產，我只是看不到任何人或任何人
2:36
公司從這裡趕上特斯拉，現在我知道這聽起來像是一個稀釋的特斯拉迷
2:42
展望未來並思考有人如何趕上特斯拉他們今天需要如何部署百萬
2:48
需要在現實世界中部署數以百萬計的機器人（無論是車輛還是類人機器人）來啟動數據
2:54
飛輪和訓練效果，我沒有看到它發生，嗯，我想實際上你提到了
2:59
最近在一個視頻上，我想我看到了一個與 Emmett 辣椒的聊天，你可以得到你需要的一切
3:05
知道互聯網上人類知識的總和，這很棒，但你也需要我看不到的真實世界經驗
3:11
任何人都捕捉到現實世界的東西，在這個時間點，我認為很多人都被抓住了
3:16
裝飾著門面，並沒有真正了解這裡發生了什麼，但我對未來感到非常興奮
3:21
特斯拉的我，我昨晚睡不著覺，我躺了好幾個小時，只是想著所有這些可能性
3:28
嗯，是的，我認為在人們真正理解這件事上發生了什麼之前，可能需要很長時間，可能需要十年或更長時間
3:35
是的，但是是的，特斯拉似乎是通向 AGI 的一條非常清晰的道路是的，是的，我想嗯，是的，有很多東西
3:42
繼續我覺得人們錯過了你所知道的十年的故事情節
3:47
嗯，這將影響下一個你知道的世紀，我想如果人們會看到特斯拉的業績記錄
3:53
埃隆看到他所顛覆和信任的行業，只相信特斯拉所說的他們正在做的事情
4:00
現在你知道我認為在 Aid 問題是他們沒有出現
4:05
可能比已經存在的更好的東西，人們就像是正確的，但是
4:11
他們所展示的是他們正在使用設計的執行器所做的工作，所有這些你都知道人工智能大腦所有這些東西，呃
4:18
尤其是同類最佳，就像沒有人試圖按照他們正在做的規模做特斯拉所做的事情，我認為
4:24
如果你只是停頓一分鐘，忍住你的憤世嫉俗，你的懷疑，說好吧，如果
4:32
特斯拉真的在建設大腦，他們真的在建設你們所知道的
4:38
車身和大規模製造的所有設計，你把這兩者放在一起會發生什麼，你
4:45
可以為接下來的 5 年 10 年 20 年 30 年以及更多
4:50
把它畫得更瘋狂，讓你知道這就像我什至不想談論過去的 15 年，因為
4:58
它只是你會迷失在互聯網上是的，是的，嗯
5:04
是的，所以我很好奇我想剖析一下你所做的假設，因為
5:11
很明顯，你正在看到某些事情，並且你假設嗯，這些事情正在將這些想法聯繫起來
5:18
你認為特斯拉要去你認為這個機器人的成功將釋放你的大腦
5:25
知道基本上讓每個人都感到驚訝，但你認為你所做的關鍵假設或關鍵是什麼
5:32
你知道點，你相信也許其他人看不清楚 嗯，我認為
5:38
正如我在介紹中提到的，最大的一個是特斯拉以特斯拉的願景建立了 AGI 的基礎
5:45
在通往 AGI 的道路上最重要的事情是視覺感知規劃和行動
5:52
現實世界和特斯拉還沒有完全確定這一點，但他們已經種下了種子，看起來
5:57
對我來說不可避免的是，如果你把時鐘向前跑得足夠遠，特斯拉今天所擁有的
6:03
繼續改進和發展，變得更好、更有能力和更智能，它只是繼續從
6:09
你知道對特斯拉 Q 到低於平均水平的智力到最終人類水平然後最終
6:16
超人，似乎看不到任何盡頭，除了特斯拉已經建立了基礎
6:23
我幾乎無法想像一個現實的場景，任何人都可以趕上特斯拉，他們只是
6:29
就像我說他們需要自己的艦隊他們需要飛輪發生所以我的論點是在某個時候 AGI 發生
6:35
在某個時候，在智能方面比人類更有能力的類人機器人將會出現，這只是時間和誰做的問題
6:41
而且我不知道我們可能離那個點還有幾十年的時間，但不管具體時間如何
6:47
當它發生時，似乎很明顯特斯拉有一個無懈可擊的領先優勢
6:52
做到這一點，我知道現在調用它還為時過早，但需要發生一些非凡的事情
6:57
我要改變我的立場，我願意改變主意，這也是我
7:02
想想現在 Elon 已經意識到 AGI 的價值的一點，一個開關在某個點上他去了哦
7:08
等一下，等一下我們用特斯拉的願景得到了什麼，他多年來一直擔心人工智能
7:14
長手上的人工智能，我也失眠了，我認為埃隆有
7:19
到了現在他覺得在特斯拉而不是你知道的情況下實現這一點的道德和道德義務
7:26
讓 Facebook 嘗試做同樣的事情，現在他們正在積極招募人才
7:31
作為一家公司，展示非常早期的技術原型併計劃招聘工程師的風險很大，我認為
7:38
在每個 AI 日期，Elon 都提到了每月更新播客的可能性，似乎 Elon 和 Tesla
7:45
絕對一心要做到這一點，並招募需要在那裡實際執行的人才
7:51
在這一點上，沒有任何其他公司如此認真地對待這一點，也沒有任何其他公司擁有這樣的基本組成部分
7:57
嗯，我認為它有
8:03
這裡的證明是我有點展望未來，我可以看到不可避免的事情，但我認為這需要很多時間
8:08
在其他人開始看到足夠的證據得出相同的結論之前的幾年，嗯，所以我認為有一個巨大的
8:14
在這裡斷開連接，我的意思是 EVS FSD 能量 沒什麼 得到它沒什麼 我
8:21
意味著它實際上什麼都不是，與人形機器人的潛力相比，它只是一個規範，是的，論文
8:27
本質上是如果你能用有用的勞動力無限擴大經濟規模
8:32
變得超級荒謬是的，是的，我想深入研究一下，嗯，但在此之前，這是一件有趣的事情
8:39
想想特斯拉如何嘗試解決 FSD 和 FSC 花了這麼長時間並且基本上是一個比這更難的問題
8:45
預期它導致他們走上這條道路，他們必鬚髮展這種瘋狂
8:51
世界級的人工智能團隊、基礎設施和計算，基本上他們必須
8:57
就像您知道的那樣，咀嚼的次數比他們預期的要多得多，但是這樣做會導致埃隆走上這條路
9:04
就像等一下下一步就像一個人形然後
9:10
你有潛在的 ATI 而這正是他所擁有的
9:16
嗯，從某種意義上說，過去他希望通過開放式人工智能來管家，你知道亞洲的方向
9:23
以正確的方式，但現在他在特斯拉獲得了這個機會
9:29
知道你不可能預料到這可能就像你知道的那樣，是的，六七年前有點難
9:35
我完全同意，我認為在某些時候，就像我說伊隆腦袋裡的燈熄滅了，他就像哦，好吧，這是
9:41
完美，我們走吧，嗯，你為什麼認為特斯拉會加入
9:47
與穀歌相比，可能解決 AGI 的領先者
9:53
你知道特斯拉與穀歌賽跑還是開放AI
10:00
現實世界的應用程序，正如我提到的，你可以在互聯網上爬行，有很多可用的知識
10:06
那裡可以訓練圖像，識別和分類，等等等等等等，但是當
10:12
談到在現實世界中的互動，我沒有看到像谷歌這樣的人亞馬遜會選擇你的大型科技公司
10:18
任何需要在那裡實際收集足夠數據以產生有意義影響的規模的東西，所以如果
10:24
谷歌想嘗試趕上特斯拉，他們可能需要建立自己的自動駕駛車隊，我的意思是數百萬
10:30
他們中的一些人開始以這種形式獲取一些數據，或者如果他們真的想要，他們可以從人形機器人開始，但我認為他們會處於
10:37
缺點，因為有很多有價值的有用信息，而且他們真的可以開發出很棒的計算機
10:42
僅通過車輛的願景特斯拉已經領先了這麼多年，這只是一個反饋循環
10:49
他們擁有的時間越多，數據越多，他們的神經網絡就越高效，我發現他們的神經網絡變得更好、更有能力
10:54
很難想像像谷歌這樣的公司或另一家大型科技公司有合法的機會讓特斯拉倒閉，讓我們走吧，你知道
11:00
如果它盡快投入 1 億美元到這個項目上怎麼辦 讓我們做任何事情
11:06
需要開始嘗試擊倒特斯拉，我認為這樣做所需的勇氣
11:11
您可以想像如果 Google 的投資者說您將在今年投入 1000 億美元來啟動我們自己的飛輪，他們會作何感想
11:18
嘗試在這個瘋狂的人形機器人上抓住特斯拉，大多數人認為這不會發生我認為特斯拉公司的敏捷性
11:25
公司的 DNA 幾乎允許他們承擔這些風險，而不會受到與
11:31
另一家公司會讓他們仍然擁有那種創業文化，我認為大多數大型科技公司都是
11:36
現在的龐然大物，他們並沒有真正願意將公司押注於某事或承擔這些巨大的風險並讓您投資
11:43
是的，你總是有一些很棒的人工智能東西，但我只是他們有很小的元素他們有很棒的語言和搜索和類似的東西，但是
11:50
我只是沒有看到任何其他大型科技公司擁有與特斯拉用他們的 Vision 系統構建的產品相媲美的東西，你對此有何看法
11:56
你看是的你怎麼說是的我的意思是我正在和一些呃訂閱者的追隨者共進午餐
12:04
嗯，有人評論說蘋果永遠不會像特斯拉在公共 FSD 測試版上所做的那樣，就像它只是
12:11
好吧，享受他們的形象，一切都好，是的，金廚師會嘔吐
12:17
只是粗暴地思考這樣一個想法，而你所知道的那種想法就是差異的例證
12:24
在特斯拉和你知道埃隆喜歡的蘋果之間，我嘗試了最新的
12:29
你知道最瘋狂的測試版可能會殺了我，他是因為他自己，這與
12:37
經營一家規模龐大的典型公司，比如說價值 1000 億美元的公司
12:42
嗯，是的，我想問題是你知道谷歌開放人工智能其他人會
12:50
計算機上的 AGI 與現實中的 AGI 不同
12:55
生活還是你認為它們會相交並成為一個你認為最終特斯拉
13:03
可以競爭和超越你知道那些公司的在線人工智能專家和
13:10
他們在做什麼他們在現實世界中的人工智能發展得如此之快以至於它只是
13:16
吞沒你知道的在線你也知道存在或者你知道你怎麼樣
13:21
預見到這種情況會發生，你知道接下來的五年或十年，我絕對認為會有一個
13:27
這些能力的融合我不確切知道它是如何發揮作用的，但我確實有一個一般假設，即
13:33
特斯拉或其他任何人在智能方面正在建立的某些方面變得更好
13:39
學習它不僅僅是學習，而是在學習方面變得更好、更有效、更通用、更精簡
13:45
就像人類是一個很好的例子 你知道數百萬年的生物進化 我們的神經網絡是
13:51
相對高效 你知道一個嬰兒只需要多次採取相同的行動就可以意識到你知道他們
13:57
可以以某種方式穿越世界不要這樣做不要那樣做和人類一樣我們可以看到某人執行一個動作然後
14:02
很少嘗試你知道什麼時候這件事在人工智能日那個小愛我試圖做正確的手我以前從未嘗試過這樣做而且我
14:08
就像你知道的那樣，我花了三到四次嘗試才得到粗糙的形狀，然後大約五次十次嘗試
14:14
我可以把它猛烈抨擊，所以我們在學習方面擁有相對高效的神經網絡，我看到它達到了
14:19
嗯，你知道特斯拉正在開發的人工智能，它將能夠學習和學習更多東西
14:25
高效而迅速，我們只是我們現在還很早，現在訓練效率不高我
14:30
認為最終會在那個時間點出現
14:37
填寫那些是的，所以我認為這些東西會融合我認為這絕對是一種可能性，也許有一個超級在線 AGI
14:45
有能力，也許特斯拉的人形機器人在現實世界中更有能力，但我看到了這些能力
14:50
隨著時間的推移，你知道在我看來你可以有效地利用你的網絡和
14:55
將它們分層就像隨著時間的推移而進化的大腦我們有我們的基本大腦然後事情有了
15:01
更複雜，有點上，稍後再上，我認為這可能是 AGI 將如何開始
15:07
你知道一些非常基本的能力，並且不斷地堆疊東西，所以我不知道是的，是的，我想
15:12
嗯，我想呃開齋節 Q a Elon 說了一些關於我認為擁有數據的事情
15:18
在計算中，嗯，你知道這就像他們在裡面
15:23
至少在現實世界的人工智能中處於領先地位，但當你考慮到它時，你的數據量有可能
15:29
在現實世界中積累，嗯，與你被孤立的情況相比，它可能是如此巨大
15:37
一個在線環境，因為例如使用 open ai 的 gpt3 就像確定他們你知道刮掉所有 reddit 的你
15:44
知道所有頁面，但特斯拉也可以做到這一點，你知道這就像它不是相互排斥
15:50
se 但要像擁有讓我們說潛在的數億
15:56
機器人與日常任務交互並進行交流和實際勞動
16:03
人類這樣做並且實際上幾乎像人類一樣行事太正確了
16:09
數據量我認為您無法與僅在線類型的數據進行比較
16:15
我同意這就像我覺得它就像在線幾乎就像一個二維世界是的與
16:21
三維世界，你知道在線上缺少的東西太多了 看起來就像在線
16:27
AIS 似乎非常擅長掌握某些任務，當然很多信息似乎相當
16:33
很簡單，你可以暴力破解你的方式，它對人工智能來說非常快，有一點
16:39
最近有人發給我的人工智能，它可以轉錄一個視頻，然後試著找出視頻的信息和含義給你一個小小的
16:44
小總結，它寫道這個視頻是關於等等等等，我在我的一些視頻上運行了它
16:51
兩到三秒，它分析整個視頻會吐出一個響應，但與
16:57
穿越複雜、混亂、不確定的現實世界，那裡有無限的多樣性
17:04
知道一加一的正確答案只有這麼多，二是正確的，而它進入現實世界
17:09
事情就呈現出一個全新的維度是的，是的，我認為我認為你在這一點上是對的
17:16
這也是你在現實物理世界中學習事物的方式與你在網上學習的方式不同
17:23
比如說有人對你說了些什麼，他們給了你一張奇怪的臉，還有更多的背景和細微差別
17:29
和更多的信息要接受，而不僅僅是你知道你在網上說了一些有趣的話，你會得到一個表情符號反應或
17:37
你知道與你在物理世界中攝入的數據量相比，我認為數據量非常有限
17:43
有一個嗯，你可以做一個假設或
17:49
你需要擁有真實世界的人工智能才能進入 AGI 的論點，就像你需要獲取的數據量一樣
17:56
物理世界，因為那是大部分數據所在的地方，如果你把自己放在電腦裡
18:02
就像你知道很多其他公司一樣，你會達到你不會有的本地最大值
18:07
如果是這樣的話，像特斯拉這樣的現實世界人工智能公司所擁有的數據
18:13
這就像哦，我的天哪，就像特斯拉的人工智能功能要瘋了
18:18
嗯，是的，我也傾向於那篇論文，你知道，是的，所以我是如此
18:24
興奮是的，嗯，人形機器人的經濟潛力如此
18:31
嗯，你的大腦在哪裡？老實說，這可能是這個問題的答案
18:37
問題是當我們開始違反物理定律以及太陽系中可用物質和能量的數量時，我知道這聽起來
18:43
荒謬，但這就是我的大腦首先要去的地方，好吧，這就是真正的物理限制
18:49
我們必須從那裡退後一步，是否存在任意限制我不知道我通常沒有
18:56
回答這個問題，我的懷疑是歷史，世界 GDP 和技術在某種程度上建立在
19:02
回到以前的創新，似乎沒有合理的限制
19:08
經濟 如果我們可以創造有用的勞動力 我們可以創造有用的東西 我沒有看到任意限制和
19:16
這是一個非常令人興奮或令人不安的現實，因為如果你
19:23
想想如果我們可以將全球經濟規模擴大 10 倍或 100 倍或 1000 倍的潛力
19:28
或 10 000 X 或更多，我相信這只是一個問題
19:34
當它變得荒謬時，我的意思是它真的變得絕對荒謬
19:39
嗯，是的，令人興奮的是，你對限制有什麼想法嗎？
19:45
就像實際發生的事情一樣，因為我已經完成了
19:52
嗯，對於人工智能和機器人，人們對人類有這種偏見，我認為
19:58
這是一種幾乎是自我保護的自然偏見，很難想像機器人比你更聰明
20:05
你知道你認為你是你所知道的結構宇宙的頂端
20:10
嗯，你是最聰明的物種，並且認為你知道機器人、人工智能或計算機可以
20:16
比你聰明我認為這太不自然了，就像你知道的人幾乎要死一樣
20:24
人們更容易認為機器人可以比他們更強壯，就像你可以說好吧你
20:31
看看他們的金屬機身，你就像哦，我的天哪，你知道的，但要更聰明，這就是我想的地方
20:37
嗯，我會說大部分是悲觀主義，但沒有利用我的想法
20:43
如果你看不到機器人會更聰明，機器人的潛力將是人工智能
20:49
比起你，你不會看到你所知道的未來的潛力，但如果你把它放下並說好吧
20:54
如果這些神經網絡發展得如此之快以至於機器人更聰明人工智能會怎樣？
21:00
比我聰明，幾乎所有我仍然認為人類會做的事情
21:06
幾乎我會說我會說神秘或超自然，但超越
21:13
也許是人類獨有的原因或理由我仍然認為這會發生，但就純粹而言
21:21
智力和能力是的，你知道這很有可能
21:26
這些神經網絡一旦它們 Eclipse 人類你知道智能然後它就會以指數方式上升到你無法做到的地方
21:32
甚至將你知道他們的智力與人類智力進行比較，然後你就會進入他們可以做出的場景
21:39
更多機器人自己 他們可以創造自己的因素 他們可以設計我們甚至無法理解的東西
21:44
他們正在設計嗯，這是對的，是的，這基本上是自我複制的，你知道的
21:50
物種差不多，這就是你喜歡的地方哦，是的，沒有盡頭
21:55
可能，這就是當我們談論經濟潛力時我喜歡的地方
22:00
是的，設計自我複制機器人的潛力是什麼？
22:05
設計比你越來越好你甚至不理解人類的設計是的，是的，我的意思是
22:12
當然，這會帶來一些潛在的危險和所有這些東西，但就純粹而言，你知道經濟上的潛力
22:19
我們談論的是世界歷史上從未發生過的事情，你甚至可以比較一下
22:25
幾乎你知道發生的任何事情，這太瘋狂了，嗯，我在想這個，嗯，也許
22:31
你會對此有一些想法，但我在想，過去一周人類有這個
22:37
繁殖的限制，不僅繁殖而且照顧你的後代都非常困難
22:44
有很多時間和精力，比如健康和安全，所有這些
22:49
問題嗯，你不能只複製你知道然後擁有的一千個後代
22:55
他們每人做一千次，就像那不是人類的工作方式，嗯，是的，但這種限制並不
23:01
適用於您認識的機器人，因為您不必為它們打掃衛生您不必知道母乳喂養牠們
23:08
你必須做任何他們就在那裡的事情，隨著大腦變得越來越聰明和更好，
23:13
無論如何，正如他們所知道的那樣，他們可以創建自己的機器人，然後您就擁有了這種瘋狂的無限複製
23:21
情況真是令人難以置信，是的，我不確定你是否提到過
23:27
但我肯定已經得出結論，特斯拉人形機器人是最終產品
23:33
是的，因為正如你提到的，能夠創造出更好的自己，而不需要接受培訓，而不是
23:39
嬰兒剛開箱即用的能力是的，只是得到的含義是
23:45
真的很荒謬是的，是的，我稱它為結束所有產品的產品，就像結束一樣，是的，結束了
23:52
真的是 嗯 有趣的也像 Elon 選擇不像狗一樣生產或
23:59
一些其他類型的機器人，他正在尋找人類機器人，比如類人機器人
24:06
嗯，我認為部分原因是他的專注，他就像你知道我們有機會
24:13
在這一點上，我們會去做的，他主要了解大腦，嗯，另一部分是我認為他
24:19
一旦他們把它拿出來就明白了，他們可以領先並得到數以百萬計的東西這些東西你怎麼能喜歡
24:26
誰能趕上你知道這就像誰是第一個以及為什麼在比賽即將開始時從狗開始
24:33
通過一個人形機器人出來，你知道這就像讓我們玩這個就像用真實的東西把它拿出來給數百萬人然後就像我會說
24:40
比賽結束了，但就像是的，你的領先優勢越來越大，似乎是的，我想每個人都說了些什麼
24:47
AI 第 2 天演示文稿，介紹實現有用的最快路徑
24:53
不管那是什麼機器人，所以他有這樣的想法，為什麼要從最終產品以外的任何東西開始，讓我們開始吧
24:59
盡快有用的東西，然後我們可以迭代和改進，是的，極度緊迫感
25:05
我真的從 AI 第 2 天得到。看起來 Elon 還不錯，這就是
25:11
絕對優先，我們必須盡快完成這項工作，他感覺到這個先發優勢有一個
25:17
如果您獲得足夠的優勢，那麼它是無懈可擊的，您只是不會被抓住
25:22
嗯，你知道的，這就像一隻蛞蝓試圖與你認識的人競爭，一旦你達到一定的能力水平
25:28
為什麼還要打擾是的，是的，我完全明白那裡的人工智能也是埃隆沒有明確地看到它，但它是
25:36
暗示人形機器人是一個不同層次的機器人
25:42
比你知道的其他任何事情都明顯，所有這些東西都只是準備，是的，你知道踏腳石可以到達這個 um 到機器人
25:49
有趣的是，埃隆在這方面的想法是多麼清晰，你知道
25:55
嗯，對我來說，這很明顯，很清楚，但只是大多數首席執行官不會這麼清楚地思考它
26:02
並做出下一個合乎邏輯的步驟，即押注整個公司，或者您知道自己的一切，這似乎
26:09
就像他看得很清楚，他按照邏輯做了他需要做的事情，嗯，這很有意義，但這是
26:16
像這樣不正常，好像你知道這個行業或者科技世界嗯
26:21
是的，時間線怎麼樣？你對特斯拉的期望是什麼？
26:27
開始在工廠中以一種有用的方式使用它，但不僅僅是鋪路石是什麼樣的
26:34
下一步是你認為它只是工廠然後你知道賣給企業還是你
26:39
認為他們會用它來做最後一腳的送貨火車，你知道機器人在回家之前會這樣做，或者
26:46
是的，我的意思是你認為像埃隆說的那樣三到五年是一個現實的時間表，可以像什麼
26:52
就時間線而言，您對時間線的想法是否很好？實際上，我的測試評估模型是第一個
26:58
特斯拉機器人在三年內最好的情況下被出售給客戶，我預計不會
27:05
大約五年的有意義的交易量，這是一個相當樂觀的估計，因此時間表非常接近我
27:11
還期望硬件大約 25 000 所以 Elon 和我有點猜測
27:17
大約在同一時間，但當然有很多未知的未知數，很難確定何時
27:22
這將是商業上可行的
27:29
勞動，但我認為如果埃隆和特斯拉的團隊能夠確定他們可以快速發揮作用的任何其他領域
27:36
也許生活方式的傳遞實際上並不像你想像的那麼難，他們實際上可能會追求這些機會
27:41
好吧，我認為特斯拉可能會想要，一旦他們在自己的工廠裡有了有用的人形機器人，他們真的會不顧一切地得到它們
27:48
進入其他應用程序，不僅是其他工廠，而且下一件事在哪裡，因為一旦你在工廠做
27:53
有用的勞動力 你可以變得越來越好 你不需要成為現在 你可以開始關注我們還能添加什麼新功能
28:00
目前根本無法做到的事情，並且在那些方面非常激進，所以我不確定我猜
28:06
可能在這個十年後期，最初會有一些銷售給其他商業客戶，我認為這真的是
28:11
2030 年代及以後，我們開始看到特斯拉機器人被用於更廣泛的應用和
28:17
變得越來越有能力找到進入家庭的方式可能是 2040 年代事情變得超級的地方
28:24
瘋狂的地方有機器人可以在家裡照顧老人無處不在你是什麼意思所以如果我能得到一個
28:31
特斯拉咬了我的房子，讓我們說五年它能夠做什麼然後你認為不是很多我認為它能夠做到
28:38
一些有用的東西是的，我認為不要誤會我的意思，因為它會很有用，但我不希望在
28:45
超級好，就像讓你的孩子在家裡幫忙一樣，他們可能很有用，但他們
28:51
也不是很有幫助 我認為如果特斯拉可以完成一些基本任務那就太好了，但我的期望
28:57
也許我對此有點悲觀，我認為特斯拉車身不會有重大用途
29:03
在最好的情況下，可能要到本世紀末期才回家
29:09
如果您只想要一個可以倒垃圾的機器人，它可能對某些人沒有用，您也許可以擁有一點
29:15
早一點，但是你希望家裡的機器人能夠完成的不同任務的數量
29:21
從消費者的角度來看，在您真正認為它有用之前，您知道洗衣裝卸的事情範圍很廣
29:28
折疊衣服這類事情我認為特斯拉在這方面真正有用還需要幾年的時間，因為有一個
29:35
有時候，如果你有一個機器人，它不能正確折疊你的衣服，就會弄得一團糟，
29:40
它正在卸載洗碗機，但它把叉子扔得滿地都是我認為特斯拉的那種水平還需要幾年時間
29:46
尤其是在操縱的時候，我認為人們不理解人類的手是多麼不可思議的敏感方式
29:53
我們可以撿起一個物體，只要看看就知道如何牢牢抓住和塞進特斯拉沒有的東西
29:59
真的有任何進展，但就像現在還很早，他們只是在弄清楚那種
30:05
你知道非常基本的動作，我認為這需要一段時間才能解決這些問題，但機器人越多
30:11
飛輪部署得越多，我認為這些能力會增長得非常快，並且會讓人們感到驚訝，我認為它會
30:17
最初幾年讓人們失望，但後來不知從何而來，就像我們看到人工智能的突破一樣，你知道
30:22
圖像分類，然後使用 chesco 這類東西，功能還不錯，然後就不行了
30:27
無處你只是有這個垂直的時刻，它幾乎變得超人你對時間的想法是什麼我很想知道是的你知道的
30:34
實際上，我對其中的一些產生了共鳴，因為一些家庭任務
30:42
超級難，就像做飯一樣，你知道切東西，就像是的，只是和
30:48
有用才能在家裡真正有用你需要做的那種事情你知道你不能只是用吸塵器吸塵
30:54
可以有一個 Roomba，或者你知道，嗯，所以高價值的東西是超級難的
30:59
我的問號，我正在研究一些新的人工智能
31:07
佔用網絡，他們稱它們為 Nerfs 神經輻射場，但它們基本上是在渲染
31:16
3D 空間現在超級高效，我想知道未來三四個
31:23
幾年後，這個領域基本上會有巨大的進步
31:30
3D 你知道建模渲染渲染你知道控制環境中的所有東西到什麼程度
31:38
我們現在擁有的就像孩子們的地方一樣，幾乎沒有你所知道的那種類型
31:43
突破將使機器人在其
31:50
在像現在這樣的環境中四處移動的能力很難想像機器人
31:56
超人和殘疾四處走動，因為我們只是沒有什麼可以將其與正確的比較，但是如果您跟踪進度和
32:04
AI 進步的速度有多快，這個領域是如何突破的
32:09
經歷了很多事情，尤其是現實世界的人工智能，只是會有一段時間我不知道
32:15
當那是四三年或四年或五年或六年時，似乎不會需要更長的時間
32:21
雖然比 Phi 嗯，但我認為這可能是突破之一，可能還有另一個
32:26
需要突破才能真正讓你知道以超人靈巧的方式發揮作用才能成為超級你知道的
32:34
有幫助，但似乎有一個我們離得不遠
32:39
嗯，但我大腦的一部分，你知道它仍然像現在比較，所以我想等一下
32:45
看起來不太現實，嗯，但我認為接下來你知道五年或十年，這將是一個迷幻
32:51
就像每年我們都在談論這裡發生的事情一樣，是的，我現在有這些時刻
32:57
像 Dali 和 Dali 這樣的 AI 也取得了進展，是的，我的腦子壞了，我覺得我們已經從 AI 中走了出來，基本上有點酷
33:04
下一分鐘你就會知道的語言內容 這是一個提示，你創造了一些獨特而新穎的藝術，我沒想到
33:10
你將能夠在 2022 年輸入一些文本提示並讓 AI 渲染一個
33:16
一堆場景和場景我只是沒想到會發生，所以你絕對正確
33:21
誰確切知道時間的突破，但有一個能力點，它無處不在
33:26
只是從體面到超人，我認為當它發生時會讓很多人感到驚訝我只是不知道確切的時機是什麼樣的
33:33
是的，是的，你期待什麼，比如明年在特斯拉援助 2023 上說什麼樣的樂觀主義者可以機器人
33:40
你會期望進入演示這是一個很好的問題嗎？我認為這些功能將是巨大的
33:46
在移動和操縱物體的能力方面有所改進，所以我們可能會看到一些更精細的演示
33:53
電機控制 第一個版本看起來很棒，但顯然需要一些工作，但我認為
33:59
我們將看到更好的能力在環境中移動以操縱對象，也許還有其他一些
34:06
看起來很聰明或可能會讓一個人感到驚訝的行為示例
34:11
很少有人會在這個時間點操縱需要瘋狂數量的東西
34:16
靈巧我認為這將是一個很大的說明是特斯拉在那裡取得了多少進步我認為大腦本身將繼續變得更好我們已經
34:23
看到了 FSD 的進展，我們知道那裡發生了什麼，但它實際上是在世界上移動並操縱我認為是的物體
34:29
將成為一個大焦點，因為我們在 2022 年人工智能日看到的大多數人都是
34:34
指著點笑著他們並不真正理解這個項目有多快完成，但我認為這將是我們看到的巨大改進
34:40
至少我們可以注意到，是的，你怎麼想，你的想法是什麼，是的，我認為
34:46
對於那些看到潛力的人來說，接下來的幾年將是令人興奮的，但我會說有點慢，但喜歡
34:54
我不期待明年會有一個瘋狂的機器人你知道它是的，只要
34:59
他們是的，他們解決了一些主要問題，比如運動，他們實際上可以有一個
35:04
體面地在機器人周圍移動 嗯，機器人可以撿起你知道的東西，自信地把它放在地方
35:12
嗯，做一些基本的任務，實際上他們在做自己的演員
35:17
他們自己的 你知道你自己的 這就是所有這些東西都在起作用 然後我認為這是一個巨大的成功
35:24
明年的下一個構建塊 有權對其進行迭代，但是是的，這就像我認為人工智能的角色
35:31
將成為像這條瘋狂的 S 曲線這樣的完美例子，你知道我們看到了一些進展，而那些
35:39
看到未來會對此感到興奮，但你認識的其他人就像未來幾年的每個 AI 日一樣
35:45
就像 嗯 我們會 嗯 嗯 測試隊列觸發器 是的 II 100 同意 我
35:53
認為是的，如果你知道進展，如果你真的知道發生了什麼，你就會明白你會被吹走
35:58
但我認為，未來幾天的 AI 日子裡，公眾的看法是，我真的希望
36:04
你花了整整一年，這就是它所能做的，嗯，但是我想如果你理解一些潛在的困難
36:10
問題 我認為特斯拉將取得令人難以置信的進步 嗯 是的 很有趣
36:17
現在可能在嘲笑特斯拉擎天柱的人誰知道 10 年後
36:22
他們可能會，也許樂觀主義者會給他們做午餐，你知道，是的，這是對的，或者和他們一起吃
36:29
床，嗯，你說我認為它在其中之一
36:34
你的視頻或者一條推文，你說 Facebook 和 Google 嗯，如果他們搬家了，如果他們有
36:41
人工智能機器人的先發優勢對你來說就像一場噩夢
36:47
背後的推理，呃，我不認為他們是最終會發生什麼的有能力的保管人
36:53
成為 AGI 嗯，我認為無論你在哪裡，我認為兩家公司都有可怕的危險信號
37:00
不完全貶低他們 我的意思是谷歌製造了一些很棒的產品 Facebook 也製造了一些很酷的東西但是
37:06
Facebook 我的意思是我什至不應該真的需要解釋為什麼我們不希望他們開發人工智能我的意思是
37:11
錯誤的數量對馬克扎克伯格來說並沒有冒犯，但他不是我想要的那種CEO
37:18
那種技術，即使他的意思很好，但在我看來，當時的記錄並不那麼好
37:24
谷歌也被 Facebook 徹底感染了工作思維病毒，我知道這樣說很有爭議
37:30
但我認為這些公司的大多數人都沒有清楚地思考我認為他們知道什麼對世界絕對是最好的
37:36
什麼是真、道德、是非的仲裁者，我不想要人
37:41
有這種感覺的人確切地知道對整個世界來說什麼是正確的
37:47
可能你知道擾亂我們所知道的生活，那裡有很多我認為是善意的場景
37:52
那些公司裡好心的人可能會把事情引向一個非常消極的方向，想像一下，如果有
38:00
你知道谷歌控制下的類人機器人，你知道幾年前下一刻的瘟疫
38:06
谷歌就像是的，我們可以幫助發送這些機器人讓人們留在家裡你知道我只是有很多這樣的場景我知道埃隆會
38:12
就像得到 effed 我不是不，那是錯的，是的，所以我只是覺得像 Elon
38:18
馬斯克的心在正確的地方，他有良好的意圖，他做出了正確的決定，並考慮了盡可能多的
38:24
可能的影響，因為他可以合理地考慮，我已經看到很多公司喜歡的例子
38:30
谷歌 Facebook 沒有這樣做，是的，我不認為他們意味著傷害我只是認為有時你可以成為一個
38:37
善意的，或者你可以善意並做非常糟糕的危險事情所以是的，我最糟糕的情況是任何公司
38:45
開發有任何危險信號的 AGI 聽起來像 Elon Fanboy 出來
38:52
但在這個星球上的所有人中，我認為埃隆是我們正在開發的東西的保管人
38:57
AGI 在這裡 是的 我曾經 我曾經認為對 AI 沒有害處 一切都會好起來的 我曾經長期狂熱
39:03
15 年前和我父親的辯論 10 我開始更多地閱讀它並完全改變了我的立場
39:08
就像我會沒事的，就像一些人工智能人今天仍然說等一下，有一百萬種方法可以
39:14
出錯了，我現在非常清楚，即使他們是善意的人，事情也會發生在錯誤的手中
39:21
在某種程度上我們真的不想
39:27
它創造了很多這樣的東西，我認為大多數人可能不會在我認為
39:34
接觸，我可能錯了，但我的意見是典型的路線
39:39
人工智能進入 HEI 所有這些東西 典型的路線不是
39:46
非常好的路線，就像您在談論一家由典型官僚機構經營的典型公司，由所有這些東西經營，例如
39:53
與政府勾結，我們的隱私很少，權利很少，你有一個場景
40:00
只是在惡化，但這將是你所知道的正常路徑
40:06
技術，我認為我們這裡有一條不尋常的道路，不知何故，一個人
40:13
誰是善意並且實際上喜歡他的大腦在正確的方向
40:19
看清楚事情已經到了可以真正影響未來方向的位置
40:27
從某種意義上說，這個世界是如此的不正常，對世界如此樂觀
40:35
未來是對的，因為否則你會看到一個非常嚴峻的情況，而我
40:40
不要認為人們理解你所知道的兩條路徑以及你知道的多麼不正常的路徑不太可能這條路徑
40:47
真的是直到最近，我們在這裡，我們正處於這個風口浪尖
40:52
可能會發生，這就像我認為被低估的情況之一
40:58
錯過了整個十年的故事，嗯，是的，雖然是的，但我很迷人
41:05
同意，最後一點，我認為這是 Elon 以這樣的方式鎖定這一點的部分原因
41:11
緊迫感，因為他知道他是那個人，是的，正如你所說，他是異類
41:17
謝天謝地，這不是 zako 或那樣的人，這一切都是優先事項
41:23
因為我認為埃隆現在正在與這種情況的風險競爭，在這種情況下，這最終會落入壞人之手
41:28
即使他們是善意的，是的，所以我認為這增加了他對這需要多快的感覺
41:34
發生以及特斯拉需要如何致力於實現這一點
41:39
嗯，如果我們回顧大流行，我的意思是我認為至少對我來說是
41:45
嗯，其中一些公司的真實面目是在政府審查的封閉下出現的，哦，是的，所有的
41:51
這些東西，看到它就像是驚人的，是的，我知道我知道你的眼睛是睜著的，即使你不認識你
41:58
不要繼續你不想知道的咆哮會影響你的整個聽眾，但我注意到你的眼睛是睜開的
42:04
這是一個非常令人不安的時刻，真的是我剛去了，你是認真的嗎，這真的發生了嗎？
42:11
現在，對我來說最糟糕的是人們沒有註意到是的，就像你們是你們一樣
42:17
真的沒有看到現在發生了什麼，這根本不會打擾你所以是的，我的意思是他們從來沒有
42:23
同時消失了，是的，你知道的狂妄自大，比如其中一些，而不僅僅是一些最大的技術
42:30
公司也是最大的你知道你有製藥政府都聚集在一起呃口授你知道什麼
42:39
它只是它只是啊它只是是的，想像他們會是
42:46
HEI的管家你知道會發生什麼是的那是你知道你是一場噩夢我猜
42:52
嗯，場景是的，完全正確，是的，嗯，我會說原因之一
42:58
為什麼你可能和我自己都被你如此興奮和
43:04
嗯，你相信你知道特斯拉在埃隆領導下人工智能機器人的潛力只是
43:10
因為這是為數不多的看起來可能不錯的場景之一
43:15
嗯，是的，是的，這是一個非常重要的情況，我們發現自己處於是的
43:22
嗯，如果 Elon 發生什麼事會發生什麼，你認為特斯拉將來可能會像管家一樣嗎？
43:30
在道德上道德上和以一種好的方式，或者你認為你只會惡化
43:36
我認為特斯拉確實擁有正確的 DNA 埃隆是徹底透明的超級
43:42
老實說，在特斯拉工作的每個人都知道這傢伙在做什麼，我認為特斯拉的文化是獨一無二的
43:48
地球上的公司至少在規模上要了解一家公司我認為這是一個好兆頭當然誰知道
43:55
隨著時間的推移，如果 Elon 明天消失會發生什麼，但我願意打賭這可能是 50 50 的機會
44:02
僅僅因為特斯拉作為一家公司的 DNA，事情繼續朝著正確的方向發展，我認為絕大多數工程師
44:08
在特斯拉工作的人真的想對世界產生積極的影響他們可以在其他地方工作獲得更多的錢這會讓你失去一份他們真正關心的工作
44:14
關於他們在做什麼，我認為公司的所有文化都是自上而下的，所以我
44:21
認為特斯拉文化的建立方式可能是一個非常好的跡象，即使埃隆做到了
44:26
明天消失 Tes 將繼續做大部分正確的事情，從一個好地方開始，你又知道什麼 20
44:32
多年前，不作惡是谷歌的座右銘，我們現在發現自己處於一種境地
44:37
我知道我可以通過只說一個這樣的句子從 YouTube 上刪除這個採訪
44:44
永不言敗，但我相當希望特斯拉現在確實擁有正確的 DNA，即使埃隆要
44:49
消失或後退或發生某些事情我認為他們至少有 50 次機會做正確的事
44:55
從長遠來看，我也有一種希望，他們已經足夠遠了
45:02
自主性和他們仍然應該首先到達那裡的人形機器人顯然會更多
45:09
具有挑戰性，所以也許我很天真，有點樂觀，但我現在很有希望，幾年前我真的輸了
45:16
像睡了很多覺一樣思考事情可能會去哪裡所以你認為你認為特斯拉能夠做到什麼他們有正確的 DNA 或
45:22
是的，我的意思是我認為隨著時間的推移，我可能不那麼樂觀了
45:28
嗯，我認為只要負責一家大公司，你就有錢，就像你知道的那樣，季度報告你
45:35
有壓力儲備所有這些東西需要一套獨特的價值觀和世界
45:40
考慮到一種反擊，所有的壓力和
45:47
是的，我想你知道我們已經習慣看到 Elon 反擊，他說，嘿，我是
45:54
不要試圖增加股票我不在乎你知道什麼，很多人可能不相信他，但結束了
46:00
時間他的信息是一致的，他的注意力也集中在他正在做的事情上，就像他不需要做他正在做的事情一樣
46:06
真的，所有的時間和痛苦，你知道這就像瘋了一樣
46:12
嗯，但是是的，那種類型的人就像是一個艱難的人
46:18
嗯，我們正在談論幾十年後的人類未來
46:23
前進，所以可以肯定地說，未來 10 年或 20 年的情況都很好，但是
46:29
那之後是什麼，嗯，這絕對是我個人關心的事情，我認為還有什麼
46:36
有趣的是我從未想過投資一家公司
46:42
不僅會成為一項投資，而且就像它與未來的未來息息相關
46:48
從某種意義上說，我更關心世界的未來，而不是我對所在公司的投資
46:57
更重要的是你知道公司的使命，有趣的是公司的使命
47:02
公司在將世界轉變為可持續能源之前，這本身就是一項巨大的使命
47:07
結果，但是當你添加這個時，整個 HEI AI 機器人就像這個任務這就是這就是我
47:15
不會說沒有任務，但是是的，這幾乎是真的，真的是任務的結束，就像
47:22
是的，是的，當你這樣想的時候，我不認為它就像男人一樣
47:28
有一家公司，你顯然知道這是必然的，嗯，我們在這些瘋狂的瘋狂
47:35
你知道的歷史時刻，很多事情都會發生，我們
47:40
不知道結果會怎樣
47:50
嗯所以嗯嗯嗯我很好奇如果你
47:55
嗯，我想听聽你對股市特斯拉股票的看法，所以我們
48:01
週一晚上錄製，今天特斯拉股價下跌了 8% 以上
48:07
交貨或交貨數量不佳，是的，至少這並不意味著
48:14
AI 確實令人失望，是的，是的，所以你知道這些年來你一直很善良
48:20
你們中的一些人知道在你喜歡的地方採取這種反直覺的方法
48:26
從長遠的角度來看，你知道逢低是機會你只是買買買買哇什麼是什麼
48:32
你目前對特斯拉股票的想法是你仍然知道用你能來的任何錢購買嗎
48:38
嗯，你對此有何看法，是的，所以我確實有一些信件在這裡
48:44
從稅務局那裡我不會詳細說明，但我優先考慮特斯拉股票
48:50
嗯，就購買而言，此時此刻，我總是著眼於長期的未來，而在這一點上
48:55
正如我在我的頻道上說過一百萬次的時間，我個人不知道有更好的風險調整機會
49:00
股市 這並不是說它們不存在 我只是不知道任何東西 我正在展望未來 看看正在發生的事情 看看
49:07
最壞情況下的潛在因素 我認為可能的最好情況
49:13
沒有什麼可以比較的 所以如果我有一些錢可以投資
49:19
在這個時間點，一切都在進入特斯拉，我預計在未來十年或更長時間內，可能會達到 10 倍
49:25
在閾值附近的某個地方給或帶你知道幾個 X，但是如果我們在另一個上加上另一個十年
49:32
如果你在 2016 年問我，我認為我的特斯拉股票投資在接下來的十年裡會做什麼
49:37
30 年後，今天將是一個非常不同的答案特斯拉繼續想出他們只是將 S 曲線疊加到
49:43
一條 S 曲線到一條 S 曲線上，我正在私下為特斯拉建模，直到 2069 年
49:49
顯然它變得非常不確定，但我必須至少做一下練習，如果這些機器人確實擴大了規模
49:56
一群機器人得到了它是怎麼回事，它變得絕對淫穢，所以我不想看
50:02
回到 2040 年 2050 年或 2060 年，思考的人，我真希望我知道你在特斯拉上投資了一點點
50:08
你知道我個人並不需要資金，但我確實打算在未來做一些資本密集型的慈善事業，所以我
50:15
試圖盡可能多地滾雪球，我認為這給了我更多的風險承受能力，因為
50:20
好吧，因為我不是你知道，如果我失去它，好吧，不管它不會改變我的生活方式，嗯，但是是的，我就像我說的那樣
50:26
我所知道的股票市場上最好的風險調整機會，在股票動蕩的時間點上說這一點毫無爭議
50:32
有便宜貨左右和中心你知道是的你現在可以採取分散方法然後砰這個抓住這個抓住這個抓住這個你會
50:38
五年後可能會很開心，嗯，但特斯拉只是在下一個之後發生了一系列事件
50:45
接下來它只是在打擊股價，我個人認為該股票是人為的
50:50
壓制我認為它有點像彈簧一樣，在某些時候你知道宏觀經濟環境何時平靜
50:56
稍微下降一點，恐慌和恐懼就會減少一點，人們開始關注特斯拉的收益，關注他們的增長，然後等一下什麼是
51:03
你說真的，在這個時間點，這就像特斯拉一樣便宜，我覺得這是在廉價地下室領域，當然
51:09
它仍然可能會下降很多
51:15
真的是我只是我可能積累的每一分錢都直接進入，我是你知道我什至
51:20
考慮到大 M 字邊距，是的
51:27
嗯，不，是的，所以考慮到這也是一種選擇，但有趣的是，我並不著急，是的，你知道特斯拉
51:35
股票太有趣了，因為你知道，當你開始對擎天柱進行任何評估時，如果有的話
51:42
否定的，嗯，是的，我同意，你知道在接下來的五年或十年內，他們將主要是你知道汽車和 FSD
51:51
嗯，但是當你看到更大的圖景時，讓我們說 10 年和
51:58
特斯拉用人工智能機器人設置的東西幾乎就像接下來的五個或
52:04
十年就像股票價格一樣無關緊要，或者至少如果你有足夠長的視角你知道什麼
52:09
我的意思是，是的，如果未來五年或十年的股價走勢真的很重要
52:15
特斯拉實際上是我的想法
52:22
我知道你可能提前知道它會回來，但我有一個長期的我不在乎我是否支付 1200 預拆分那是什麼 400
52:28
現在你知道 200 500 800 我只是不在乎，因為我不考慮五年 十年 我在想 10 20 30 年
52:35
未來，你是對的 我不相信華爾街股市總體上會給任何
52:41
完全相信他們只是他們完全忽略了特斯拉機器人和 AGI 的可能性他們只是他們不存在他們沒有看遠
52:47
足夠的未來，如果這真的發生了怎麼辦，所以我認為特斯拉將成為
52:54
一家公司在很長一段時間內，在股市認為特斯拉所處的位置與他們的未來看起來之間存在巨大差距
53:00
喜歡和現實大規模的大規模脫節，我不認為特斯拉機器人會發生什麼
53:07
最終成為 AGI 市場將真正開始獲得它至少五年或更長時間
53:12
因為他們會說，嘿，我知道我們會填補它不是很你知道這看起來不會真的有很大的進步，只有我
53:19
感覺就像華爾街和大多數普通散戶投資者，即使他們需要看到今天發生的事情
53:25
明白這將在明天發生，他們無法對未來看得足夠遠，無法衡量發生這種情況的概率，並且
53:32
想想這是否可能，如果不能，其他人可以抓住他們等等等等等等，我明白為什麼，因為什麼
53:38
特斯拉要做這件事需要你需要相當廣泛的知識基礎和許多真正書呆子的話題才能做到
53:44
理解發生了什麼你看到一個機器人，你也只是人工智能日的普通大眾
53:50
波士頓動力公司已經做到了，你知道我不明白，但如果你有額外的上下文層和
53:56
理解你等一下他們在這裡用大腦做什麼只是在另一個層面上，所以我
54:01
認為市場不理解只是很長一段時間，我同意這就是為什麼我只有當我有錢時才會購買特斯拉股票我不想
54:07
時間到了，我知道我可能會坐在這裡等待一場大崩潰並加載，但這就像我寧願在我有現金的時候買，我真的不在乎
54:14
如果特斯拉的市值達到 5 萬億美元，我今天仍然會購買，你知道價格是多少
54:20
而且我仍然對前景感到非常興奮，但現在比現在要少得多，所以回報更好是的
54:26
有趣的是，你對加密有什麼最新的想法嗎？我知道你知道你在
54:31
過去看到了加密貨幣的潛力，但你在投資方面有點保留，有什麼地方可以
54:38
加密貨幣下來，你會說，嗯，有趣，嗯，你知道我會的
54:43
是一項很好的投資我認為我們正在接近那些點，但我不可能投資
54:49
在加密貨幣中，當特斯拉仍然在這個價格時，現在一些額外的背景一些觀眾可能不知道我做了
54:54
實際上之前在 2013 年和 14 年投資過比特幣。我分配了大約 15
54:59
淨資產我說這是高風險高回報我可以看到這東西可能會變成零或十萬美元
55:06
我讓你在不同的交易所知道比特幣 長話短說我被騙了 我被黑了，最後我輸了
55:12
幾乎所有的比特幣，這對我來說遠遠超過一百個比特幣
55:17
從心理上考慮重新進入一個價格點而不是 200 300 比特幣或類似的東西，我只是
55:24
鑑於未來的不確定性，看不到那裡的回報 我認為加密貨幣將繼續存在 比特幣可能就在這裡
55:30
也可以留下來，但現在還為時過早，如此不確定，而且資產類別如此波動，我沒有足夠的信心
55:37
說好吧，這里肯定有一些東西我願意分配資本，即使我認為潛在的上行空間可能是巨大的
55:43
這對我來說仍然非常投機，所以仍然每天都購買特斯拉股票
55:48
備用美分，但我很想知道發生了什麼我一直在關注 nft 市場出於興趣我有點我有一個假設
55:54
看起來大部分時間將下降 95 100 幾乎從哪裡開始
56:00
我們是今天，我什麼也沒說，但我總是檢驗我的假設與現實，我認為我們是加密貨幣，這個
56:07
整個太空將經歷一個系列，你剛到月球，它就會一次又一次地回到現實，我不知道還有多少次
56:13
這會奏效，但我認為從長遠來看，它會一直存在，但很難選出贏家，我做不到
56:19
任何具有高度信心的陳述是的是的，當你比較投資機會時，我同意
56:24
對特斯拉來說，這是不同的，你知道我認為有更多的不確定性
56:31
嗯，我的意思是比特幣 Maximus 會告訴你有 100 個確定性是對的
56:36
在這裡，但向邁克爾大喊一聲，是的，是的
56:41
嗯，這是投資建議
56:48
與特斯拉不同的水平嗯是的雖然嗯嗯我知道我正在閱讀最近的
56:54
你關於火山口經濟的話題，你似乎非常看好那種
57:00
內容創作者的未來在某種程度上接管了更多的市場，因為他們
57:05
就像你知道的發布產品已經只賣給他們的觀眾
57:12
嗯，你覺得這在哪裡？
57:17
這以造物主經濟為首 就像它會討厭一切 我想我真的 它會吃掉一切
57:25
我的意思是我真的真的相信那個我曾經做過的野獸先生的一個很好的例子
57:30
開始在 YouTube 上觀看很多內容創作者不是因為我真的想觀看內容，而是我想看看他們在做什麼並了解如何
57:36
他們正在做他們正在做的事情 野獸先生是一個很好的例子 我敢肯定你聽說過他 是的 最大的 YouTuber 基本上是
57:44
對渠道進行再投資，沒有人能真正與之競爭，所以 Mr Beast 是該頻道推出的品牌 Beast burger 和
57:51
一夜之間，他們開了一堆像鬼一樣的餐館，他們只是用別人的廚房做飯
57:56
在 Beast Burger 的第二年裡，訂單和不知從何而來的同時開了幾百家店
58:02
賺取超過 1 億美元的收入，並且每次 Beast 先生髮布
58:08
新視頻他有 50 多萬觀眾，每個視頻都有 10 個超級碗廣告
58:15
嘿，我們有一個新產品，看看吧，如果你是粉絲，我已經要買漢堡了，我為什麼要去
58:20
麥當勞當我可以買一個野獸漢堡而不是我認為人們不理解這一點
58:26
成本實際上是零 客戶獲取幾乎沒有你已經在這裡得到了信任，我沒有
58:31
為傳統品牌尋找一種競爭方式，除非他們自己在創造內容，因為每個人
58:37
看這個就會明白有 YouTube 是他們真的很喜歡它 還有其他的內容創作者他們真的很喜歡看 如果這個人
58:43
要放棄他們已經購買的產品，並且在質量和價格方面具有可比性，您願意大公司嗎？
58:49
faceless Corporation 獲得收入，或者您更願意幫助您已經喜歡觀看其內容的人
58:54
呃，我看不出傳統品牌將如何與之競爭，除非他們創建自己的內容，你知道，也許可口可樂有自己的播客
59:01
並且有一個巨大的東西，人們喜歡可樂播客，他們仍然會忠於這個品牌，但如果不是
59:06
一個接一個地，我們開始看到各種奇怪的融合，YouTube 明星們正在做正確的拳擊，
59:13
製作的盒子比大多數專業盒子在他們的職業生涯中所做的還要多
59:19
30 4000 萬美元，呃，我認為人們嚴重低估瞭如何
59:25
我想很大的力量可能是創作者可以擁有的影響這個詞，我只是看不到傳統品牌
59:31
如果您是 YouTuber，則能夠與您知道的某些特定事物競爭
59:36
你談論健康和營養每個人都想知道你吃什麼下一分鐘你釋放什麼補充劑
59:41
你自己的補充 它實際上是好的人們會購買它 如果你有 1000 萬訂閱者並且你是一個
59:47
YouTube 上的全科醫生突然間你不再需要作為全科醫生工作你實際上可以建立一個巨大的補充
59:52
公司我知道大衛人類或呃安德魯休伯曼或大衛辛克萊是
59:58
開始發布產品，他們幾乎在一夜之間就會擁有 1 億美元的公司，因為他們擁有廣泛的
1:00:03
觀眾 他們對健康很感興趣 每個人每次我提到一些我用來作為能量的東西 每個人都在求我你是什麼產品
1:00:09
使用您使用的品牌，您知道那裡的潛力只是天文數字
1:00:14
我不認為人們會明白你的想法是什麼 你的遊戲計劃是什麼
1:00:19
嗯，我知道金融投資特斯拉是一個領域，一個利基，你快樂嗎
1:00:26
與那種利基呆在那裡，做你的事，或者你有沒有被誘惑到那種你知道的去其他地方
1:00:32
可能會更廣泛和更大的領域是的，未來我肯定會擴大很多不同的領域
1:00:39
關於我感興趣的話題的事情我想討論我的限制因素直到公平
1:00:45
自從我們上次談話可能是兩年前以來，最近一直精力充沛，我的精力增加了一倍多
1:00:50
衡量這是我的工作能力和恢復能力，因此我現在每天可以做的事情是兩年前的兩倍
1:00:57
而且我現在確實有足夠的時間和精力來開始添加和下降
1:01:02
其他兔子洞我全神貫注於這個特斯拉內容，我將在幾十年內報導這個不會去任何地方，但我正在嘗試排序
1:01:09
砍掉我的生活，如果我能從一天中提取更多的能量，提高生產力，做更多的事情，我就走了
1:01:16
隨著時間的推移，不斷地不斷地添加東西，有很多我非常熱衷的領域
1:01:21
有點知識淵博，但還沒有接近專家水平，但在接下來的幾年裡，你可能會看到一些新的頻道突然出現
1:01:27
他們中的一些人可能無法出現在 YouTube 上，但其他人會出現，是的，我想我只是在想
1:01:33
我不會對盡可能多的人產生積極影響，而且你知道我對特斯拉有一些了解
1:01:38
這對某些人可能有用，但是我在 patreon 上傳達了很多其他的東西，但我並沒有真正開始談論
1:01:43
關於在 YouTube 上，我知道大多數觀眾不在乎，所以如果我開始發表太多離題的咆哮，我會趕走人們，但我想
1:01:50
將來分享這些東西，嗯，是的，肯定還有很多東西要來，我們才剛剛開始，是的
1:01:56
就像是的，我看過野獸先生他過去幾年的採訪我看過很多他的和
1:02:05
從某種意義上說，這讓我很困擾，因為他說的就像你知道他抓住的機會一樣
1:02:13
這麼大而且非常有意義的東西
1:02:19
嗯，這讓我有時會質疑我在做什麼，就像我正在做的那樣
1:02:25
不負責任的喜歡我應該喜歡你知道基本上為你拍攝星星
1:02:31
知道，我也和這個作鬥爭，是的，但我有同樣的想法，是的，因為
1:02:36
我覺得他肯定在利用一些你知道的內容
1:02:43
不同規模和水平的創作者，但你知道
1:02:50
II，我很舒服，我喜歡我所處的位置，感覺很好，我不喜歡
1:02:56
必須強調我沒有員工或公司，我只是喜歡冷靜，你知道想出內容，這很好
1:03:03
對我來說，但我的腦後確實有這個問號
1:03:08
我喜歡我會說錯過但不負責任，比如不去追求更大的東西和挖掘
1:03:15
進入別的東西，但嗯，是的，它是它買的東西，它就像一個常數
1:03:22
像你知道的那樣煩擾就像我腦後的東西總是它只是是的所以無論如何我會說話
1:03:28
關於內容
1:03:35
嗯，是的，我肯定知道有一個，因為我是一樣的，我想你意識到有一些未開發的潛力
1:03:41
你所擁有的，而且你在專注於一個非常小的狹窄利基和我與之抗爭的事情上非常有限
1:03:48
最重要的是我不想對任何事情不誠實，我什麼也做不了，只能100%做我自己
1:03:54
而且我知道有很多事情不利於限制增長的最佳實踐，如果我想開始
1:03:59
未來更通用的渠道我可以很容易地知道嘗試成為 Graham Stefan 或認識 Kevin 並在金融領域進行擴張
1:04:06
位，除此之外還有很多層次，我非常有信心無論如何我都能執行這個策略
1:04:11
我所關注的可能會吸引到 10 100 倍的觀眾，但為了做到這一點
1:04:18
如果我必須遵守算法或不遵守算法，我就看不到路徑
1:04:23
我完整而真實的自我，嗯，但我絕對認為我有點像戰鬥
1:04:30
好吧，因為是的，我的意思是你有一個巨大的大腦
1:04:37
把煙吹起來，所以我通常會看到你的能力 我了解你在 App Store 的歷史 看到機會並執行，我知道
1:04:43
如果你要去你知道什麼讓我們現在正在做的事情只有 100 倍你會做你會執行它並且
1:04:49
做一些事情，這只是值得走一條不同的道路去接觸更多的人，並擁有
1:04:55
更多的影響和一些我與之抗爭的東西，因為我不能不是我，但我知道我是否能夠拒絕它並且
1:05:00
做的事情可以走不同的道路，接觸更廣泛的受眾，可能會產生更大的淨積極影響是的，這是一個艱難的過程是的
1:05:07
絕對嗯我想問這個我打算離線問你但我想現在就問你
1:05:13
嗯，您的商品銷售情況如何，您的體型相當可觀
1:05:18
知道收入進來是否值得像你知道的那樣值得嗎
1:05:24
不是時間能量的東西實際上不好，如果我把時間投入進去
1:05:29
當我去年積極推廣商品作為實驗時，我只是想我會不斷地推銷商品，但我沒有
1:05:34
我的意思是插入我的意思是喜歡在每個視頻中唯一地談論商品，而不僅僅是複制和粘貼相同的內容
1:05:40
促銷這樣做並整合它會更有效，是的，當我這樣做時，我每天都在賣數百美元的商品，
1:05:47
如果我們每天發布一段視頻，那就很多了
1:05:53
而不是自己鞭打它們，我可能會從商品中獲得更多收入
1:05:58
僅銷售額就比所有其他綜合收入來源高出一大截，只是我對創造不感興趣
1:06:05
定期推出新商品並不斷推廣它我會做新的事情我更專注於視頻而且我在桌面上留下了很多錢
1:06:11
那裡，但我認為對於大多數創作者來說，如果你沒有，如果你有
1:06:16
粉絲而不是觀眾，或者這是人們非常熱衷的主題
1:06:21
人們喜歡商品，他們喜歡支持創作者，如果你有引人注目的設計，而且它們是新鮮的，而且
1:06:26
獨特的人喜歡它，您無需進行大量轉換即可獲得合理的信息
1:06:32
您發布非常頻繁的視頻，並且每次都提到您的商品，如果您有 5 個或
1:06:38
每個商品 10 美元的利潤才能真正開始加起來，我認為有很多 YouTube 創作者
1:06:45
即使他們有廣告、贊助和展示位置，他們仍有一半以上的收入來自商品
1:06:51
以及諸如此類的巨大未開發潛力我認為很多創作者是的，顯然我有
1:06:57
就 Elon 和 Tesla 書呆子而言，這是不公平的優勢，因為如果你想要一件 Tesla 的 meme 襯衫
1:07:02
沒有那麼多地方你會找到它們，或者無論如何也沒有它可能已經改變了很多現在我看到很多人抄襲我的設計
1:07:08
諸如此類的東西，但是有一段時間，如果您不想打賭 Elon 襯衫，我的意思是只有其中一個定義了我
1:07:15
認為僅該單位就已售出超過 4 000 件單品，令人難以置信
1:07:21
他們是很好的談話開始者，雖然是的，是的，是的，你做了一些很棒的設計，我絕對
1:07:26
欣賞它，嗯，是的，有趣的東西，嗯，這是一個很好的追趕，尤其是在之後
1:07:32
幫助並了解您的想法，嗯，我相信人們會很高興進入
1:07:38
你的頭有點嗯，像這樣的談話只是為了退後一步，看看更大的
1:07:44
圖片我會繼續在下面的視頻描述中鏈接到您的 YouTube 頻道任何其他類型的最後一個
1:07:50
最後我可能會鼓勵每個真正不購買的人觀看之前的文字或評論
1:07:57
圍繞董事會和 AGI 的論文，盡可能多地嘗試和教育自己，以真正做好
1:08:03
評估特斯拉是否在路上，以及他們實現這一目標的可能性有多大，因為即使你
1:08:09
現在不要買它，只要你能把自己置於一個你可以明確知道他們錯了什麼的境地
1:08:14
我不反對我們有足夠的信息來真正做出判斷我認為你會處於一個更好的位置來看到潛力
1:08:19
機會在這裡，因為正如我所提到的，我真的認為你需要很多關於廣泛的相當專業的知識
1:08:25
抓住機會的一系列真正書呆子的事情，正如我所說的機器人
1:08:30
AGI 讓特斯拉未來的其他一切都相形見絀
1:08:35
規模和影響是荒謬的你知道市值它只是變得奇怪但我認為很多
1:08:41
人們很難實現這一飛躍，因為他們的知識沒有太多差距，無法真正畫出任何合理的
1:08:47
關於他們所看到的結論，呃，但不要錯過了解機會，伙計們，這是我的最後一句話，好吧，好東西，嗯，是的，謝謝安娜
1:08:56
謝謝再次聊天，嗯，很喜歡，我們會再見的，稍後再見，看看其他正在觀看的人，謝謝
1:09:02
見到你們
沒有留言:
張貼留言