Latest / Elon Musk Thinking / NVIDIA CEO Jen Hsun Huang Interviews Elon Musk, 10 Years Ago!!!

Transcript

NVIDIA CEO Jen Hsun Huang Interviews Elon Musk, 10 Years Ago!!! · Elon Musk Thinking

• 0:01
  Speaker 1: Now now you know, we made it a point not
• 0:04
  Speaker 1: to not to rehearse anything, and so as I just
• 0:07
  Speaker 1: want to just as a as a as a just reminder,
• 0:10
  Speaker 1: you're you're my last thing. Okay, okay, could you not
• 0:13
  Speaker 1: ruin the whole thing?
• 0:14
  Speaker 2: All right?
• 0:15
  Speaker 1: All right? So now speaking of that, speaking of that,
• 0:19
  Speaker 1: I think everybody would like to before we get into
• 0:21
  Speaker 1: all of the good stuff, okay, and they want to
• 0:24
  Speaker 1: go directly to the juicy stuff.
• 0:26
  Speaker 2: Okay, okay.
• 0:27
  Speaker 1: And the juicy stuff is this. Look, you know, you
• 0:31
  Speaker 1: were quote as a saying that that artificial intelligence is
• 0:37
  Speaker 1: more dangerous than nuclear weapons, and I said potentially, and
• 0:44
  Speaker 1: and well it goes on, it goes on, and you say,
• 0:48
  Speaker 1: you say that it's like summoning the demon could be.
• 0:53
  Speaker 1: How do you consolidate, rationalize the the the conflict between
• 0:56
  Speaker 1: artificial intelligence of course deep learning that that obviously is
• 1:00
  Speaker 1: going to be very important to self driving cars. How
• 1:01
  Speaker 1: do you think through that?
• 1:03
  Speaker 2: Well, I don't think.
• 1:04
  Speaker 3: We're to worry about autonomous cars because that's sort of
• 1:07
  Speaker 3: like a narrow form of AI, and it's not something
• 1:12
  Speaker 3: that I think is very difficult. Actually, I think the
• 1:14
  Speaker 3: to do autonomous driving to agree that's much safer than
• 1:18
  Speaker 3: a person.
• 1:19
  Speaker 2: Is much easier than people think.
• 1:21
  Speaker 3: Yeah, right, and yeah, I think it's going to just
• 1:26
  Speaker 3: become normal, Like it be like an elevator, Like nobod
• 1:29
  Speaker 3: they used to have elevator operators and then we you know,
• 1:33
  Speaker 3: we've developed some simple circuitry to have elevators just automatically
• 1:38
  Speaker 3: come to the floor that you're at and you can
• 1:40
  Speaker 3: just press the button. Nobody needs to operate the elevator.
• 1:44
  Speaker 3: The car is just going to be like that.
• 1:46
  Speaker 1: And the elevators these days are even smart. I mean
• 1:48
  Speaker 1: it knows it knows where to position an elevator, so
• 1:52
  Speaker 1: that if you were to need an elevator, it's pretty
• 1:55
  Speaker 1: close to you. Cars in the future will be pretty
• 1:56
  Speaker 1: smart about that too.
• 1:57
  Speaker 3: Yeah, you'll be able to tell your car like take home,
• 2:01
  Speaker 3: go here, go there, anything, and it'll just do it
• 2:06
  Speaker 3: at an order of mag to safer than a person.
• 2:09
  Speaker 2: In fact, in the in the distant future, I.
• 2:11
  Speaker 3: Think it's probably going to be if people may outlaw
• 2:15
  Speaker 3: driving cars because it's too dangerous, Like you can't have
• 2:19
  Speaker 3: a person driving a two time death machine.
• 2:24
  Speaker 1: Now, if we if we have the right type of
• 2:26
  Speaker 1: intelligence in a car, we we also don't have to
• 2:29
  Speaker 1: make the cars that heavy. I would think, you know,
• 2:32
  Speaker 1: cars are getting heavier and heavier, and it's got more
• 2:34
  Speaker 1: and more stuff in it because it needs to survive
• 2:37
  Speaker 1: all these incredible collisions and things like that. If I
• 2:39
  Speaker 1: wonder if if we were to design cars that just
• 2:43
  Speaker 1: simply don't collide as much. I wonder if we could,
• 2:46
  Speaker 1: we could relax on some of those laws and and yeah,
• 2:49
  Speaker 1: make cars more fuel efficient and lighter and better to.
• 2:51
  Speaker 2: Right, you could definitely do that.
• 2:52
  Speaker 3: If you could count on not having an accident, then
• 2:54
  Speaker 3: you can get rid of a huge amount of the
• 2:57
  Speaker 3: crash structure and the airbags. And it'll be We're a
• 3:02
  Speaker 3: long way from that, because there's always gonna be some
• 3:04
  Speaker 3: for a very long time, there'll be some amount of
• 3:05
  Speaker 3: legacy cars on the road. And I think it is
• 3:09
  Speaker 3: important to just appreciate the size of the automotive industrial base.
• 3:14
  Speaker 3: Like it's not as though like when somebody makes an
• 3:18
  Speaker 3: autonomous car that suddenly all the cars will be autonomous.
• 3:20
  Speaker 2: It's like there's two billion of them.
• 3:23
  Speaker 3: Okay, So the total total number of cars and trucks
• 3:26
  Speaker 3: on the road is two billion. In climbing, the capacity
• 3:31
  Speaker 3: of car and truck production is about one hundred million
• 3:34
  Speaker 3: a year. So if tomorrow all cars were autonomous, it
• 3:38
  Speaker 3: would take twenty years to replace the fleet, assuming the
• 3:40
  Speaker 3: fleet stayed the same size. Arguably it could get smaller
• 3:43
  Speaker 3: if things are autonomous, but still it's it's still you know,
• 3:47
  Speaker 3: maybe fifteen years or something, and it's not all going
• 3:50
  Speaker 3: to transition immediately. It'll take quite a while. So and
• 3:55
  Speaker 3: it's the same for electrification of carsing that industrial base
• 4:01
  Speaker 3: to be electric. I mean, if if all cars were
• 4:04
  Speaker 3: suddenly fall cars produced were electric tomorrow, it would still
• 4:07
  Speaker 3: take twenty years to replace the fleet.
• 4:10
  Speaker 2: And right now it's less than one percent.
• 4:12
  Speaker 1: So now you you you're you mentioned just now about
• 4:16
  Speaker 1: about self driving cars being easier than people think. Now
• 4:19
  Speaker 1: you have your vision of how to go from where
• 4:21
  Speaker 1: we are today. Now, my model, my P eighty five
• 4:24
  Speaker 1: D has lane detection and so it gets a little
• 4:28
  Speaker 1: you know, when I get close to a lane, yep,
• 4:31
  Speaker 1: it detects the uh uh the speed signs and then
• 4:34
  Speaker 1: uses uses computer vision technology to do that. And but
• 4:39
  Speaker 1: and that's today's adas. What is your what is your roadmap?
• 4:43
  Speaker 1: You know, how is that different than other people's roadmap?
• 4:45
  Speaker 1: How do you think about how to get to self
• 4:46
  Speaker 1: driving cars.
• 4:48
  Speaker 2: Yeah, well.
• 4:50
  Speaker 3: You kind of need the hardware foundation, the sort of
• 4:52
  Speaker 3: sensor and computing foundation, and then you can keep uploading
• 4:55
  Speaker 3: new software. At least you can with the TAILS because
• 4:57
  Speaker 3: it's it's always connected. So the car that you have,
• 5:01
  Speaker 3: you'll notice that it's the features are steadily improving. We
• 5:06
  Speaker 3: now you know, have active cruise control, so it'll it'll
• 5:09
  Speaker 3: use radar and camera fusion to track the car in
• 5:12
  Speaker 3: front of you. It's also looking at with some of
• 5:16
  Speaker 3: the things that are coming out. It's got it looks
• 5:18
  Speaker 3: at the brake lights, so it anticipates that the car's
• 5:20
  Speaker 3: got the brake lights are active. It's going to get
• 5:23
  Speaker 3: basically smarter and smarter even with the current hardware suite.
• 5:26
  Speaker 3: So the current hardware suite is three hundred and sixty
• 5:28
  Speaker 3: degree ultrasonic sensors that go off to about just over
• 5:31
  Speaker 3: five meters. It's a forward camera to Ford radar. So
• 5:34
  Speaker 3: we'll make even with just that sensor suite, we can
• 5:38
  Speaker 3: actually make a huge progress in autonomy. We can certainly
• 5:44
  Speaker 3: make the car steer itself on on a freeway, do
• 5:49
  Speaker 3: lane changes. It's really autonomy is about what level of
• 5:54
  Speaker 3: reliability and safety do you want?
• 5:58
  Speaker 2: Even with the current sensorus.
• 6:00
  Speaker 3: We could make the cargo fully autonomous, but only to
• 6:03
  Speaker 3: but not to a level of reliabily that would be
• 6:05
  Speaker 3: safe in say a complex urban environment at thirty miles
• 6:12
  Speaker 3: an hour, where the lanemarking's not there and children could
• 6:14
  Speaker 3: be playing and things could be coming at.
• 6:17
  Speaker 2: You from the side.
• 6:18
  Speaker 3: So in order to solve that, you need a bigger
• 6:23
  Speaker 3: sensor suite, and you need more computing power. And I
• 6:27
  Speaker 3: think what you're doing actually with the tigers in the
• 6:31
  Speaker 3: futures is super interesting and will really be a big
• 6:33
  Speaker 3: enabler for autonomous driving. So I think, you know, we're
• 6:37
  Speaker 3: in video is doing really great stuff on that front.
• 6:38
  Speaker 1: I appreciate that. Yeah, And so some of the challenges
• 6:41
  Speaker 1: that you see, what are the what are some of
• 6:43
  Speaker 1: the technological hurdles that And there's all kinds of researchers
• 6:49
  Speaker 1: in the room, they're all kinds of engineers in room.
• 6:50
  Speaker 1: What are some what are some of the technological hurdles
• 6:52
  Speaker 1: that you think are really important for us to go tackle. Surely,
• 6:56
  Speaker 1: surely we're going to get to some better cruise controls
• 7:00
  Speaker 1: on highways, but beyond that, what are some of the
• 7:03
  Speaker 1: things that you would like is to go focus on
• 7:05
  Speaker 1: the tackle for the car.
• 7:06
  Speaker 3: Industry, Well, it's it. You know, where it gets tricky
• 7:11
  Speaker 3: is is just the is that sort of urban environment
• 7:16
  Speaker 3: around thirty or forty miles an hour. So like right
• 7:19
  Speaker 3: right now, it's fairly easy to deal with, say, things
• 7:22
  Speaker 3: that are sub five to ten miles an hour, because
• 7:25
  Speaker 3: we can do that with the ultrasonics. We just make
• 7:28
  Speaker 3: sure it doesn't hit anything right, you know, because you.
• 7:30
  Speaker 1: Can always just the right thing to do. Largely, that's
• 7:32
  Speaker 1: why would.
• 7:33
  Speaker 3: You want to hit anything with your exactly, So at
• 7:36
  Speaker 3: five ten miles an hour, you can stop within the
• 7:38
  Speaker 3: range of the ultrasonics, and that then from let's say
• 7:43
  Speaker 3: ten miles an hour to.
• 7:46
  Speaker 2: You know, call it sort of fifty miles an hour.
• 7:48
  Speaker 3: That that that that area in complex suburban environments, that's
• 7:54
  Speaker 3: that's where you can get a lot of unexpected things happening,
• 7:59
  Speaker 3: like let's say this of like a road closure or
• 8:01
  Speaker 3: a manhole cover open. Children playing is a big issue bicycles.
• 8:07
  Speaker 3: Once you get about fifty miles an hour and you're
• 8:09
  Speaker 3: in kind of a freeway environment, then it also gets
• 8:11
  Speaker 3: easier again, like the set of possibilities is much reduced,
• 8:17
  Speaker 3: so highway crews is easy. Low speed is easy, intermediate
• 8:22
  Speaker 3: is hard. And so being able to recognize what you're
• 8:27
  Speaker 3: seeing and make the right decision in the suburban environment
• 8:31
  Speaker 3: in that ten miles an hour to fifty mile an
• 8:33
  Speaker 3: hour zone is the challenging portion. But I really think
• 8:40
  Speaker 3: it's I mean, I almost this may sound a little complacent,
• 8:42
  Speaker 3: but I almost viewed it as like a solve problem,
• 8:44
  Speaker 3: Like we know exactly what.
• 8:45
  Speaker 2: To do, and we'll be there in a few years.
• 8:48
  Speaker 1: Right, It's just like Mars. That's quite side. That's kind
• 8:56
  Speaker 1: of the spirit of a I mean, in a lot
• 9:01
  Speaker 1: of ways in your mind, you kind of you kind
• 9:03
  Speaker 1: of see things solvable or arguably arguably solved, and and
• 9:07
  Speaker 1: a lot of it is really about getting there.
• 9:10
  Speaker 3: Yeah, we'll take autonomous costs for granted, in quite a
• 9:13
  Speaker 3: short period of time. It's amazing how comfortable you get
• 9:16
  Speaker 3: at how quickly you get comfortable with it.
• 9:19
  Speaker 2: So, now, what.
• 9:21
  Speaker 1: About government government policies? Like one of the things that
• 9:24
  Speaker 1: I would like to do is I would I would
• 9:26
  Speaker 1: just like to keep working on my email as I'm
• 9:28
  Speaker 1: driving to work. Sure, you know, there's there's a thirty
• 9:30
  Speaker 1: people do that already. Like I said, I would like
• 9:36
  Speaker 1: to do it without without without breaking the law. So
• 9:41
  Speaker 1: where where where do you where do you think government
• 9:45
  Speaker 1: intervention falls in some of this stuff? Because you know, obviously,
• 9:48
  Speaker 1: if you car drivee by itself and it does it
• 9:50
  Speaker 1: even better than people, you would like it to drive
• 9:53
  Speaker 1: by itself, but largely the laws don't allow you to
• 9:55
  Speaker 1: do that today, right, absolutely, So how do we cross
• 9:57
  Speaker 1: that bridge? And how do you think about government invention regulations?
• 10:02
  Speaker 2: Right? So, I think.
• 10:05
  Speaker 3: It'll be from the point at which a car is
• 10:07
  Speaker 3: definitely safer than a person, there's probably at least another
• 10:12
  Speaker 3: two or three years after that before regulators will allow
• 10:15
  Speaker 3: that to be the case, because they will want to
• 10:17
  Speaker 3: see a large amount of statistical proof that it's not
• 10:21
  Speaker 3: merely as safe as a person, but much safer. So
• 10:25
  Speaker 3: I think what you can do is you can run
• 10:27
  Speaker 3: run it in shadow mode and essentially say, okay, this
• 10:30
  Speaker 3: is this is what the computer would have done in
• 10:32
  Speaker 3: all these circumstances, and was there a crash or was
• 10:35
  Speaker 3: there not?
• 10:36
  Speaker 2: Like what are the false parts of false negatives?
• 10:39
  Speaker 3: And then you know, it's achieve a large population group
• 10:44
  Speaker 3: and then and then make a really clear statistical argument
• 10:47
  Speaker 3: with the regulators, and then they're going to digest that,
• 10:50
  Speaker 3: observe it for a while, see if they agree with it,
• 10:52
  Speaker 3: and then I think they will because the evidence will
• 10:54
  Speaker 3: be overwhelming.
• 10:55
  Speaker 1: Yeah, and the evidence is actually already quite overwhelming that
• 10:58
  Speaker 1: if you, if you, if you would have would have
• 11:01
  Speaker 1: noticed a break line in front of you in the
• 11:03
  Speaker 1: highway and you didn't, you didn't crash into a rear
• 11:07
  Speaker 1: in collision. Right, A lot of laser sate, you know, ideally, ideally,
• 11:10
  Speaker 1: hopefully people don't don't overreact with this, with this unknown technology,
• 11:16
  Speaker 1: and uh and prematurely regulate no premature regulations.
• 11:23
  Speaker 2: Well, I mean regulation that was a joke.
• 11:31
  Speaker 3: I think it when when it comes to public safety,
• 11:33
  Speaker 3: I think there's there's an argument for being quite cautious
• 11:37
  Speaker 3: and making sure that things are okay before before there's
• 11:41
  Speaker 3: a change.
• 11:43
  Speaker 2: And I mean, and I.
• 11:45
  Speaker 3: Don't think it's the case that right now there's a
• 11:48
  Speaker 3: fully autonomous system and regulators are not approving it, that
• 11:51
  Speaker 3: that that really be a substitute for people. But they
• 11:53
  Speaker 3: will be in a few years now.
• 11:55
  Speaker 1: As we get more computer rized technology into these cars
• 11:58
  Speaker 1: and this car becomes really a software defined car. I
• 12:01
  Speaker 1: mean a lot of your engineers are software engineers.
• 12:03
  Speaker 2: I mean that's yeah, absolutely.
• 12:04
  Speaker 1: One of the great things about Tesla. You guys right
• 12:06
  Speaker 1: here in Silicon Valley, you're rich with software engineers and
• 12:09
  Speaker 1: you have that you' have that computer sensibility about architecting
• 12:13
  Speaker 1: a computer properly, designing the software, properly designing the software
• 12:17
  Speaker 1: for many generations of cars, so where he refines and
• 12:19
  Speaker 1: gets better and better. And it has been getting better.
• 12:21
  Speaker 1: I mean the software from the first time you send
• 12:24
  Speaker 1: me my Tesla to the now it's just like unrecognizable software.
• 12:28
  Speaker 2: Right, big improvement.
• 12:30
  Speaker 3: So that's why the first thing we try to do
• 12:32
  Speaker 3: is establish the hardware platform, make sure that we have
• 12:36
  Speaker 3: the sensors and compute power. And so we do that first,
• 12:40
  Speaker 3: even though the software is only taking advantage of a
• 12:43
  Speaker 3: small percentage of the sensors and compute power. And then
• 12:47
  Speaker 3: we do continuous updates to make the car more and
• 12:51
  Speaker 3: more capable.
• 12:52
  Speaker 2: And we're going to see a lot of that happen
• 12:54
  Speaker 2: later this year.
• 12:57
  Speaker 3: If I didn't have an announcement on Thursday morning, I
• 12:59
  Speaker 3: would be saying a lot more of it.
• 13:00
  Speaker 1: Yeah, the audience doesn't understand why they have to wait
• 13:06
  Speaker 1: until Thursday morning. You tweeted it already. You're announce that
• 13:09
  Speaker 1: you're going to do an Ota. What kind of announcement
• 13:11
  Speaker 1: is that I'm going to do an Ota on Thursday.
• 13:14
  Speaker 1: That's like a new product announcement.
• 13:15
  Speaker 3: These days, it's just it's just a well, it's just
• 13:17
  Speaker 3: saying that there's going to be a cool on Thursday morning,
• 13:20
  Speaker 3: and I'll describe what's going to be in version six
• 13:22
  Speaker 3: point two for anyone who's interested.
• 13:25
  Speaker 1: That's so awesome. Though I'm interested. I get excited every
• 13:27
  Speaker 1: time I get an Ota, and it's you know, one
• 13:29
  Speaker 1: of the things that was really interesting is in the beginning,
• 13:32
  Speaker 1: when we first built the first Tesla together, the tegra
• 13:36
  Speaker 1: in it, we thought was more than enough. And recently
• 13:41
  Speaker 1: you said, can we just squeeze more performance out of
• 13:44
  Speaker 1: that platform? And it just happened in literally two years,
• 13:47
  Speaker 1: you know, several versions of your software updates all of
• 13:49
  Speaker 1: a sudden, the computing platform is not powerful enough, right,
• 13:53
  Speaker 1: And it's because you want to add more features, and
• 13:54
  Speaker 1: a lot of features these days are based on software.
• 13:56
  Speaker 2: True.
• 13:57
  Speaker 1: Yeah, And so one last question, and it's it has
• 14:01
  Speaker 1: to do with I guess something that a lot of
• 14:04
  Speaker 1: people are very concerned about, which is your car becomes
• 14:07
  Speaker 1: a software platform and software platforms get hacked. How do
• 14:10
  Speaker 1: you think about that, How do you think about security?
• 14:12
  Speaker 1: And what are some of the things that we could
• 14:13
  Speaker 1: do to try to make make make the car more
• 14:17
  Speaker 1: resilient to security attacks.
• 14:20
  Speaker 3: Yeah, I think that that becomes really important when the
• 14:22
  Speaker 3: cars are fully autonomous. I mean, the way the cars
• 14:25
  Speaker 3: work right now, every system of the car, it's assumed,
• 14:29
  Speaker 3: could actually have a mechanical failure of some kind or
• 14:32
  Speaker 3: a logic failure, a fundamental logic failure. So you can
• 14:35
  Speaker 3: always overwhelm the breaking of the car with your foot,
• 14:39
  Speaker 3: and you can overwhelm the steering wheel with your hands. So,
• 14:43
  Speaker 3: but but when when there is a steering wheel, or
• 14:44
  Speaker 3: there isn't you know, a brake, pedal or something in
• 14:46
  Speaker 3: that like you know many is from now, then it's
• 14:49
  Speaker 3: really really dangerous, you know, because but even as it
• 14:54
  Speaker 3: is right now, what we spend most of our time
• 14:57
  Speaker 3: on is making sure that it's it's very difficult to
• 14:59
  Speaker 3: do a multi car hack. Like if you have direct
• 15:03
  Speaker 3: access to a car, just like if you've got direct
• 15:05
  Speaker 3: access to a computer or even a conventional car, you
• 15:08
  Speaker 3: can do a lot of things to it.
• 15:11
  Speaker 2: But but that's less of.
• 15:12
  Speaker 3: A concern than somebody being able to hack an arbitrary
• 15:14
  Speaker 3: car or multiple cars. So that's what we focus our
• 15:18
  Speaker 3: energy on is making sure that that in that ways
• 15:21
  Speaker 3: it's it's it's a lot like a like a cell
• 15:23
  Speaker 3: phone or a laptop, you know, you focus on making
• 15:26
  Speaker 3: sure that they can't or that it's very difficult for
• 15:29
  Speaker 3: that to be any kind of system wide hack. So
• 15:33
  Speaker 3: we put a lot of effort into that, and we
• 15:35
  Speaker 3: have third parties try to attack it. And in certain
• 15:40
  Speaker 3: parts of the car at the very fundamental level, like
• 15:43
  Speaker 3: the drive unit controller or the steering controller have an
• 15:47
  Speaker 3: additional level of security. So somebody may be able to uh,
• 15:53
  Speaker 3: you know, hack something that's cosmetic, but it's much harder
• 15:57
  Speaker 3: to hack something that's that's actually physically agerous.
• 16:01
  Speaker 2: There's multiple levels of.
• 16:02
  Speaker 1: Secure and so this way, if you if you weren't
• 16:04
  Speaker 1: able to penetrate maybe the entertainment system, it doesn't allow
• 16:08
  Speaker 1: you quickly as a result of.
• 16:09
  Speaker 3: That, right, I may display a funny message or something,
• 16:12
  Speaker 3: but it would not you would not be able to
• 16:14
  Speaker 3: then control the steering or the motor.
• 16:16
  Speaker 1: Yeah. Well, the future of cars is so exciting and
• 16:19
  Speaker 1: the work that you guys are doing are so exciting,
• 16:21
  Speaker 1: and it's it's it's great to see you guys pioneering
• 16:24
  Speaker 1: these computer rized cars. I mean a lot of people
• 16:26
  Speaker 1: think about think about Tesla as the electric car, but
• 16:30
  Speaker 1: I think it's obviously more than that. It's an electric car,
• 16:33
  Speaker 1: but it's a whole computer platform on top of that.
• 16:36
  Speaker 3: Yeah, I think I think Tess is sort of the
• 16:38
  Speaker 3: leader in electric cars, but I think we'll also sort
• 16:40
  Speaker 3: of be the leader in autonomous cars, at least autonomous
• 16:43
  Speaker 3: cars that people can buy. And and so we're I mean,
• 16:48
  Speaker 3: if there's anybody interested in working on autonomous cars, would
• 16:50
  Speaker 3: love to have you work at Tesla, by the way.
• 16:56
  Speaker 3: So we're gonna put a lot of effort into automotive
• 17:00
  Speaker 3: autonomous driving because it's just going to be the default
• 17:02
  Speaker 3: thing and it could save a lot of lives.
• 17:05
  Speaker 1: Yeah, so save a lot of lives and hopefully, hopefully
• 17:08
  Speaker 1: one of these days, it would be nice if Nvidia's
• 17:10
  Speaker 1: campus has no parking lot, yeah right, that it drops
• 17:13
  Speaker 1: us off and the meanders off to a place where
• 17:15
  Speaker 1: the land's a little cheaper and you know, and parks
• 17:17
  Speaker 1: a whole bunch of cars there and when it's time
• 17:19
  Speaker 1: to go home, were someone at to.
• 17:21
  Speaker 3: Come, it will be extremely transformative, that's for sure. But yeah,
• 17:26
  Speaker 3: I mean when it comes to AI, I'm not really
• 17:27
  Speaker 3: worried about narrow AI like like autonomous cars or like
• 17:32
  Speaker 3: you know, a smart air conditioning unit at the house
• 17:36
  Speaker 3: or something. It's more like sort of the deep intelligent
• 17:39
  Speaker 3: stuff that is where we need to be cautious. I
• 17:43
  Speaker 3: actually think there's many potential flavors of AI. And you know,
• 17:49
  Speaker 3: it's odd that we're at we're so close to the
• 17:51
  Speaker 3: advent of AI, Like it's it seems strange that we
• 17:53
  Speaker 3: would be alive in this in this time.
• 17:57
  Speaker 1: We'll come back every year, come back every year, and
• 17:59
  Speaker 1: you'll see the work that this group was going to do.
• 18:02
  Speaker 1: I mean, there's so much deep learning work being done here.
• 18:05
  Speaker 1: You have a lot of engineers here as well, and
• 18:07
  Speaker 1: there is fantastics to see the whole community focused on
• 18:12
  Speaker 1: advancing this field. And along the way, we're gonna spin
• 18:14
  Speaker 1: off a whole bunch of new capabilities. As you know,
• 18:16
  Speaker 1: that's gonna make cars just safer and more fun to drive.
• 18:19
  Speaker 1: Long before we have to get to essentially a self
• 18:22
  Speaker 1: driving car, there's going to be a lot of versions
• 18:24
  Speaker 1: along the way that's just going to bring joy to
• 18:26
  Speaker 1: a lot of people.
• 18:28
  Speaker 2: Yeah. Absolutely, I just hope there's something left for used
• 18:30
  Speaker 2: humans to do.
• 18:32
  Speaker 1: Well. I'm not gonna let let go of my steeringwell,
• 18:34
  Speaker 1: you know, I've got mine on the craziness mode and
• 18:37
  Speaker 1: the sports steering mode. Is that the way you have it?
• 18:41
  Speaker 1: You get driven to work? Now?
• 18:42
  Speaker 3: No, well I drive half the time actually, And which
• 18:47
  Speaker 3: mode do you have it in? I always have it
• 18:48
  Speaker 3: insane mode?
• 18:49
  Speaker 1: Yeah right, all right, thank you,