#1470 - Elon Musk

你知道，隧道公司占的脑力资源还不到1%。

You know, Boring Company does not take you know, it's, like, less than 1% of brain cycles.

还有Neuralink，这个嘛，我也不太好说。

And and then this Neuralink, which is I don't know.

大概占5%左右吧。

Maybe it's, like, 5%.

再加5%。

And then 5%.

加起来就是

That's that's

不小的一块了。

a good chunk.

确实是一大块。

It's good chunk.

是的。

Yeah.

是的。

Yeah.

上次我们聊过这个，你当时还在琢磨它到底什么时候能上线、什么时候能用上。

We were talking about that last time, and you were trying to figure out when it was actually going to go live, when it's actually going to be available.

你们现在已经在人身上做测试了吗？

Are you testing on people right now?

还没有。

No.

还没在人身上试，但我觉得不会等太久。

We're not testing on people yet, but I I think it won't be too long.

我觉得一年内就能给人体植入Neuralink了。

I think we may be able to implant a Neuralink in less than a year in in a person, I think.

而且

And

你们做这个之前，会不会先做些测试，看看有多少人身体会排斥这些装置？

when you do this, is there any test that you have to do before you do something like this to to see what percentage of people's bodies are going to reject these things?

有没有可能被身体排斥呢？

Is it put is it there is there a potential for rejection?

排斥的可能性非常低。

It's it's a very low potential for rejection.

你可以想想，很多人体内植入的是心脏监测器，或者癫痫治疗设备、深部脑刺激装置之类的。

I mean, you can think of it like people put in, you know, heart monitors and, you know, things for epileptic seizures and deep brain stimulation.

还有人工髋关节，对吧？

Obviously, like, you know, artificial hips and Right.

人工膝关节，这类东西都是常见的。

Knees and that kind of thing.

所以，什么情况会导致排斥、什么不会，这方面已经有很明确的了解了。

So the the probability of I mean, like, it's so it's well known, like, what will cause rejection and what what will not.

不过，当设备需要读取和写入神经元信号时，情况确实更复杂一些。

It's definitely hotter when you've got something that is re re sort of reading and writing neurons.

因为它要产生电流脉冲，同时还要读取神经的电流信号。

That's that's generating a a current pulse and reading current pulses.

这比被动设备稍微复杂一点，但依然完全可行。

That's that's a little hotter than than a than a than than, say, a passive device, but it's still, you know, very doable.

而且确实现在就有人在脑子里植入了初级设备。

And, yeah, that there that there are people who have primitive devices in in their brains right now.

是什么样的设备？

What kind of devices?

比如深部脑刺激，主要用于帕金森病，已经极大地改善了患者的生活，这其实挺了不起的，因为这相当于用电流刺激你的大脑。

I would like deep deep brain stimulation is for, I think, for Parkinson's is like, has really changed people's lives in in a big way, which is kinda remarkable because we're it it it it kind of like zaps your brain.

就像拍打电视机一样。

It's like kicking the TV type of thing.

你可能会想，拍打电视机怎么可能有用呢？

And you think, like, man, kick kicking the TV shouldn't work.

但有时候还真管用。

It does sometimes.

是的。

Yeah.

是的

Yeah.

老式电视机那时候。

The old the old TVs.

确实管用。

It did.

我爷爷以前总拍电视顶上。

My grandpa used to slap the top.

没错。

For sure.

是的

Yeah.

有时候还真能修好。

It would work sometimes.

是的

Yeah.

这种深部脑刺激植入设备彻底改变了人们的生活，带来了根本性的积极影响。

So this deep brain simulation implanted devices in the brain that have changed people's lives for the better, like, fundamentally.

咱们聊聊你能谈的关于Neuralink的内容吧，上次你来的时候，我根本没法讨论这个话题。

Well, let's talk about what you can talk about to what Neuralink is because the last time you were here, I really couldn't discuss it.

后来好像出了个新闻稿之类的，大概讲了一下情况。

And then there was a, I guess, a press release or something that sort of outlined.

是的。

Yeah.

当然。

Sure.

上次你来之后，这种情况发生过好几次。

That that that happened quite a bit after the last time you were here.

那它到底是什么东西？

So what exactly is it?

你们是怎么做的？如果真有人做了Neuralink植入，会发生什么？

How do you do what what happens if someone ultimately does get a Neuralink installed?

会发生什么？

What will take place?

第一代设备会植入你的头骨里。

Well, for version one of the device, it would be it it basically implanted in your skull.

而且会和头骨表面齐平。

So but it would be flush with your skull.

basically 就是取下一块头骨，把 Neuralink 设备放进去。

So you basically take out a chunk of skull, replace put put the Neuralink device in there.

然后小心地把电极线插入大脑，缝合伤口，外人根本看不出来你植入了设备，它就能和你大脑的任何部位进行交互。

You you put the the electrode you'd insert the electrode threads very carefully into the the brain, and and then you, you know, stitch it up and and you wouldn't even know that somebody has And then and and so then it it it can interface basically anywhere anywhere in your brain.

比如它可以用来治疗视力问题，即使视神经已经受损，也能帮你恢复视力。

So it could be something that, you know, helps cure, say, eyesight, like, give you returns your eyesight even if you've, like, lost your optic nerve type of thing.

真的吗？

Really?

是的。

Yeah.

是的

Yeah.

当然

Absolutely.

听觉，显而易见。

Hearing, obviously.

我的意思是，原则上，只要大脑出了问题，它几乎都能解决。

I mean, pretty much anything that where that that it it could, in principle, fix almost anything that is wrong with the brain.

它还能恢复肢体功能。

And it it could it could restore limb functionality.

只要能接入运动皮层，再配合植入肌肉群里的微型控制器，就能搭建一条神经通路，让截瘫患者完全恢复行动能力。

So if you've got interface into the motor cortex and then an implant that's, say, that's like a microcontroller in in your muscle groups, you you could then create a sort of a neural shunt that restores somebody who is a quadriplegic to full functionality.

他们能走路，像正常人一样。

Like, they can walk around, be normal.

天啊。

Woah.

是的

Yeah.

所以

So

可能更好一点

Maybe slightly better.

更好一点？

Slightly better?

随着时间推移

Over time.

对

Yes.

你是说未来的升级版本？

You mean with future iterations?

就像那个价值六百万美元的男人。

Like, the, you know, $6,000,000 man.

不过现在看来，这笔钱已经不算多了。

Although these days, that would that doesn't seem much.

是的。

Yeah.

对。

Yeah.

六十亿美元的人。

$6,000,000,000 man.

所以。

So

你得钻一个洞，再换上这块东西，那洞得打多大？

the the hole would be small how big would the hole be that you have to drill and then replace with this piece?

就一个洞？

It's only one hole?

嗯，我们目前正在做的这个设备，直径大概一英寸左右。顺便说一句，你的头骨可不薄。

Well, yeah, the device we're working on right now is about it's about an inch in diameter, and your your skull's pretty thick, by the way.

头骨嘛

So skulls are

我的就是。

Mine is.

当然了。

For sure.

说实话，如果你眼睛比较大，头骨其实还挺厚的。

It might actually literally be I mean, if you're a big if you're a big eye, your skull is actually fairly thick.

头骨厚度大概在七到十四毫米之间。

Skulls, like it's, like, seven to 14 millimeters.

那差不多有几厘米厚。

So That's probably a couple inches.

大概半英寸左右。

A half inch.

你

You

知道吗？

know?

头骨大概有半英寸厚。

Half inch thick skull ish.

所以啊，确实挺厚的，咱们这头骨简直像椰子一样结实。

So, yeah, yeah, so that that's a fair bit of like, our our we got quite a coconut going on there.

可不是什么蛋壳那么薄。

It's not it's not like some eggshell.

哦，是啊。

Oh, yeah.

我信你。

I believe you.

所以呢，

So the yeah.

你基本上就是把设备植入进去。

You basically implant the device.

所以你的设备大概是一英寸见方，或者一英寸直径？

And so you would be, like, a one inch square or one inch in diameter?

对。

Yeah.

大概像

Like a

一个一英寸的圆圈。

So an inch circle.

是圆形的吗？

Like a circular?

是的。

Yeah.

我觉得就像一块智能手表那样的大小。

I think like a like a smartwatch or something like that.

或者

Or

好的。

Okay.

对。

Yeah.

行。

Okay.

所以你是说像冰钓那样，一英寸直径的？

So you take this one inch diameter like ice fishing.

对吧？

Right?

你去过冰钓吗？

You ever go ice fishing?

没去过。

No.

但我挺想试试的。

But I'd like to.

这太棒了。

It's great.

对。

Yeah.

特别有意思。

It's really fun.

对。

Yeah.

你基本上会用一个钻孔器，然后在冰面上打个洞。

So you basically take an auger, and you you drill

你用

You take

穿过冰面，没错。

through the surface of the ice Yeah.

然后弄出一个小洞，就可以把钓线放下去了。

And you create a small hole, and then you can dunk your line in there.

所以这就像那样。

So this is like that.

你是在自己的头顶上玩冰钓，然后把它封起来。

You're ice fishing on the top of your skull, and then you cork it.

对。

Yeah.

你把头骨上那一小块一英寸直径的骨头替换成这个Neuralink设备，它内置电池、蓝牙和无线充电功能。

And you replace that, say, one inch diameter piece of skull with this Neuralink device, and that has a battery and a and a Bluetooth and a inductive charger.

然后，不，不是那样。

And then you and and no.

你还需要把电极精准地植入进去，我们开发了一台机器人来完成这个操作，它会非常小心地放置电极，避开任何血管和脑组织。

Then then you you also gotta insert the electrodes so that electrodes are very carefully inserted with with our with a a robot that we developed that look it's, you know, very carefully putting in the electrodes and avoiding, you know, any baines or arteries.

所以整个过程不会造成创伤。

So it's it's you know, it doesn't create trauma.

通过这个一英寸直径的开口，是的。

So through this one inch diameter Yeah.

设备和电极植入后，会自动定位到目标区域

Device, electrodes be inserted, and they will find their way

其实就是一些细小的导线。

Like tiny wires, basically.

细小的导线。

Tiny wires.

细小的

Tiny

导线。

wires.

它们会自动找到大脑的特定区域进行刺激吗？

And they'll find their way to specific areas of the brain to stimulate?

不是。

No.

你得直接把它们精准地放到该放的位置。

You literally put them where where they're supposed to go.

哦，明白了。

Oh, okay.

对。

Yeah.

你刚才说‘插入’

You said insert

这些导线大概有多长？

How long will these wires be?

一般来说，它们会插进去，具体长度要看位置，大概两到三毫米左右。

I mean, they usually go in, like, you know, depending on where it is, like, you know, two or three millimeters.

所以它们自己就能找到正确的位置。

So they just find the spots.

是的。

Yeah.

真厉害。

Wow.

然后呢，

And then yeah.

接着把设备放进去，它就会替代之前取走的那小块头骨。

And then you you put the device in and that that gets that that replaces the the little piece of skull that was taken out.

然后再把伤口缝上，就留下一点点疤痕，完事了。

And then you you you stitch up the the hole and and you just have a little like a little scar, and that's it.

那这个装置是可以更换或者取出来的吧？

Well, this should be replaceable or reversible?

当然可以。

Yes.

比如有人实在受不了了？

Like, if someone can't take it anymore?

我太聪明了。

I'm too smart.

我实在受不了了。

I can't take it.

是的

Yeah.

你现在完全可以取出来。

You can totally take it now.

除了恢复肢体功能、视力和听力这些很棒的效果外，你认为这种技术还能带来什么认知方面的益处吗？

And what is the besides re restoring limb function and eyesight and hearing, which are all amazing, is there there any cognitive benefits that you anticipate from something like those?

是的

Yeah.

当然可以，本质上这是一种通用的解决方案，理论上能修复各种脑损伤，比如你要是有严重的癫痫，它就能直接阻止癫痫发作。

I mean, you could for sure I mean, basically, it's it's a generalized sort of thing for for fixing any kind of brain injury in in in principle, like, you or if you've got, like like, severe epilepsy or something like that, it could it could just it could just sort of stop the epilepsy from occurring.

它能实时检测到异常，然后释放一个反向脉冲，把癫痫掐灭在萌芽状态。

Like, it could detect it in real time and then fire a counter pulse and stop the epilepsy.

实际上，脑损伤的类型非常多。

If I mean, there's there's a whole range of brain injuries.

比如有人中风了，可能会丧失说话能力。

Like, people somebody gets a stroke, they could lose the ability to speak.

这个也能解决。

That that that also that could also be fixed.

比如中风后受损，或者面部某个部位失去肌肉控制之类的情况。

So if you get, like, stroke damage or if you you lose, say, you know, muscle control over a part of your face or something like that.

我觉得，等年纪大了，要是得了阿尔茨海默病之类的，记忆力会衰退，这个技术或许能帮人恢复记忆，诸如此类的效果。

I think and then when when when get old, you tend to if you get, like, you know, Alzheimer's or something like that, then you lose memory, and this could help you with, you know, restoring your memory, that kind of thing.

恢复记忆。

Restoring memory.

那到底是什么机制让它能做到这一点？

And what what is happening that's allowing it to do that?

就是这些细小的导线，没错。

Like, the wires these these small wires Yes.

刺激大脑的这些区域。

Stimulating these areas of the brain.

那么，是这些脑区失去了某种电活动吗？

And then is it that the areas of the brain are they're they're losing some sort of electrical force?

那到底发生了什么？

Like, what what is happening?

对。

Yeah.

是吗？

Yeah?

这就像是很多电路，有些电路断了，我们可以把它们修好。

It's it's it's like it's like think of just like a bunch of circuits, and there's some, like, circuits that are broken, and we can, like, fix those circuits.

这是在替代那些断掉的电路。

It's substitute for those circuit circuits.

所以会有特定频率通过这里？

And so a specific frequency will go through this?

对。

Yeah.

就只是

Just a

具体到这个人身上，得怎么确定该刺激多少、还是太少呢？

Specific her in that would is the process figuring out how much or how little has to be?

这些脑区需要被激活到什么程度？

How how much these areas of the brain have to be juiced up?

是的。

Yeah.

我的意思是，还有很多工作要做。

I mean, there's still a lot of work to do.

我说我们有可能在一年内把这项技术用到人身上，就是这个意思。

So when I say, you know, we got a shot at probably putting it in in in a person in, you know, a within a a year.

我指的就是这个意思。

I think that that's that's what that's exactly what I mean.

我们有机会把信号输入进去，让大脑恢复健康，重新获得一些已经丧失的功能。

Think we we have a chance of putting input into one end, having them having them be healthy and and restoring some functionality that they've they've lost.

担心的是，最终你可能得把人的整个头顶切开，换一个新的顶盖。

The fear is that eventually you're gonna have to cut the whole top of someone's head off and put a new top

当然。

Sure.

如果你想搞个真正的高性能版，也就是脑刺激的P100D，那就得接上一大堆线。

With a whole bunch of wires if you wanna get, you know, the real turbocharged version, the p 100 d of of brain stimulation.

我的意思是，如果你真想实现和人工智能的深度融合，大概就得这么做。

I mean, ultimately, if you if you wanna go with full AI symbiosis, you'll probably wanna do something like that.

说到人工智能，‘融合’这个词听起来挺吓人的。

Symbiosis is a scary word when it comes to AI.

这完全是自愿的。

It's optional.

我。

I

但愿如此。

would hope so.

对。

Yeah.

我的意思是，一旦你习惯了曼哈顿博士那种生活，一旦你成了神，似乎就几乎不可能再回去了。

It's just I mean, once you enjoy the doctor Manhattan lifestyle, once you once you become a god, seems very, very unlikely.

你肯定不想再变回那个笨笨的自己了。

You're gonna wanna go back to being stupid again.

我是说，你完全可以从根本上改变人类彼此交互的方式。

I mean, you literally could fundamentally change the way human beings interface with each other.

没错。

Yes.

对。

Yes.

你都不需要开口说话了。

You wouldn't need to talk.

我既觉得挺害怕的，但同时又特别兴奋。

I'm so scared of that, but so excited about it at the same time.

这会不会很奇怪？

Is that weird?

是的。

Yeah.

我的意思是，我觉得这可能是AI不断进步的其中一条路径，你懂的，AI变得越来越厉害了。

I mean, the the I think this is one of the paths to, you know, like, think, what what are for AI is getting better and better.

那我们就假设一个比较友好的AI场景吧。

So now let's assume it's sort of like a a benign AI scenario.

即便是在这种友好的情况下，我们还是被甩在后面了。

Even in a benign scenario, we're kinda left behind.

我们根本跟不上节奏。

You know, we're we're we're not we're not along for the ride.

我们太落后了。

We're just too dumb.

没错。

Right.

那我们该怎么跟上这趟车呢？

So so so how do you go along for the ride?

所以与其对抗，不如融入其中。

And so you can't beat him, join him.

其实我们某种程度上已经是赛博格了。

So and we're we're already we're already a cyborg to some degree.

对吧？

Right?

你有手机啊。

Because you've got your phone.

你有电脑啊。

You got your laptop.

还有眼镜。

Glasses.

是的。

Yeah.

对。

Yeah.

吉他。

Guitar.

对。

Yeah.

电子设备啊。

Know, electronic devices.

对。

Yeah.

现在呢，要是你出门没带手机，感觉就像少了条胳膊一样。

And, I mean, we're today, if you've your your phone if you if you don't bring your phone along, it's like you have missing limb syndrome.

就是觉得特别不踏实，好像少了什么重要的东西。

That's like you know, feels like something's really, really missing.

所以我们其实已经算是一半的赛博格，或者说是和人工智能共生的生物了。

So we're already partly part you know, partly a cyborg or an AI symbiote, essentially.

只是现在和电子设备之间的数据传输速度还太慢了。

It's just that the data rate to the electronics is slow.

所以特别是输出方面。

So especially output.

你就光靠拇指打字。

Like, you're just going with your thumbs.

我的意思是，你的数据传输速率到底有多少？

I mean, like, what what's your data rate?

乐观点说，大概每秒一百位。

Optimistically, a 100 bits per second.

这已经算很高了。

That's being generous.

可电脑的传输速度能达到每秒上百万兆位，你知道吧？

And and now the computer can can communicate it, like, you know, 100 a 100 terabits.

你明白我的意思吧？

You you know?

所以现在，吉比特级别的速度根本不算什么。

So so, certainly, you know, gigabits are a trivial at this point.

所以这简直就是如此。

So this this is like yeah.

基本上，你的电脑能快上一百万倍来做事情。

Basically, your computer could do a a mill do things a million times faster.

或者到了某个阶段，你跟人工智能交流，就像在跟一棵树说话。

Or or you're you're at a certain point, it's like talk the AI is like talking to a tree.

明白了。

Okay.

这太无聊了。

This is boring.

你可以跟一棵树说话。

You can talk to a tree.

根本没什么意思。

It's very not very entertaining.

所以，如果你能解决数据传输的问题，尤其是输出端，同时兼顾输入端，就能进一步提升人与机器之间已经存在的协同关系。

So so if you if you can solve the the data rate issue and you're especially output but input two, then you can improve the symbiosis that is already occurring between man and machine.

你说不用再互相说话了，那具体是怎么改进的呢？

So you you can improve it in what when you said you won't have to talk to each other anymore.

我们以前常拿这个开玩笑。

We used to joke around about that.

在这个播客里，我早就说过无数次了，总有一天，人们能直接读懂彼此的想法。

I I've joked around about that a million times in this podcast that one day in the future, there's gonna come a time where you can read each other's minds.

到时候，你们能以某种非语言的方式直接交流。

And you'll you'll be able to interface with each other in some sort of a nonverbal Mhmm.

不用动嘴、不用发出声音，就能互相传递信息。

Nonphysical way where you will transfer data back and forth to each other without having to actually use your mouth and make noises.

没错。

Exactly.

比如说，当你想向别人表达一个复杂的想法时，会发生什么？

So when you like, what happens when you when like, let's say you've got some complex idea that you're trying to convey to somebody else.

你该怎么做到这一点呢？

And how how do you do that?

你的大脑在把复杂的想法压缩成语言时，要花很多精力。

Well, your your brain spends a lot of effort compressing a complex concept into words.

把复杂概念压缩成语言的过程中，信息损失是不可避免的。

And there's a there's a lot of a lot of loss or information loss that occurs when compressing a complex concept into words.

然后你把话说出来，听的人再把这些话解码、还原成自己的理解。

And then you say those words, those words are then interpreted, then they're decompressed by the person who is listening.

但到最后，对方对你的意思往往只能得到一个很不完整的印象。

And they they all at best get a very incomplete understanding of what you're trying to convey.

想精确传达一个复杂的想法很难，因为你得经过压缩和解码两个环节。

It's very difficult to convey a complex concept with precision because you've got compression, decompression.

说不定你的话对方都没听全，沟通起来自然就更困难了。

You may not even have heard all the words correctly, and so communication is difficult.

这根本就是沟通失败啊。

You know, what we have here is a failure to communicate.

这挺酷的，卢克。

It's cool, and Luke.

是的。

Yes.

还有一部特别棒的电影。

And there's a great movie.

对。

Yeah.

很棒。

Great

电影。

movie.

还有一个解读的因素在里头。

And there's an interpretation factor too.

比如，同样的词句，你可以有不同理解，这跟语调有关，也跟语气有关。

Like, you can choose to interpret certain series of words in in different ways, and they're dependent upon tone, dependent upon Yeah.

还有社交线索，甚至表情、讽刺这些都会影响。

Social cues, even facial expressions, sarcasm.

影响因素太多了。

There's there's a lot of variables.

讽刺最难把握了。

Sarcasm is difficult.

没错。

Yes.

确实如此。

Yeah.

我曾经说过，也许未来会出现一种由计算机创造的通用语言，年轻人尤其是孩子，学起来会特别快。

And so one of one of the things that I I've said is, like, that there could be potentially a universal language that's created through computers that, particularly young kids, would pick up very quickly.

我孩子整天玩TikTok，各种乱七八糟的东西，我根本看不懂他们在干嘛。

Like, my kids do TikTok and all this jazz, and I I don't know what they're doing.

但他们就是知道该怎么操作。

They just know how to do it.

而且学得特别快。

And they know how to do it really quickly.

他们学得特别快，还会教我怎么剪辑。

Like, they learn really quickly, and they show me how to edit things.

而且

And

对。

Yeah.

如果你从一年级就开始教孩子一种全新的通用语言，就像一个能解读思维的罗塞塔石碑，让你表达想法时完全不会产生歧义。

It's if you taught a child from first grade on how to use some new universal language, I mean, essentially, like a a Rosetta Stone and something that's done with that interprets your thoughts, and you can convey your thoughts with no room for interpretation Mhmm.

彼此的意思清清楚楚，你想说什么，对方一听就懂，根本不需要那些语气词。

With clear very clear that where you know what a person's saying, and you can tell them what you're saying, and there's no need for noises.

不用任何语气词。

No need for mouth noises.

也不用

No need for

是的。

Yes.

我们逐渐形成了一些被普遍接受的发声方式，大家也都默认了。

These sort of accepted ways that we've sort of evolved to make sounds that we all agree.

通过我们文化中的共同理解

We through our cultural dictionary

没错。

and Right.

其实我们可以绕过这一切。

We agree or certain we we could bypass all that.

是的。

Yeah.

但为了好玩，我们还是可以继续用吗？

We could still do it for For fun?

出于情感上的原因。

Sentimental reasons.

对。

Right.

就像篝火一样。

Like campfires.

是的。

Yeah.

没错。

Exactly.

我不需要篝火。

I don't need campfires.

我的意思是，我不非得烤棉花糖。

I mean, I don't need to roast marshmallows.

挺有意思的。

It's kinda fun.

对。

Right.

所以嘛。

So yeah.

是的。

Yeah.

我觉得原则上，你能够非常迅速且精准地传递想法。

I think you would in principle, you would be able to communicate very quickly and with far more precision ideas.

语言会变成什么样，我不太确定，但像这种情况，你可能就像《黑客帝国》里那样，直接就能传输信息。

And and language would I'm not sure what would happen to language, but you you could probably with a situation like this that you would be able to just you'll be kinda like the matrix.

你想用另一种语言来解决一个问题。

You you wanna speak a different language in a problem.

对。

Right.

所以我才下载了那个程序。

That's why I was downloaded the program.

对。

Right.

所以至少在最初的几次迭代中，我们就能做到这样——我知道谷歌的一些Pixel Buds已经能实时翻译语言了。

So at least for the first iterations first few iterations, we'll just be able to like, I I know that Google has their some of their pixel buds have the ability to interpret languages in real time.

当然。

Sure.

对。

Yeah.

你能听到对方说话，它会根据你选择的语言把内容播放给你听。

You can hear it, and they'll it it'll play things back to you in whatever language you choose.

所以大概就是类似这样的功能。

So it'd be something along those

这种方向，是的。

lines Yeah.

在最初的几次迭代中？

For the first few iterations?

最初的几次迭代，我的意思是，我讲的是长远来看，经过大量开发之后的情况。

Well, the first few iterations are I mean, what I'm talking about is, like, in the limit over time, you know, with with a lot of development.

在最初的几个版本里，我们真正想解决的其实是脑损伤问题。

The first few iterations really, in the first few versions, all we're gonna be trying to is is solve brain injuries.

所以啊，你不用担心这事会突然冒出来吓你一跳。

So so it's like, don't don't don't worry that that's not gonna sneak up on you.

这事儿得花上一段时间。

This this this will take a while.

要过多久你才不用开口说话？

How many years before you don't have to talk?

如果研发进度持续加快，大概五到十年吧。

If the if the if the development continues to accelerate, then maybe, like, five years, five to ten years.

这么快？

That's quick.

真的很快。

That's really quick.

这还远不是最好的情况。

That's not the best case scenario.

五年后就完全不用说话了。

No talking anymore in five years.

最好的情况。

Best case scenario.

但话说回来，十年后人们可不会喜欢那样。

But, man, ten's ten years won't like it.

我一直推测，外星人可能是我们未来的模样，你看他们脑袋那么大，肌肉却极少，连嘴都用不着了。

I've always speculated that aliens could potentially be us in the future because if you look at, like, the size of their heads and the fact that they have very little muscle and they're they're they don't use their mouth anymore.

他们就剩这么点小嘴，我的意思是，你想想《第三类接触》里那种经典外星人形象，从南方古猿或者远古人类进化到我们，差别在哪？

They have this tiny little I mean, the archetypal alien that you see in, like, close encounters of the third kind, they they're like if you went from, like, Australopithecus or ancient Hominid to us, what's the difference?

毛少了，肌肉弱了，脑袋变大了。

Less hair, less muscle, bigger head.

然后你就继续这么发展下去。

And then you just keep going.

一千年后，一百万年后，或者就五年后，等Neuralink上线，我们就会慢慢适应这种全新的存在方式，对。

A thousand, a million, whatever you or five years, whatever whatever happens when Neuralink goes on online, and then we slowly start to adapt to this new way of being Mhmm.

到时候，我们连肌肉都不用了。

Where we don't use our muscles anymore.

我们的脑袋变得特别大。

We have this gigantic head.

我们甚至不用说话就能交流。

We can talk without words.

你还可以保存状态。

You you could also save state.

保存状态？

Save state?

对，保存状态。

Save state.

就像把大脑的运行状态存下来，跟电子游戏里存档一样。

Like, save your brain state, like like a saved game in a video game.

哇哦。

Woah.

比如说，你想从 Windows 95 切换到

Like like, if you wanna swap from Windows 95

嗯，

Well

比如说你啊，

to like, you

可能比那还要强一点。

Probably a little better than than that.

不过也是。

But yeah.

我觉得我们现在就像Windows 95。

I think we are Windows 95 right now.

对。

Yeah.

从未来的角度看，大概吧。

From a future perspective, probably.

没错。

Yeah.

不过呢。

But yeah.

我的意思是，从理论上讲，未来你完全可以把意识状态保存下来，再还原到一个生物体里。

I mean, you you could save state and restore that state into a biological being if you if you wanted to in the future in principle.

从物理学的角度来看，根本没有任何障碍。

It's like nothing, like, from a physics standpoint that prevents us.

当然，你可能会有点不一样，但其实你每天早上醒来，跟昨天的自己也已经不一样了，这很正常。

Now you you'd be a little different, then but you're also a little different when you wake up in the morning from yesterday, and you're a little different.

事实上，你五年前和现在的自己，差别可大了。

In fact, if you say, like, you five years ago versus you today is quite a big difference.

没错。

Yes.

所以你还是那个你。

So you'd be substantially you.

我的意思是，你肯定会觉得自己就是你。

I mean, you'd be you would you'd certainly think you're you.

但把你自己保存下来，再转化成某种生物状态，比如能跟三十岁的自己待在一起，这个想法听起来怎么样？

But the idea of saving yourself and then transforming that into some sort of a biological state, like, could hang out with 30 year old you?

我的意思是，可能性是无穷无尽的。

I mean, the possibilities are endless.

这也太奇怪了。

That's so weird.

我的意思是，你就想想你的手机能录像这个功能。

I mean, just think think like how your phone can you can record videos on your phone.

对。

Mhmm.

比如，你不可能像手机或相机那样，把视频记得那么清楚，对吧。

Like, there's no no way you could remember a video Right.

毕竟，手机或相机能记录得那么准确。

As accurately as your phone or a camera, you know, could.

所以，如果你以后用了比如 Neuralink 10 代之类的设备，在遥远的未来，你就能像看电影一样，把一切都回忆起来。

So now if if you've got, like, a you know, some some, you know, version 10, Neuralink, whatever, and far in the future, you you could you could you could recall everything, but just like it's a movie.

因为

Because

它包含了所有的感官体验。

it's Including all the entire sensory experience.

情绪。

Emotions.

所有东西。

Everything.

一切。

Everything.

一切。

Everything.

然后还能回放。

And play it back.

而且你觉得自己还能剪辑它。

And and and you think you can edit it.

改一改。

Edit it.

对。

Yeah.

所以你能改变你的过去吗？

So you can change your past?

你可以改变你认为的过去是什么样子。

You could change what do you think was your past.

是的。

Yeah.

如果你现在经历了一些创伤性的事件，

Well, so if you had, like, traumatic experience now

那可能只是记忆在重放。

could be a replayed memory.

有可能。

It could be.

是的。

Yeah.

有可能。

It may be.

这会是记忆重现的概率有多大？

What's the odds of this being a replayed memory?

如果你

If you

必须猜的话。

had to guess.

超过五成。

It's more than 50%.

这里没法准确估算概率。

There's no way to assign a probability with accuracy here.

对。

Right.

不过大概来说。

But roughly.

如果你只是凭直觉的话。

If you just had a just gut instinct.

嗯，我脑子里没有神经链接，所以我现在说是0%。

Well, I don't have a neural link in my brain, so I'd say right now 0%.

但当你真正拥有神经链接的时候，这个概率就会超过0%。

But at the point at which you do have a neural link, then it rises above 0%.

就算我们经历的是某种被保存下来的记忆，这种想法其实并不让人安心，对吧？

The idea that we're experiencing some sort of a preserved memory is even though it's still the same, it's not comforting.

对吧？

Right?

不知怎么的，人们一谈到模拟理论，聊的都是“我们现在可能正活在模拟里”这种可能性。

For some reason, when we people talk about simulation theory, they talk about the potential for this currently being a simulation.

哪怕你生活美满、正在恋爱、热爱自己的工作、还有很棒的朋友，但一想到这一切其实并没有物质实体，没法去敲一敲、摸一摸，心里还是觉得不踏实。

It even though your life might be wonderful, you might be in love, you might love your career, you might have great friends, but it's not comforting to know that this experience somehow or another doesn't exist in a material form that you can knock on.

感觉这么真实，对吧？

Feels real, doesn't it?

确实很真实。

Feels real.

但问题是，明明它可能是假的，这想法不知怎么的反而让人不安。

But but if it's not but the idea that it's not is for some strange reason disconcerting.

是啊，这确实该让人不安，因为如果这一切都不是真的，那什么才是真的？

Well, yeah, I'm sure it should be disconcerting because then if if this is not real, what is?

没错。

Right.

还有那个老掉牙的思想实验，比如你怎么确定自己不是个泡在营养液里的大脑？

But but the you know, there's that that old sort of thought experiment of, like, how do you know you're not a brain in a vat?

你懂的。

You know?

但关键是这么一回事。

And now here here's the thing.

你就是个大脑泡在罐子里，而那层脂肪就是你的头骨。

You are a brain in a vat than that fat is your skull.

对。

Yes.

你所看到、感受到、听到的一切，所有感官体验，都只是电信号。

And everything you see, feel, hear, everything, all your senses are are electrical signals.

全部都是。

Everything.

所有的一切，都是给泡在罐子里的大脑传递的电信号，而这个罐子就是你的头骨。

Everything is an electrical signal to a to a brain in a vat where the vat is your skull.

你体内的所有激素、所有神经递质，这些东西都是药物。

And all your hormones, all your neurotransmitters, all these things are drugs.

肾上腺素就是一种药物。

Adrenaline's a drug.

多巴胺也是一种药物。

Dopamine's a drug.

你就是一个药物工厂。

You're a drug factory.

你一直在用爱、催产素和美感改变自己的状态。

You're constantly changing your state with love and oxytocin and and beauty.

没错。

Sure.

改变你的状态。

Changes your state.

好音乐能改变你的心境。

Great music changes your state.

完全正确。

Absolutely.

但这里还有一个有意思的观点，就是你说，意识到底是从哪儿来的？

And yet, here's another sort of interesting idea, which is because you say, like, where where did consciousness arise?

如果你相信物理学的基本观点是成立的，那宇宙最初只是由夸克和轻子组成，很快便形成了氢、氦、锂，也就是元素周期表里的那些基本元素。

Well, assuming you believe the belief in in physics, appears to be true, then, you know, we the universe started off as basically quarks and leptons and quickly became hydrogen and helium, lithium, like, basically elements of the periodic table.

但那时候主要就是氢，基本上。

But it was, know, like, mostly hydrogen, basically.

然后经过很长一段时间，大概一百三十八亿年后，这些氢竟然变得有意识了。

And then and then over a long period of time, you know, 13,800,000,000 later, that hydrogen became sentient.

那么问题来了，意识到底是在哪个环节出现的？从什么时候开始，氢变成了有意识的东西？

Where so where along the way did where is consciousness what's the line of consciousness and not consciousness

没错。

Right.

从氢到现在的我们，中间的分界线到底在哪？

Between hydrogen and here?

没错。

Right.

那我们到底该在什么时候定义它？

When do we call it?

什么时候才能说，这就是意识了？

When do we call it consciousness?

今天我看了个视频，就是我们之前在播客里放过的，一只猴子骑着摩托车在街上跑，然后跳下车，想偷个婴儿。

I was watching a video today that we played on a podcast earlier of a monkey riding a motorcycle down the street, jumps off the motorcycle, and tries to steal a baby.

是的。

Yeah.

我看过那个。

Saw that one.

那时候没火起来。

They weren't viral.

那只猴子有意识吗？

What is that monkey conscious?

看起来它确实有。

It seems like it is.

它好像早有打算。

It seems like it had a plan.

它居然骑着摩托车，还跳下来想偷孩子。

It was riding a fucking motorcycle and then jumped off the motorcycle to try to steal a baby.

看起来挺明显的

Seems pretty

就是那个把婴儿拖了挺远的那次。

The one that just dragged baby down the street pretty far.

是的

Yeah.

对

Yeah.

明显是有意识的。

Seems pretty conscious.

对吧？

Right?

那里确实有某种程度的意识。

There's definitely some degree of consciousness there.

是的

Yeah.

这可不是普通的虫子。

It's not like it's not a worm.

确实不是。

It's not

它已经处在另一个层次了。

Seems to be on another level.

对。

Yeah.

而且它还会继续发展。

And it's gonna keep going.

当人们思考人类未来可能的形态时，尤其是考虑到与人工智能的共生关系，那种变化将让人完全认不出来。

And that that's the real concern when when people think about the potential future versions of human beings, especially when you consider symbiotic relationship to artificial intelligence, it will be unrecognizable.

总有一天，我们会和现在这个状态相差太远。

That one day, we'll be so far removed from what this is.

那时我们回看现在，就像现在回看那些我们祖先所由来的简单生物一样。

We'll look back on this the way we look back now on, you know, simple simple organisms that we evolved from.

再过不久，我们就会以这样的视角回望现在了。

And then it won't be that far in the future that we do have this this view back.

我希望意识能延续到未来，变得越来越复杂、精妙，并且懂得提出关于宇宙的深刻问题。

Well, I hope consciousness propagates into the future and gets more more sophisticated and complex and and that understands the questions to ask about the universe.

你觉得真是这样吗？

Do you think that's the case?

作为一个人，作为你自己，你显然在有意识地努力成为更好的版本。

As as a human being, as yourself, you're clearly trying to make conscious decisions to be a better version of you.

对吧？

Right?

这就像一种理念：放下身外之物，意识到自己并不喜欢某些东西。

This is the idea of, like, getting rid of your possessions and realizing that you're trying to like, I don't like this.

我会试着去改进它。

I will try to improve this.

我会努力以更好的方式去与现实互动。

I will try to do a better version of the way I interface with reality.

只要你朝着提升自己的方向走，事情本来就是这样。

That this is always the way things are if you're if you're moving in a a some sort of a direction where you're trying to improve things.

你总会进入一个新阶段，回头看过去，会觉得那时候自己做错了。

You're always gonna move into this new place where you look back in the old place and go, I was doing it wrong back then.

所以这只不过是那种情况的加速版。

So this is an accelerated version of that.

一种特别快的加速版。

A super accelerated version of that.

你并不总是能进步，但你可以追求进步。

I mean, you don't always improve, but you can aspire to improve.

你可以追求少犯错。

You can aspire to be less wrong.

对。

Yeah.

我觉得，物理学的那些工具确实非常强大。

This is like, I think a good the tools of physics are very powerful.

不如直接假设自己是错的，目标就是尽量少犯错。

Like, just assume you're wrong and you're asked for your goal to be less wrong.

我不觉得你每天都能成功地少犯错，但如果你大多数时候都能做到，那已经很棒了。

I don't think you're gonna if if succeed every day in being less wrong, but, you know, if you're gonna succeed in being less wrong, most of the time, you're doing great.

这话说得真好。

That's a great way of putting it.

努力让自己少犯错。

Aspire to be less wrong.

对。

Mhmm.

不过，当人们怀旧地回想过去 simpler 的时光时，这种情况也存在。

But then when, you know, people look back in nostalgia about simpler times, there's that too.

回想起篝火晚会，确实很浪漫，也很令人向往。

It's very romantic and exciting to look back on campfires.

但你依然可以点起一堆篝火。

But you can still have a campfire.

是的。

Yes.

对。

Yeah.

但当你成为一个超级极客，把大脑接入网络，头顶装上脑机接口，与全宇宙通用的交流语言相连时，你就会真正体会到它的价值。

But you will you appreciate it when you're a super nerd, when you connect it to the grid, and you have some skullcap in in place of the top of your head, and it's interfacing with the inter international language that the rest of the universe now enjoys communication with people.

而且我们。

And we're

对。

Yeah.

当然。

Sure.

我觉得是这样。

I I think so.

对。

Yeah.