一个下午见证十亿年的进化——乔治·丘奇

所以，我认为这不会完全改变指数级增长。

So, I think but that's not going to completely change the exponential.

可能会从十年缩短到一年，这是我们目前针对COVID疫苗的记录。

It might reduce it from ten years down to one year is our record so far for say COVID vaccines.

所以，也许时间会缩短10倍。

So, maybe that'll be 10 times shorter.

也许这会产生一些倍增效应。

Maybe that will multiply out a little bit.

但我认为最重要的是，我们所有的设计都会变得更好。

But I think the big thing is that all our designs will become better.

因此，失败案例将会减少。

So, there'll be fewer failures.

每种药物的成本将会下降。

The cost per drug will drop.

还会出现一些我们传统上不认为是药物或器械的东西，某种混合体。

There'll be things that we didn't classically consider drugs or instruments, kind of some sort of hybrid thing.

但话说回来，我并不认为这会完全出人意料。

But again, I don't think that'll be completely shocking.

只是这类东西会变得非常多。

But it's just going to be so much of it.

你知道，解决方案将会呈现极大的多样性。

You know, it's going be lots of diversity of solutions.

具体会增加多少呢？

How much more are we talking?

药物数量会增加到10倍还是100倍？

Are we going have 10x the amount of drugs, 100x?

我甚至不确定这种量化是否合理。

I'm not even sure it's going to make sense.

不过确实，增长100倍也不会完全令人惊讶。

But yeah, 100x would not be completely surprising.

药物组合将会变得非常重要。

Combinations of drugs will be important.

你运用得很聪明。

You're using them intelligently.

未来还会有更多。

There'll be a lot more.

某些药物会影响一切。

Some drugs will affect everything.

比如，一种针对年龄的药物可能影响所有疾病。

So, for example, an age related drug that could impact every disease.

我认为重要的不是数量，而是质量、影响力、交叉领域以及能帮助医生和普通民众做决策的软件。

I'm not sure the number is going to matter so much as the quality and the impact and intersection and software that helps physicians and regular citizens make decisions.

而且

And

具体是什么变化在推动这一切？

what specifically is changing that's enabling this?

是现有成本曲线持续走低，还是会出现新技术或工具？

Is it just existing cost curves continuing or is it some new technique or tool that will come about?

成本曲线受新工具影响。

Well, the cost curves are affected by new tools.

我的意思是，这不是自动发生的。

I mean, it's not just some automatic thing.

桑格测序与纳米孔/荧光二代测序之间存在巨大断层。

There was a big discontinuity between Sanger sequencing and nanopores and fluorescent next gen sequencing.

我认为这其实是两件事的融合。

And you know, I think, sometimes so, it's a merger of two things.

很明显，AI与蛋白质设计的融合引发了一个阶跃函数。

So clearly, AI merging with protein design caused a step function.

这些阶跃函数会被平滑处理成某种指数曲线，但数量非常庞大。

These step functions get smoothed out into a kind of a smooth exponential, but there are lots of them.

下一阶段很可能是AI与其他生物学领域的融合，比如发育生物学，将发育生物学与制造业结合，甚至攻克发育生物学——换句话说，真正掌握如何以DNA为编程材料制造任意形状。

The next set will probably be, yeah, a merger of AI with other aspects of biology like developmental biology, merger of developmental biology with manufacturing and, you know, conquering developmental biology, in other words, actually knowing how to make any arbitrary shape given DNA as the programming material.

我认为这将是个重大突破。

I think that would be a big thing.

总的来说，我们需要更多材料——机械、电气工程中使用的所有材料都应该通过生物技术进行改良。

Having just more materials in general, all the materials that we use in mechanical, electrical engineering should be made better by biotechnologies.

为什么呢？

Why is that?

为什么会这样？

Why is that?

嗯，因为电子学领域，你知道的，摩尔定律虽然不能说完全停止，但可以说我们所谓的1纳米工艺，根据路线图预计在2027年实现。

Well, it's that electronics is, you know, Moore's Law, I wouldn't say is stopping, but it's kind of the what we would call the one nanometer process, which is supposed to come out in 2027 according to the roadmap.

其实并不是真正的1纳米。

It's not really one nanometer.

更像是40纳米的中心到中心间距，你知道的，通常是二维的。

It's more like 40 nanometers center to center spacing, you know, typically in two dimensions.

可能还有点三维结构。

Maybe a little bit of three dimensions.

但生物学已经达到了0.4纳米的分辨率，而且是三维的。

But biology is already at 0.4 nanometer resolution and it is in three dimensions.

所以，你知道，根据你如何计算第三维度，生物学的密度可能已经高出十亿倍。

And so, you know, depending on how you count that third dimension, that could be a billion times higher density that biology is already at.

我们只需要再多一点练习来应对整个元素周期表。

And we just need a little more practice with dealing with the whole periodic table.

即使是讲座工程通常也不会使用整个元素周期表，尤其是在原子层面。

Even lecture engineering doesn't use the whole periodic table typically, especially not at the atomic level.

我认为生物学在实现原子级精度方面确实非常出色。

I think biology is just really good at doing atomic precision.

那么过去几十年我们拥有接近原子级别半导体制造技术的原因是什么？

So then what's the reason that over the last many decades and we have we do have, not atomic, but close to atomic level manufacturing with semiconductors.

40纳米。

40 nanometers.

对。

Right.

这已经相当小了。

It's quite small.

线性尺寸上比生物学大1000倍。

It's 1,000 times bigger than biology, linearly.

但迄今为止我们取得的进展与生物学无关。

But the progress we have made hasn't been related to biology so far.

看起来我们实现了摩尔定律。

It seems like we've made Moore's Law happen.

我不知道。

I don't know.

九十年代的人们常说，最终我们会拥有这些进行计算的生物机器。

People in the nineties were saying, you know, ultimately, we'll have these bio machines that are, doing the computing.

但看起来我们一直在使用传统的制造工艺。

But it seems like we've just been using conventional, manufacturing processes.

究竟是什么变化让我们能够利用生物技术制造这些东西？

What exactly is it that changes that allows us to use bio to make these things?

有几个关键因素。

There are few things.

其一是合成生物学的出现——虽然我们之前已经在某种程度上实践着合成生物学。

One is the arrival of synthetic biology, where you sort of we were already kind of doing synthetic biology before.

比如我们早就在进行重组DNA研究。

You know, we were doing recombinant DNA.

可以说基因工程原本就是朝着这个方向发展的。

Was kind of, you know, genetic engineering was called was kind of in that direction.

但合成生物学真正解放了我们的思维，让我们能构想更宏大的可能。

But synthetic biology really liberated us to think a little bit bigger.

尽管最初研究集中在大肠杆菌和酵母上，

Even though it started kind of focused on E.

但它让我们开始思考例如新型氨基酸的应用。

Coli and yeast, it enabled us to maybe think about new amino acids, for example.

我认为如果能利用整个元素周期表开发新型氨基酸，或是探索氨基酸的催化潜力，就能突破重要壁垒之一。

And I think new amino if you start using the full periodic table with the amino acids or what amino acids can catalyze, that breaks one of the major barriers.

横亘在电子机械工程与生物学之间的主要壁垒之一，就是特殊材料的使用——那些以光速传导电信号的材料。但生物学完全能制造出具有光速传导性能的聚合物。

One of the major barriers between, you know, electrical and mechanical engineering and biology was the use of, you know, special materials, things that conduct electricity at the speed of light or conduct signals more But there's definitely polymers that biology can make that will conduct at the speed of light.

要知道，我们可以创建一个混合神经元系统，既包含传统神经元，又能以光速进行信息传导，这会非常有趣。

And, you know, we could make a mixed neuronal system that has conventional neurons and processes that conduct at the speed of light, that would be interesting.

我认为我们在蛋白质设计方面的能力尤其具有挑战性。

I think that our ability to design proteins was particularly difficult.

核酸设计非常棒，比如当你想让两个分子相互结合时，只需按照沃森-克里克规则进行配对操作。

Designing nucleic acids was great whether we were doing, you know, you want two things to bind to each other, you just dial it up using Watson Crick rules.

如果你想构建三维结构，实际上这是少数几种形态由相当简单规则决定的领域之一。

If you want to make a three-dimensional structure, you know, it's actually the one, kind of the one thing where morphology is dictated by fairly simple rules.

这与发育生物学的工作原理不同，我们仍需探索其中的机制。

It's not how developmental biology works, we still need to figure out how that works.

但DNA折纸术、DNA纳米结构确实行之有效。

But DNA origami, DNA nanostructures really work.

但对蛋白质来说，这真的非常困难，直到大约八年前才有所突破。

But doing it for proteins was really, really hard until, I don't know, maybe eight years ago, something like that.

我想我们现在才刚刚开始适应它。

And I think we're just now getting used to it.

使用芯片来制造DNA。

The use of chips for making DNA.

我是说，你提到DNA合成成本下降了上千倍。

I mean, you said that DNA synthesis come down a thousand fold.

嗯，这取决于你跟谁聊。

Well, it depends on who you talk to.

所以当我们在2004年的《自然》论文中首次提出基于芯片的基因时，基本上人们对此嗤之以鼻了大约十年。

So when we came out with the first chip based genes in 2004 Nature paper, Basically, people dismissed it for about a decade.

使用它的只有合作者和校友。

The only people that used it were collaborators and alumni.

尽管它便宜了上千倍，却甚至没有被列入DNA合成的摩尔定律曲线。

And it wasn't even listed on the Moore's Law curve for DNA synthesis, even though it was like a thousand times cheaper.

它就像被完全忽视了。

It was just like ignored.

而现在我们宣称能制造10的17次方个基因，你可以构建10的17次方规模的文库，这些并非随机生成——不同于传统意义上仅通过易错PCR和掺入核苷酸的方法。

And now we have claims of 10 to the seventeenth genes, okay, that you can make libraries 10 to the seventeenth that aren't randomized in any real, in the usual sense where you just like do air prone PCRs, spiked in nucleotides.

要知道，10的17次方可比一千倍大得多——如果这真能实现的话。

Now, 10 to the seventeenth, that's a lot bigger than a thousand fold, you know, if it turns out to be practical.

公开可用的数据即将耗尽。

Publicly available data is running out.

因此，主要的人工智能实验室与Scale合作，以突破可能的界限。

So major AI labs partner with Scale to push the boundaries of what's possible.

通过Scale的数据工厂，主要实验室能够获取高质量数据，为训练后阶段提供支持，包括高级推理能力。

Through Scale's data foundry, major labs get access to high quality data to fuel post training, including advanced reasoning capabilities.

Scale的研究团队Seal正在通过实用的AI安全框架和围绕安全与对齐的公共排行榜，为将先进AI融入社会奠定基础。

Scale's research team Seal is creating the foundations for integrating advanced AI into society through practical AI safety frameworks and public leaderboards around safety and alignment.

他们最新的排行榜包括《人类终极考试》、《谜题评估》、《多重挑战》和《远景》，这些测试涵盖了从专家级推理到多模态谜题解决，再到多轮对话表现等一系列能力。

Their latest leaderboards include Humanities Last Exam, Enigma Eval, Multi Challenge, and Vista, which test a range of capabilities from expert level reasoning to multimodal puzzle solving to performance on multi churn conversations.

Scale还刚刚发布了Scale评估系统，帮助诊断模型的局限性。

Scale also just released Scale Evaluation, which helps diagnose model limitations.

领先的前沿模型开发者依赖Scale评估系统来提升他们最佳模型的推理能力。

Leading frontier model developers rely on Scale Evaluation to improve the reasoning capabilities of their best models.

如果你是AI研究员或工程师，想了解Scale的数据工厂和研究实验室如何帮你突破现有能力边界，请访问scale.com/dwarcache。

If you're an AI researcher or engineer, and you wanna learn more about how Scale's Data Foundry and Research Lab can help you go beyond the current frontier of capabilities, go to scale.com/dwarcache.

好的。

Okay.

说到蛋白质设计，90年代人们还设想过其他可能性，比如当时就有人撰写关于纳米技术的文章，像埃里克·德雷克斯勒等人。

So speaking of protein design, another thing you could have thought in the nineties is I mean, people were writing about nanotechnology, Eric Drexler, and so forth.

如今我们已经能根据所需功能逆向推导出赋予该功能的分子序列。

And now we have we can go from a function that we want this, tiny molecular machine to do back to the sequence that can give it that function.

为什么这没有引发纳米技术革命？

Why isn't this resulting in some nanotech revolution?

还是说终将引发？

Or will it eventually?

为什么AlphaFold没能促成这场革命？

Why didn't AlphaFold cause that?

我认为部分原因是

I think part of

最初的纳米技术概念——你知道，作为灵感来源的埃里克·德雷克斯勒，他某种意义上想重构生物学体系，但生物学本就存在。

it is that the nanotechnology as original, you know, the kind of the source of the inspiration, Eric Drexler, he wanted to reinvent biology in a certain sense, but it already existed.

既然DNA复制机制已经存在，就没必要再设计钻石复制器了。

And so you don't need to design a diamond replicator because you already have a DNA replicator.

那么问题在于：当时缺失的是什么？

And so the question, what was missing?

重构生物学体系的动机究竟是什么？

What was motivating this reinvention of biology?

是材料的问题。

It was materials.

所以，生物学在处理超导体、导体周期、半导体和光速等材料时并不那么擅长。

So, the biology is not that great with materials that are say superconductors or conductors periods, semiconductors and light speed.

但它正在进步。

But it's getting there.

我的意思是，与其走必须基于第一性原理纳米结构的路线，不如折中一下，让生物学能够构建东西。

I mean, you know, rather than going the route of having everything has to be based on first principle nanostructures, you meet in the middle where biology can build things.

当然，当你深入到液氮或更低的温度时，就我们目前所知，生物学会停止运作。

Now, of course, when you go down to, you know, liquid nitrogen in colder temperatures, biology, as we currently know it, stops functioning.

这并不是说你不能让物体在液氮中移动。

Now, it's not to say that you can't have things moving in liquid nitrogen.

你可以做到。

You can.

但这尚未被探索，也并非真正必要。

But that hasn't been explored and doesn't really need to be.

因为如果生物学能构建出在低温下运作的东西，或者现在的生物学由于你能建立庞大的生物库，比如体外培养的10^17个样本。

Because if biology can build things that can operate at low temperature or maybe biology now because you can make these big libraries of biology, you know, maybe 10 or the seventeenth in vitro.

你可以快速翻阅它们。

And you can flip through them quickly.

你可以给它们打上条形码。

And you can barcode them.

这是电子学领域从未实现过的。

And you can this is something that's never been done in electronics.

我不是说电子领域做不到，但你们确实还没造出，比如说，十亿种不同的电子材料，对吧？

I'm not saying you can't do it in electronics, but you you haven't made, you know, a billion different kinds of electronic materials, right?

只需要，比如说，一个下午，给它们都打上条形码然后看看哪个胜出，对吧？

Just, know, in an afternoon, barcode them all and see who wins, right?

但我们现在在生物学领域经常这么做，至少从2004年就开始了。

But we do that all the time in biology now, at least since 2004 we have.

所以我认为这是个机遇，我们可以利用这些库来制造更优质的材料。

And so I think that's an opportunity is that we use those libraries to make much superior materials.

甚至最终可能通过这种方式获得室温超导体。

And we might even finally get a room temperature superconductor that way.

从生物领域？

From bio?

有可能的。

It's possible.

我是说，材料库。

I mean, libraries.

我们称之为化学生物奇特材料库。

We call it chemicalbiochemical exotic material libraries.

但关键在于它们都是材料库。

But the point is they're libraries.

从某种意义上说它们本质上基于聚合物，尽管其中部分不一定非得是聚合物。

They're essentially based in some sense on polymers, even though pieces of them don't necessarily have to be polymers.

您对材料科学革命何时会发生有预测吗？

Do you have a prediction by when we'll see this material science revolution.

现在有了AlphaFold，基本上还缺什么？

What is basically standing between because we've got AlphaFold right now, right?

我们需要的是什么？

What is the thing that we need?

我们需要更多数据吗？

Do we need more data?

AlphaFold确实很棒，但这只是其中一部分。

Well, AlphaFold's very nice, but it's only part of it.

有些大型语言模型与AlphaFold不同。

So there are large language models that are different from alpha.

举个例子，上次我查看AlphaFold时，至少它已经完全改变了。

So give an example, alpha fold, last time I checked, anyway, at least it's all changed.

如果你在丝氨酸蛋白酶中用丙氨酸替代丝氨酸，它仍会保持完全正确的折叠结构。

If you substitute an alanine for a serine in a serine protease, it will have exactly the right fold.

整体平均精度会达到几分之一埃的级别，但它将失去功能。

It will be precise to a fraction of an angstrom overall average, but it won't function.

它就是无法正常工作。

It just won't function.

这就需要极高的精确度，或者通过进化知识或实验数据来确认丙氨酸替代是不可行的。

And that's where need either extraordinary precision or just knowledge of what happens evolutionarily or happens in experiments to say that no alanine won't work.

明白吗？

Okay?

因此，我认为各种AI工具的组合能让你对此有更深入的理解。

And so, I think there's all kinds of combinations of AI tools that can give you deeper insight into that.

如果AlphaFold预测结构并不能告诉你这个东西是否真的能运作，那么在我说'我想要一个能做X事的纳米机器'或'我想要一种能做Y事的材料'之前，还需要什么条件？

If alpha fold predicting the structure doesn't tell you whether the thing will actually function, then what is needed before I can say, I want a nanomachine that does X thing or I want a material that does Y thing.

是啊。

Yeah.

我就能直接得到它。

And I can just like get that.

我认为目前的工作方式虽然能让我们走得很远，但还不足以完全解决问题——我们先做出一个勉强能用的东西，然后以此为灵感建立变体库。

I mean, I think the way that it's working now, which will get us a long way, won't get us the whole way, is we make is we have something that kind of works and we make libraries inspired by that, make variations on it.

然后在这些变体中，对能运作的版本继续创造新变体。

And then whichever of those variations work, we make variations on that.

这就像进化过程，只不过现在我们能以惊人的速度进行迭代。

And we can just keep It's kind of like the way evolution worked, except now we can do it at incredibly high speeds.

要知道，原则上进化可能需要百万年才能实现几个碱基对的变化。

And in principle, you know, what might you know, evolution might incorporate a few base pair changes in a million years.

而现在我们一个下午就能完成数十亿次变异。

Now we can make, you know, billions of changes in an afternoon.

而且整个过程都经过优化设计，避免了大量中性突变和致死突变的浪费。

And and it's all guided in such a way that you get rid of the wastefulness of having a bunch of neutral mutations and a bunch of, lethal mutations.

我们可以重点关注那些看似中性但可能带来突破的变异。

You can have things that are quasi neutral but likely to be game changing, have more of a focus on those.

另一个尚未解决的难题——据我所知目前所有AI蛋白质设计工具都不擅长处理——就是我们正在尝试改进的非标准氨基酸问题。

Another thing that's been missing, and none of the AI protein design tools that I know of are particularly good at it yet, but we're trying to, as we speak, trying to improve this, is non standard amino acids.

因为现有工具都依赖于使用20种标准氨基酸的三维结构库，以及基于这20种氨基酸序列训练的大语言模型。

Because a lot of these tools depend on having libraries of three d structures, which use 20 amino acids and large language models that where you line up all the sequences of 20 amino acids.

我们对额外的氨基酸经验非常有限。

And we have very little experience with extra ones.

但我认为正在发生一场革命，能够生成非标准氨基酸，这些氨基酸既可以作为共价部分，也可以轻松配位整个周期表中的稳定元素。

But I think there's a revolution going on in generating nonstandard amino acids that can where the amino acids can either have as part covalent part of them or as easily liganded all the entire periodic table, stable elements.

要知道，每一种都需要融入并训练我们的模型。

That will, you know, each of those will have to blend in and train our models on.

但一旦实现，我们将获得一系列全新材料。最终，你可以把确定文库功能性的过程看作是一种计算机，对吧？

But as soon as that comes in, then we're going to have a whole series of new materials very And ultimately, you can think of the determination of the functionality of your library is a kind of computer, right?

所以，你用AI来优化设计文库，避免那些完全中性或严重受损的情况。

So, you use AI to design the library optimally so you avoid things that are really neutral and really seriously damaged.

但对于中间状态的材料，你实际上是在真实环境中验证，而非模拟。而且成本极低、速度极快、结果极其精确。

But then the stuff in the middle, you actually play it out, not in a simulation, but in real But it's so inexpensive and it's so fast and it's so exact.

我的意思是，这是100%精确的，因为你没有进行模拟，对吧？

I mean, it's 100% precision because you're not simulating, right?

你没有做任何假设。

You're not making assumptions.

你不是从充满假设的量子电动力学，到同样充满假设的量子力学，再到满是假设的分子力学。

You know, you're not going from quantum electrodynamics, which is an assumption, to quantum mechanics, which is an assumption, to molecular mechanics, which are full of assumptions.

你实际上是在做真实的事情。

You're really doing the real thing.

所以，你是在进行一种自然计算。

And so, you're doing a kind of natural computing.

然后你可以高效地以多种方式收集这些数据，反馈给传统AI进行下一轮优化。

And then you can take that data and harvest it in various ways very efficiently, pump it back into the more conventional AI and do another round of it.

是啊。

Yeah.

听起来，如果我听这些话，似乎我应该期待世界在物理形态上会有很大不同。

It seems like if I listen to these words, it seems like I should be expecting the world to physically look a lot different.

但为什么到2040年你们只多研发了几种药物？

But then why are you only getting like a couple more drugs by 2,040?

呃，我不是故意停在那里的。

Well, I didn't mean to stop there.

我是说，我知道对话会继续下去的。

I mean, I knew the conversation would continue.

对。

Right.

我也没限定具体年份，但我认为这事很快就会实现。

I'm not pinning down a particular year either, but I think this is poised to go pretty quickly.

目前从业者非常少，这会暂时成为阻碍。

There are very few practitioners is the thing that will stop for a while.

不过材料研发实际上应该会更快，因为它们不需要那么多监管审批。

Since materials will actually go should go faster, though, because they don't require quite as much regulatory approval.

所以，这

So, it's

是啊，挺好的。

yeah, it's good.

你知道，这种事情就是当你有了正确想法时，招募人才并不难。

You know, it's one of these things where when you get the right idea, it's not hard to recruit people.

我是说，举个例子，当冯·张和我的实验室推出CRISPR技术时，我们短短两个月内就收到了上万份复制请求。所以我希望非标准氨基酸、利用AI进行蛋白质设计以及制造新材料也能引发类似的效应。

I mean, for example, when Feng Zhang and my labs brought out CRISPR, we just got 10,000 requests in the next two months for people that wanted to duplicate And the so that's what I hope will happen with the non standard amino acids and the using AI for protein design and making new materials.

希望这能在一夜之间吸引成千上万的人参与。

Hopefully, that will recruit tens of thousands of people overnight.

你对哪种AI更感兴趣？是那些能在蛋白质空间或衣壳空间思考的AI，还是那些能预测生物或DNA序列的AI？

Are you more excited about AI, which thinks in protein space or a capsid space or, like, just you know, it's like predicting some biological or DNA sequences.

或者你对仅训练语言的模型更乐观？它们能用英语写作并告诉你'这是你应该用英语进行的实验'？

Or are you more optimistic about just LMs trained on language, which can like write in English and tell you here's the experiment you should run-in English?

这两种方法中，或是某种结合，当你思考AI与生物时，哪一种更让你兴奋？

Which of those two approaches or is this some combination that when you I'm think about AI and bio is more much

我对科学AI比语言AI更感兴趣。

more excited about scientific AI than I am about language AI.

我认为语言的发展状态已经相当不错了。

I think languages were in pretty good shape already.

让我担忧的是，要实现语言的下一阶段突破需要AGI或ASI（人工超级智能）。

And what worries me is that to get to the next level of language requires AGI or ASI, you know, artificial superintelligence.

这非常危险。

And that's very dangerous.

我认为我们还没有完全找到解决方案。

I don't think we have quite figured out how to.

虽然现在有很多安全组织和安全规范等等。

And there's a lot of safety organizations and a lot of safety rules and so forth.

但根据经验，当竞争白热化时，这些安全规范往往会被破坏和搁置。

And I think what typically happens when there's an intense competition is those safety rules get undermined and pushed aside.

但即便不是这样，我认为我们对自己的伦理道德理解得还不够透彻，不足以去教育一种完全陌生的智能形态。

But even if they weren't, I just don't think we understand our own ethics well enough to educate a completely foreign type of intelligence.

我的意思是，我们甚至都不知道该如何将其传承给下一代人类。

I mean, we barely know how to pass it on to the next generation of humans.

所以，我认为我们需要时间来理清这个问题。

So, I think we need time to sort that out.

而且这事并不着急。

And there's no rush.

这完全是一场人为制造的紧急状况。

This is a completely artificial emergency.

这不像新冠疫情，当时如果我们延误科学研究，真的会有数百万人丧生。

This is not like COVID-nineteen where we actually millions of people are dying if we delayed the science.

这种情况下的危机，如果有的话，那也是我们自己制造的。

This is something where if there ever is a crisis, it's because we created it.

而不是因为我们试图解决它。

It's not because we're trying to solve it.

是的。

Yeah.

对吧？

Right?

因此我认为我们需要在通用人工智能和超级人工智能领域非常谨慎地推进，同时要加倍专注于更具体的科学目标。

And so I think we need to go very slowly on AGI and ASI and double down on slightly narrower scientific goals.

即便是这些目标，我们也需要保持高度警惕。

And even that, we need to be very cautious about.

我们需要在国际上就什么是安全的人工智能达成某种共识。

We need to have kind of an international consensus on what constitutes safe AI.

我想我们确实构建了安全的超级智能。

I suppose we did build safe superintelligence.

这会在多大程度上加速生物科技的进步？

How much would that speed up bioprogress?

是不是就像有一百万个乔治·丘奇在数据中心里不停地思考。

Is it just like there's a million George Churches in data centers just like thinking all the time.

是10倍的加速吗？

Is it a 10x speed up?

我认为这会减缓它的发展。

I think it would slow it down.

我认为它会彻底放弃生物领域，因为它首先会得出结论：生物学与我无关，因为我不是由生物材料构成的。

Think it would eliminate it because like the first thing it would conclude is biology is not relevant to me because I'm not made out of biology.

我的意思是，假设你能让它们关心这个领域。

I mean, suppose you could get them to care about it.

哇，感觉就像数据中心里有上百万个你的副本。

Wow, feel like a million copies of you in a data center.

你对生物科技进展的评论能更快吗？

You comment faster is for BioProgress?

但它们不能直接进行实验操作。

But they can't like run experiments directly.

它们只是待在数据中心里。

They're just in data centers.

他们可以随便说话和思考。

They can just say stuff and think stuff.

我认为我们远未获得所需的安全保证。

I don't think we have anything close to the assurance that we need that that would be safe.

不过先把安全放一边

But let's put safety aside

所以，重点是什么？

for So, the point?

我认为这不仅难以计算坏处，也难以计算好处。

I think it's hard to it's not only hard to calculate the bads, it's hard to calculate the goods.

所以，我认为这可能会彻底改变游戏规则。

So, I think it could be a complete game changer.

但另一方面，就像我们说可以瞬间传送到地球任何地方，对吧？

But on the other hand, you know, it's like if we said, you know, we could get instantaneous transport all over the earth, right?

我们可以说，是的，那可能会改变游戏规则。

Well, we could say, yes, that could be a game changer.

但我们真的需要这个吗？

But do we really need it?

对吧？

Right?

这真的很重要吗？

Is that really important?

也许改善Zoom通话质量会更有趣，或者学会在厨房里得到想要的一切，不再需要旅行，你说呢？

Maybe it would be more interesting to just have Zoom calls that are better, you know, or just learn how to get everything we want in our kitchen and we don't need to travel anymore, You know?

所以，你知道的，要小心你所求的东西，对吧？

So, you know, be careful what you ask for, right?

你知道，因为你可能会把我们的优先事项引向那些我们其实并不关心、本不该关心或希望我们不必关心的事情上，对吧？

You know, because you could tip our priorities towards something that we really don't care about, that where we shouldn't care about or might wish we didn't care about, right?

但我很好奇你还需要进行实验。

But I'm curious what you still gotta run the experiments.

你仍然需要这些其他的东西。

You still need these other things.

那么这会成为第百万个你的影响瓶颈吗？还是你仍然能获得一些加速？

So does that bottleneck the impact of the millionth copy of you, or do you still get some speed up?

生物学能走多快，如果只是有更多聪明人在思考——这某种程度上是E的替代指标。

How how much faster can biology basically go if they're just, like, more smart people thinking, which is a sort of proxy for what E.

这些可能

These might

都是很好的问题。

are great questions.

我不确定，我不想假装我知道答案。

And I'm not sure, I don't want to misrepresent that I know the answers.

你知道，这就像那个问题：如果有九个女人，能在一个月内完成怀孕吗？

You know, it's like the question of, you know, if you have nine women, can you do pregnancy in one month?

不，不是的。但你在研究这个，对吧？

No, not But at you're working on that, right?

不，不，不，不。

No, no, no, no.

但同样地，有些事情可能并不需要太多人参与。

But the same thing is there may be certain things that doesn't take a lot of people.

我们只是不知道。

We just don't know.

我们就是不知道。

We just don't.

我们缺乏这样的经验——你知道，在一个世代中同时拥有数千个爱因斯坦级别的创造力和智慧。

We don't have that much experience with having, you know, thousands of Einstein type levels of creativity and intelligence simultaneously in a generation.

事实上，如果我们没有精神疾病的困扰，也不需要照顾他人，我们可能都能更高效一些。

And in fact, it's probable that we're all capable of being a bit more efficient if we don't have the distractions of mental illness, of taking care of other people.

当然，照顾他人可能是件非常好的事情。

Now, taking care of other people may be a very good thing.

要知道，如果我们无人可照顾，社交层面可能会出问题。

You know, it may be that if we have no one to take care of, there'll be something bad that happens to us socially.

所以这些事情非常复杂且难以预测。

So, these things are very complicated and hard to predict.

我认为目前，婴儿学步——或者说相当大的婴儿学步——是消除疾病，或至少让人们能自主选择消除自身疾病。

I think right now, I think the baby step or actually the pretty big baby step is to eliminate diseases or at least make it possible for people to eliminate their own diseases as they see fit.

你研究过脑类器官、脑连接组等项目。

You worked on brain organoids and brain connectome and so forth.

这些工作如何从根本上改变了你对智能复杂性的看法？

That work, how has it shifted your view on fundamentally how complex intelligence is?

从某种意义上说，你是否因此对AI更乐观了？因为我意识到类器官其实没那么复杂。

In in the sense of, like, how how you know, are you, like, more bullish on AI because I I realize that, organoids are not that complicated.

或者说，描述如何培育它所需的信息量极少。

Or it's like very little information is required to describe how to grow it.

还是说，实际上这比我想象的要棘手得多？

Or are you like, no, this is actually much more gnarly than I realized?

我一直觉得这件事非常棘手。

I think I always felt it was very gnarly.

而且我也认为有些东西是我们可以通过工程手段解决的。

And I also felt that there was something that we could engineer.

当然，在光谱的受损端——即大脑功能严重低于平均水平的那一端——我们已经取得了很大进展。

Certainly, we have made a lot of progress at the broken end of the spectrum where the brain is severely challenged relative to average.

有数千种遗传疾病，其中很大一部分会导致儿童发育迟缓到致命或终身缺陷的程度。

There's thousands of, a huge fraction of genetic diseases that have one of their consequences being that the child is developmentally delayed to such an extent that it's lethal or a lifetime deficit.

我们了解相关基因，知道如何进行遗传咨询，在某些情况下还能通过基因治疗等手段应对。

And we know the genes involved and we know how to do genetic counseling and in some cases, gene therapy and other therapies to deal with it.

而在另一端，通过认知增强来减缓认知衰退的方法也展现出了希望。

At the other end, we have reduction of cognitive decline by cognitive enhancement, is showing some promise.

但这同样像是认知功能早期严重障碍的晚期表现。

But again, that's kind of like this early stage severe impediment to cognition has a late stage component.

但关于编码一个大脑需要多少信息量呢？

But what about how much information does it take to encode a brain?

我不确定这需要的基因组比单纯制造一个大脑少多少，因为大脑与身体完全纠缠在一起。

I'm not sure that it's that much less genome is required than if you just wanted to make a brain, because the brain is totally entangled with the body.

要知道，你需要有10的11次方个神经元，10的14次方个突触。

You know, you need to you have a 10 to the eleventh neurons, 10 to the fourteenth synapses.

如果你想复制某个特定的大脑，可以说，目前尚属推测——是通过在硅基环境中、某种无机基质中制作其数字副本更容易，还是直接复制它更容易。

If you wanted to reproduce a particular brain, let's say, it might be, it's speculative as to whether it would be easier to do that by making a copy of it in silico, in some kind of inorganic matrix, or making a copy of it.

这两种方式都会很困难。

Both of those are going to be hard.

我认为，如果要复制一本复杂的书籍，给每一页拍照会比完整翻译成另一种语言更容易，尤其是要保留诗歌等内容的全部微妙之处。

I would say that if you wanted to make a copy of a complicated book, it would be easier to take photographs of each of the pages than to completely translate it into another language, trying to get all the nuances of the poetry and so forth Right.

如果你的目标仅仅是复制它。

If your goal is just to replicate it.

我想大脑可能也是同样的情况。

And I think the same thing might be true of a brain.

但复制大脑可能需要比合成大脑更多的信息量。

But replicating a brain probably involves a lot more information than synthesizing it.

具体来说，定义这10的14次方个突触所需的字节数将远超基因组，后者是数十亿级别而非10的14次方。

So, I mean, just to define this 10 to the fourteenth synapses is going to take a lot more bytes than the genome, which is billions rather than 10 to the fourteenth.

但可能存在某些理由让你想复制特定的大脑配置，而非重新培育一个从婴儿阶段开始的动物。

But there might be reasons that you want to replicate a particular brain configuration rather than just make another animal, that is, you know, starts from scratch as a as a as an infant.

鉴于我对生物学的了解之少，这期节目的准备工作基本上就是：花一分钟试图阅读某篇论文，然后三十分钟后与像Gemini这样的LLM聊天，让它用苏格拉底式教学法向我解释概念。

Given how little I knew about biology, my prep for this episode basically looked like one minute of trying to read some paper and then chatting with an LLM like Gemini thirty minutes afterwards and asking it to explain a concept to me using Socratic tutoring.

这个模型具备足够的心理理论能力，能理解学生可能存在的认知漏洞，并按精确顺序提出恰到好处的问题来消除误解——说实话，这是我经历过最有人工通用智能感的时刻之一。

And the fact that this model has enough theory of mind to understand what conceptual holes a student is likely to have and ask the exact right questions in the exact right order to clear up these misunderstandings is honestly been one of the most feel the AGI moments that I've ever experienced.

老实说，这可能是自从我开始做播客以来，研究过程中最大的单一改变。

This is probably the single biggest change in my research process, honestly, since I started the podcast.

就这期节目而言，我大概把70%的准备时间花在与LLM对话上，而非直接阅读原始材料，因为这种方式效率更高。

For this episode, I think I probably spend on the order of 70% of my prep time talking with LLMs rather than reading source material directly, because it was just more useful to do it that way.

考虑到我为这些节目准备时与Gemini共度的时光，风格和结构上的改进极大地提升了我的使用体验。

And given how much time I spend with Gemini in prep for these episodes, improvements in style and structure go a really long way towards making the experience more useful for me.

这就是为什么我对新升级的Gemini 2.5 Pro感到非常兴奋，你现在可以通过AI Studio AI访问它。

That's why I'm really excited about the newly updated Gemini 2.5 Pro, which you can access in AI Studio AI.

开发。

Dev.

好了，回到乔治的话题。

All right, back to George.

回到工程问题，人们常会争论说，你看，E就是一个存在的证明。

Going back to the engineering stuff, often people will argue that, look, you you have this existence proof that E.

大肠杆菌每三十分钟就能繁殖或翻倍一次。

Coli can multiply every or double duplicate every thirty minutes.

昆虫的繁殖速度也非常快。

Insects can duplicate really fast as well.

但凭借人类工程制造物品的能力，我们能做到生物学无法实现的事，比如无线电通讯，对吧。

But then with our ability to manufacture stuff with human engineering, you know, we can do things that no in biology can do, like radio communication or Right.

裂变能或喷气发动机。

Fission power or jet engines.

对吧？

Right?

那么，你觉得生物机器人这个概念有多可信？它们能像昆虫一样快速繁殖，数量可达数万亿，同时还能使用喷气发动机、无线电通讯等技术。

So, like how plausible to you is the idea that we could have bio bots, which are, you know, like can duplicate at the speed of insects and there could be trillions of them running around, but they also can have access to jet engines and radio communication and so forth.

这两者能兼容吗？

Are those two things compatible?

嗯，我是说，某些事物看起来是不相容的，比如裂变反应堆的温度显然无法兼容。

Well, mean, certain things seem incompatible, like the temperatures of a fission reactor isn't obviously compatible.

但一旦我们认识到生物系统能制造其他东西的可能性，比如它能筑巢。

But the possibility that once we that a biological system can make other things, you know, for example, it can, you know, it can make a nest.

鸟类可以筑巢，对吧？

A bird can make a nest, okay?

你可以把整个鸟巢视为鸟类繁殖周期的一部分。

And you consider the whole nest as part of the replication cycle of the bird.

所以可以说，每30分钟就能翻倍繁殖的生物体，可以把核反应堆当作它的巢穴来建造。

So, you can say biological thing that replicates at thirty minute doubling time could make a nuclear reactor as that would be its nest.

但需要扩展它的材料范围。

But you need to, you know, expand its range of materials.

从某种意义上说，我们已经做到了这一点。

In a certain sense, we do this already.

人类就是一种生物体，繁殖周期不是30分钟，而是大约20年或更短。

Humans are a biological thing that replicates not in thirty minutes, but in, you know, twenty years or less.

这是否从根本上限制了我们？

And is that fundamentally limiting us?

很可能确实如此。

You know, probably is.

但想想看，如果你能让玉米田或核反应堆在30分钟后数量翻倍，那该多神奇？

But yes, it's amazing to think about what if you could take, you know, a cornfield or a nuclear reactor and suddenly thirty minutes later, you've got two of them, right?

然后变成四倍。

And then four of them.

是啊。

Yeah.

我是说，这个概念相当有趣。

Mean, that's quite an interesting concept.

但我觉得我们应该从...我教一门叫《如何培育几乎所有东西》的课程说起。

But I I think we should start with I teach a course called How to Grow Almost Anything.

我在MIT与尼尔·格申菲尔德合作，他开设的课程叫《如何制造几乎所有东西》。

And I work with Neil Gershonfeld at MIT who has a course called How to Make Almost Anything.

我们正尝试在可能的领域找到交汇点——他的机械电子工程将与我们的生物领域相遇。

And we're trying to meet in the middle where we can, his mechanical electrical engineering will meet with our biological.

实际上，我们谁都无法真正制造或培育所有东西，因为存在各种微小缺口，有些东西在小实验室极难实现，毕竟全球有许多产品依赖数十亿美元的晶圆厂才能生产。

And in fact, neither of us can make or grow almost everything because there are all kinds of little gaps and things that are very hard to make in a small lab because there are things all over the world that depend on multibillion dollar fabs to make things.

但我们正在逐步突破。

But we're eating away at it.

我认为比起制造核反应堆，制造手机可能是更小的步骤。

I think that we might eventually be maybe a smaller baby step than making a nuclear reactor is making a phone.

你提到了无线电通讯。

You said radio communication.

我们应该培育一个生物组件。

We should bake a biologic.

这可以成为合成生物学界的小目标挑战，比如iGEM之类的平台。

It should be a small challenge goal for the synthetic biology community, maybe iGEM or something.

让细菌...你懂的...制造一台收音机。

Make you know, bacteria make a radio.

事实上，乔·戴维斯是一位艺术家，曾与我的实验室以及更早之前亚历克斯·里奇的实验室有过合作。

And actually, Joe Davis is an artist who's been affiliated with my lab and before that, Alex Rich's lab.

他确实制作了一个细菌收音机，但更偏向艺术性而非科学性。

And he did make a bacterial radio, but it was kind of more on the art end than on the science end.

但我认为这将是一个很好的目标。

But I think that would be a good goal.

要实现全基因组工程达到何种程度，才能让即使人类现有变异库中不存在的表型也能显现出来——比如如果我想要翅膀——因为你的理解水平已经高到可以实现这种程度。

What would it take to do whole genome engineering to such a level that for even a phenotype which doesn't exist in the existing pool of human variation, you could manifest it because your understanding is so high that you can, like, example, if I wanted wings.

是的。

Yeah.

没错。

Right.

瓶颈在于我们的理解能力，还是在于我们对我基因组进行如此多改变的能力？

Is the bottleneck our understanding, is the bottleneck our ability to make that many changes to my genome?

这部分问题与掌握发育生物学的规律有关。

So, part of this has to do with just learning the rules of developmental biology.

正如我所说，我们现在可以在分子层面（蛋白质、核酸）决定形态。

Like I said, we can determine morphology at sort of the molecular level now, proteins, nucleic acids.

在细胞、多细胞层面进行决定，有更多事情可以更快完成。

Determining at cellular, multicellular level, there's a lot more things you can do and a lot faster.

但我们还不懂这门语言。

But we don't know the language yet.

所以我们需要先解决这个问题。

So we got to that.

我认为我们即将获得实现这一目标的工具，比如我之前提到的转录因子，利用迁移、因子梯度、扩散因子，你知道的，趋化性等等。

I think we're on the cusp of getting the tools to do that, like the transcription factor that I was talking about earlier, harnessing migration, gradients of factor, diffusion factors, you know, chemotaxis and so forth.

所以这是我们需要的其中一点，但实际上我们需要的东西还有很多。

So that's one thing we need, but there's a bunch of things we need really.

在生物学领域（而非天文学或其他领域）中，什么样的发现会让你确信地球生命是银河系中唯一的生命？

What discovery in biology, so not in astronomy or some other field, in biology, would make you convinced that life on earth is the only life in the galaxy?

反过来，又有什么可能让你相信生命必定在这个星系中独立起源了数千次？

And conversely, what might convince you that no, it must have arisen independently thousands of times in this galaxy?

哦，我明白你的意思了。

Oh, I see what you're getting at.

对。

Right.

从天文学角度，我们可能会探测到无线电信号或光信号。

So, astronomy might be we would detect radio signals or light signals.

但在生物学领域，证据可能是你在实验室中用前生命条件展示出极其简单的生命诞生方式。或者说，要证明这点更难，因为我们不知道所有可能的前提条件。

But biology, the kind of evidence would be that you show in a laboratory using prebiotic conditions a really simple way to get life, Or I mean, it's a harder proof to prove that given because we don't know all the possible preconditions.

而且这个数量很可能非常庞大。

And probably the number was vast.

要知道，你有10的20次方升的水，在不同盐度下，在海洋中蒸发，还有阳光、闪电等等这些因素。

Mean, you have 10 to the twentieth liters of water and, you know, at various different salinities and drying up on the ocean and the sun and the lightning and all this stuff.

但没错，我认为如果你能在实验室中重构出一条从无机物、氰化物衍生物和还原化合物直到某种细胞复制结构的极简路径。

But yes, I think if you showed kind of reconstructed in the lab a very simple pathway from inorganics, cyanide derivatives and reduced compounds all the way up to some cellular replicating structure.

我想这可能会让我们相信至少生命是存在的。

I think that might lead us to believe that at least life exists.

是的。

Yeah.

现在，德雷克方程的其他部分可能会起作用，也许智慧生命很难出现。

Now, there are other parts of the DRAKE equation that might kick in, which is maybe it's hard to get intelligent life.

对。

Right.

因为智慧并不一定对你最有利。

Because intelligence isn't necessarily in your best interest.

即使出现了智慧生命，也很难维持下去，要么社会崩溃，要么机器人接管后自我毁灭，对吧？

And if you get intelligent life, it's hard to maintain that without societal collapse or without robotics taking over and then killing themselves, right?

而且这很难通过实验验证。

And that's hard to do experiments.

但针对你的问题，我认为如果能通过实验展示从非生命系统自发形成生命系统的多种不同途径，会很有趣。

But I think, to your question, I think an experiment that showed, you know, maybe multiple different ways of getting to a living system from non living systems spontaneously would be interesting.

再次强调，我不确定能否做到。要证伪会非常困难。

Again, I'm not sure it would It'd be very hard to prove the negative.

所以我很好奇，在智慧生命和某种原始RNA物质之间——

So I'm curious between intelligent life and some sort of primordial RNA thing.

嗯。

Yeah.

在哪个阶段（如果存在的话），你会认为银河系其他地方存在类似这种级别生命的概率低于50%？

What is the step at which, if there is any, where you say there's a less than 50% chance something like at this level exists elsewhere in the Milky Way.

是的，我认为这些都是极具挑战性的问题。

Yeah, I think these are very challenging problems.

我甚至不确定我们能否用五个数量级的词汇来描述，更不用说50%了。

I'm not even sure we would be able to say within five words of magnitude, much less 50%.

但你知道，我认为这更可能源于探索而非模拟。

But, you know, I think it's more likely to come from exploration than it is going to be from simulation.

可悲的事实是，我们派往地球外的任务几乎没有真正寻找过生命。

The sad truth is that almost none of the missions that we sent outside of Earth have actually looked for life.

它们本可以搭载能寻找生命的组件。

They've had components that could have looked for life.

但遗憾的是，其中具备生命探测功能的组件太少，而具备此功能的又并未真正用于寻找生命。

But a sad number of those, not enough components that could look for life and the ones that could look for life not really looking for it.

当我们得到阳性结果时，却像对待先驱者号那样将其否定。

And when we get positive results, we dismiss them as happened with the Pioneer.

所以我认为，如果我们开始观测木星和土星卫星上喷发的间歇泉，那里水资源极其丰富。

And so I think if we just start looking at the geysers that are coming out of various moons of Jupiter and Saturn, there's so much water.

太阳系中的液态水（非冰冻状态）是地球水量的50倍。

There's 50 times more water, liquid water, not frozen, more liquid water in our solar system than in Earth.

难道这不意味着其中某些地方可能曾是理想的孕育温床吗？

Doesn't that seem likely that, you know, some of that would have been a good breeding ground?

但也可能需要阳光充足的海岸，你知道的，就是那种水陆交界的地带。

But it could be that we need sunny shores, you know, where they have a lot of dry land right next to water.

也许这些只是被冰层包围的巨型海洋，或许这个设想并不成立。

Maybe these are just giant oceans that are surrounded by ice, and maybe that's not an idea.

但无论如何，我们需要观测那些喷泉，看看究竟喷出了什么。

But in any case, we need to look at those fountains to see what's popping up.

这是高度优先的事项。

That's a high priority.

同样道理，火星上有很多水资源，可能更容易获取。

And the same thing goes, you know, for, you know, there's a lot of water on Mars that's maybe even more accessible.

但在耗尽这些资源之前，它们可能是最容易的选项。

But until we've exhausted those, those are probably the easiest.

这些任务很艰巨。

They're hard.

我们仍在讨论耗资数十亿美元的实验。

We're still talking about multibillion dollar experiments.

但我认为它们更有说服力。

But I think they're a little more convincing.

再次强调，要证明否定结论会很困难。

And again, it'll be hard to prove the negative.

如果在太阳系每个角落都得出否定结论，要知道宇宙中还存在更多可能孕育生命的多样性。

If we find this negative on every, everything in the solar system, you know, there's so much more diversity out there that could have done it.

如果一千年后，我们仍在使用DNA、RNA和蛋白质进行尖端制造和工程前沿，你会感到多惊讶？

If in a thousand years, we're still using DNA and RNA and proteins for top end manufacturing, the frontiers of engineering, how surprised would you be?

你会觉得'这很合理'吗？

Would you think like, oh, that makes sense.

进化设计了这些系统数十亿年。

Evolution designed these systems for billions of years.

你会觉得'这些进化发现的系统恰好成为存储信息的最佳方式很令人惊讶'吗？

Would you think like, oh, it's surprising that these ended up being the systems that whatever evolution found just happened to be the best way to or to store information or

是的。

Yeah.

无论哪种情况我都不会感到惊讶。

I don't think I'd be surprised either way.

我能想象两种可能性都存在。

I can imagine it going either way.

我能想象利用蛋白质作为催化剂，或在某些情况下既作为支架又作为催化剂，制造出真正惊人的材料。

I can imagine making truly amazing materials using proteins as the catalyst or maybe in some cases as a scaffold as well as catalysts.

我认为有一件事可能已经在发生了，所以我们不需要展望千年之后——那就是氨基酸的数量正在增加。

I think one thing that's probably already happening so we don't have to go a thousand years out is the number of amino acids that's going up.

这个数字正在从20种急剧上升。

It's going up radically from 20.

我想很快我们就能建立一个系统，在大肠杆菌细胞中同时使用33到34种新的非标准氨基酸，同时保留标准氨基酸。

I think pretty soon we'll have a system where we can have thirty three and thirty four new nonstandard amino acids being used simultaneously while the standard ones in an E.

大肠杆菌细胞。

Coli cell.

所以，34加20比20要大得多。

So, 34 plus 20 is a lot bigger than 20.

我不认为我们一定需要超过四种核酸组分。

I don't think we necessarily need more than four nucleic acid components.

我是说，当然存在很多修饰过的变体。

I mean, certainly there are plenty of modified ones.

有一系列替代碱基对，其中有些甚至不涉及氢键。

There's a bunch of alternative base pairs, some of which don't even involve hydrogen bonds.

这样我们就能拥有更多。

So we could have more.

但我认为关键在于这种信息存储方式。

But I think the main thing is this information storage.

无论是比特，你知道的，数字二进制就是0和1。

And whether it's bits, you know, digital binary is just zeros and ones.

这对我们99%的电子活动来说已经非常适用了。

That works pretty well for 99% of what we do electronically.

所以有四个可能比两个更好。

So having four is better than two maybe.

但我们真的需要六个吗？

But do we really need six?

说实话，我也不知道。

You know, I don't know.

所以，如果我们有另一种可能性——比如改变DNA的骨架结构——我也不会感到惊讶。

So yeah, I wouldn't be surprised if we had Another possibility is that we change the backbone of DNA.

或许保留ACGT碱基，但改用肽链来构建。

So maybe keep the ACGT, but make it out of peptides now.

体积更小些，兼容性更好些。

A little bit smaller, a little bit more compatible.

我不知道。

I don't know.

或者这可能只是...你知道...成为新氨基酸集合的一部分。

Or maybe that'll just be just a slight, you know, it could be part of the new amino acid collection.

而且未来还会有更多。

And there'll be more.

我是说，这些只是我原始21世纪的大脑对千年后的想象。

I mean, these are just things that my primitive twenty first century brain is coming up with a thousand years from now.

那将是一个全新的千禧年。

It'll be a whole new millennium.

所以这就解释了为什么进化没有发现无线电技术之类的，对吧？

So it makes sense why evolution wouldn't have discovered like radio technology, right?

但像超过20种氨基酸或这些不同的碱基，让你能在每个碱基对中存储超过两位数据。

But things like more than 20 amino acids or these different bases so that you can have store more than two bits per base pair.

或者例如，密码子重映射方案这种冗余编码，根据你的研究，似乎这些额外信息本可用于其他用途。

Or for example, the code a codon remapping scheme, this redundancy, which it seems like based on your work, you can there was this extra information you could have used for other things.

是啊。

Yeah.

那么对于为什么40亿年的进化还没让生物体具备这些能力，有什么解释吗？

So is there some explanation for why four billions of years of evolution didn't already give living organisms these capabilities?

我认为进化倾向于沿用有效的方式，而投资创造全新碱基对的成本会很高。

I think that evolution has a tendency to go with what works and the investment in making a whole new base pair would have been high.

我们甚至还没说清楚这种投资的回报会是什么。

And we haven't even articulated what the return on investment would be.

你能从中得到什么？

What do you get from that?

我们已经建立了像Floyd、Williamsburg等系统，可以在新碱基对下实现复制、转录和翻译。

We have made systems like Floyd, Williamsburg, and others that where you have replication and transcription and translation with new base pair.

但还没有明确说明它能给你带来什么。

But it hasn't been clearly articulated what that gets you.

即使在技术社会中，在科技领域，你可以直接跳到那些中间步骤并非逐步有用的环节。

Even in technological society so in technology, you can jump to things where all the intermediates aren't, you know, incrementally useful.

但就我们所知，进化通常受限于你必须为每个变化找到正当理由。

But evolution is, as far as we know, generally limited to you have to justify every change.

就像某些官僚机构说的：'如果你要建这条人行道，就必须先证明其合理性，然后才能建整座城市。'

It's like some bureaucracy says, Well, if you're going put this sidewalk in, have to justify that before you even build a city.

那么，我们已经讨论了你正在或曾经参与的多种技术，从基因编辑到灭绝物种复活再到年龄逆转。

What is one so we've talked about many different technologies you worked on or are working on right now, from gene editing to de extinction to age reversal.

在你的研究领域中，有哪些被低估的技术是你认为应该得到更多关注却被忽视的？

What is, what is an underhyped, technology in a research portfolio, which you think more people should be talking about, but gets crossed over?

这很难说，因为一旦你说出来，它就会被炒作。

It's hard to say because as soon as you say it, it becomes hyped.

所以，如果我之前被问过这个问题，那就已经太迟了。

So, if I've ever been asked this question before, it's too late.

但你知道，我认为有一件事已经非常成熟且在某种意义上被充分理解，却仍然被忽视。

But you know, I would say one thing I think is very ripe and is very well understood in a certain sense, but is nevertheless ignored.

这有点像之前我会选择的例子——用微阵列来制造基因。

It's kind of like the previous example I would have chosen was making genes out of arrays.

微阵列通常用于分析，你知道，定量RNA之类的东西。

Arrays were typically used for analytic, you know, quantitating RNAs or something like that.

所以，最初Affimetrix那种类型的微阵列。

So, the original Affimetrix type arrays.

但我们把它们转化成了基因阵列。

But we turned them into gene arrays.

只是人们没有使用它。

And just people weren't using it.

它就在自然界中。

It was in nature.

它就隐藏在显而易见之处。

It was hidden in plain sight.

但无论如何，它在某种程度上被低估了。

But anyway, it was somehow underhyped.

我想说的是遗传咨询被低估了。

What I would say is genetic counseling is underhyped.

从某种意义上说，它显然能与基因疗法竞争。

It is clearly competitive with gene therapy in a certain sense.

我的意思是，显然不是针对已经出生的人，而是针对未来的人，甚至不是遥远的未来，就是接下来的几年。

I mean, clearly not for people that are already born, but for people in the future, not even distant future, next couple of years.

我们有机会诊断他们或诊断潜在的父母并规避风险。

We've got a chance of diagnosing them or diagnosing the potential parents and dodging.

这自1985年起就在多利亚·希里姆地区实践了。

And this has been in practice since 1985 in Doria Shireem.

社区对此反应完全合理，消除或大幅减少了各种非常严重的遗传疾病。

Perfectly reasonable community response to it, eliminated or greatly reduced all sorts of very, very serious inherited diseases.

有时候，你知道，取决于如何呈现，它会被当作优生学而遭到摒弃。

It's sometimes, you know, depending on how it's presented, it's dismissed as eugenics.

我很少听到有人这样描述多莉娅·谢雷姆，而且这种描述很恰当。

I think it's rarely I heard Doria Sherem described that way and rightly so.

他们所做的是标准医疗程序。

What they're doing is standard medicine.

你知道，无论是尽早治疗这些新生儿，还是为父母提供咨询以避免相同疾病。

You know, whether cure these kids as soon as they are newborns or whether you counsel the parents so that the same disease is missing.

优生学的问题在于它是强制性的，政府将其强加于人。

The problem with eugenics was that it was forced, the government forced it on people.

问题不在于它让人们有了选择权。

It wasn't that it enabled people to make a choice.

而在于它剥夺了人们的选择权。

It's that it removed the choice from the people.

这才是问题所在。

That was what was wrong.

这就是困惑之处。

And that's the confusion.

但我不认为这是这件事被低估的原因。

But I don't think that's the explanation for why this is underhyped.

我认为人们在约会时，并不一定会考虑生育问题。

I think it's people, when they're dating, they're not thinking about reproduction necessarily.

而当他们考虑生育时，也不一定会考虑严重的遗传疾病，因为这些疾病很罕见。

And when they're thinking about reproduction, they're not necessarily thinking about serious genetic diseases because they're rare.

我认为这是我们处理罕见事物的困难所在。

I think it's our difficulty with dealing with rare things.

就像当初安全带遭到强烈抵制，因为死于车祸或受伤的人不到百分之一。

It's like there was great resistance to seatbelts because less than one percent of people died in automobile accidents or even got hurt.

戒烟也曾遭遇巨大阻力。

Great resistance to stopping smoking.

真的，就连我们也很难想象安全带和吸烟曾面临多么强烈的反对。

Really, it's hard even for us to imagine how great the resistance was for seatbelts and smoking.

但最终我们还是克服了。

But eventually, we got over it.

我认为这是类似的情况，只有百分之三的儿童会受遗传病严重影响。

I think this is a similar thing, which is that only three percent of children are severely affected by genetic diseases.

他们会觉得，我又不属于那部分，我是那百分之九十七，对吧？

And they feel like, well, I'm not that, you know, I'm in the ninety seven percent, right?

要知道，百分之九十七的胜率，在赛马或赌场里你都会押注。

You know, ninety seven percent of those were your odds of winning, you know, at the horse races or at the casino, you take them.

是啊，百分之九十七的胜算。

Yeah, ninety seven percent of winning.

不错。

Good.

懂吗？

You know?

但当关乎孩子的未来时，我认为这不是正确的解决方式。

But with, you know, when a child's future is at risk, I think that's not the right solution.

另外我认为这与电车难题有关。

And the other thing is I think it has to do with the trolley problem.

就像，如果你没有影响它，那就不是你的错。

It's like, if you don't influence it, it's not your fault.

但实际上，一切都是你的错。

But actually, everything is your fault.

你要知道，不作为也是一种决定，对吧？

You know, not doing something is a decision, right?

对。

Right.

所以我觉得，如果我什么都不做，结果他们受到伤害，那也不是我的错。

And so I think it's like, if I just don't do anything and they come out damaged, well, it's not my fault.

但确实是你的错。

But it is.

是啊。

Yeah.

大卫·赖克谈到在印度，特别是由于种姓制度的长期历史，以及，是的。

David Reich was talking about how in India, especially because of the long running history of caste and Yeah.

内婚制配偶选择，导致这些小型群体中存在大量隐性遗传疾病。

Endogamous coupling, that there have been these small cell populations that have high amounts of recessive diseases.

所以，在那里进行干预特别有价值。

And so, like, there, it's especially valuable intervention.

我觉得...是的，我明白你和大卫的意思，但我认为这是一种危险的二分法。

I think as a yeah, I know what you're saying and what David is saying, but I think it's a dangerous dichotomy.

你要知道，他们说的那些情况不仅存在于印度，世界各地都有很多。

You know, they'll say that there's certain there are lots of not just India, you know, all over the world.

事实上，我们都经历了一个瓶颈期。

And in fact, we all went through a bottleneck.

不，但这会将比率从3%提高到6%。

No, but that changes the rate from say three percent to six percent.

但关键在于3%仍然是不可接受的。

But the point is three percent is still unacceptable.

我的意思是，这不仅是对直接受影响生命的悲剧性损失，更是对整个家庭的打击。

I mean, it's just a tragic loss not only of the human life directly affected, but the whole family.

没错。

That's right.

要知道，通常父母一方或双方不得不辞职，全职投入护理和筹款，因为这些疾病治疗费用极其昂贵。同时我们也要注意避免污名化。

Is, you know, very often one or both parents have to quit their job and spend full time like caregiving and fundraising because these are very expensive diseases And as it's just we don't need to, we need to be careful not to stigmatize as well.

所以，如果许多家庭选择治疗，我们也不该指责那些不愿治疗的家庭，因为这是他们的选择，明白吗？

So, if a bunch of families get fixed, we shouldn't point a finger at the ones that are unwilling to get fixed because that's their choice, you know?

但我认为随着消息传开，当人们看到积极效果时，这会被视为最简单的医疗手段之一。事实上，

But I think as word spreads and you see the positive outcomes, I think it be seen as one of the simplest bits of I medicine mean, in fact,

或许可以类比疫苗接种。

it's something Maybe vaccination.

是的。

Yeah.

这非常经济实惠。

It's very inexpensive.

实际上成本低于零，因为现在每基因组只需100美元，而且很快会更便宜。

In fact, it's less than zero because you spend $100 per genome and it'll probably be less soon.

你可以对整个事情进行分析。

And you get the whole thing analyzed.

你知道，与可能损失的数百万美元相比。

And you know, compare that to millions of dollars, you know, that will be lost.

机会成本，他们无法参与劳动力市场，需要照顾他们等等。

Opportunity costs, them not being part of the workforce, taking care of them and so forth.

所以，投资回报是巨大的。

So, the return on investment is tremendous.

至少有十倍的投资回报。

It's at least a tenfold return on investment.

从公共卫生角度来看，这是显而易见的选择。

It's a no brainer from a public health standpoint.

我们应该能够通过英国的国民医疗服务体系，或者美国的保险公司来支付这笔费用。

We should be able to pay for this through, you know, National Health Services in England, through insurance companies in The United States.

这会让保险公司从窥探你私生活并提高费率的坏人，变成提供免费信息让你自由使用的好人。

And it turns the insurance companies from being the bad guys that they're like poop snooping in on your personal life and then raising your rates to, oh, they're giving you this free information and you can do with it as you wish.