本集简介
双语字幕
仅展示文本字幕，不包含中文音频；想边听边看，请使用 Bayt 播客 App。
我们只是想为整个科学界提供这些工具。
We just wanna give tools to the whole scientific community.
我们希望弄清楚生物学是如何运作的。
We wanna understand how biology works.
我想了解这个人的基因情况。
I wanna understand the genetics of this person.
我想弄明白他们患各种疾病的风险有多大。
I wanna understand the risks they have to different illnesses.
我的目标是把每个人当作独立的个体来对待，理解背后的机制，然后能够进行干预。
My goal is to be able to treat the individual as an individual, understand the mechanisms, and be able to intervene.
如果我们以开源项目的方式来做，就能更快地让更多的科学家用上，影响力也会更大。
We'll have a bigger impact by getting us in more scientists hands quicker by doing it as open source projects instead.
这可不是说有个工厂，你付钱就能生产出数据那么简单。
It's not just like there's some factory somewhere that you can pay to produce the data.
你真的需要创造出全新的科学方法。
You actually need to invent new novel scientific approaches.
我们的目标并不是要彻底治愈这些疾病。
The theory isn't that we're gonna cure the diseases.
确实不是。
We're not.
我们的目标是帮助加快整个科学领域的研究进展。
It's that we wanna help accelerate the pace of progress for the
整个科学界。
whole scientific field.
我们折叠了超过十亿个蛋白质，并预测了它们的结构，但并没有专门针对抗体设计模型。
We folded over 1,100,000,000 proteins and predicted their structures, and we didn't design a model for antibodies.
我们也没设计一个专门去结合某个特定目标的模型。
We didn't design a model to be able to bind one particular target.
我们只是设计了一个能理解蛋白质的通用模型。
We just designed a model that could understand proteins.
如果我们能设计出一种能改变生理状态的蛋白质，那就能真正实现治愈。
If we could design a protein to actually change the physiology, then we can actually cure someone.
今天在《NoPriers》节目中，我们邀请到了马克·扎克伯格、普莉西拉·陈和亚历克斯·里夫斯。
Today on NoPriers, we're joined by Mark Zuckerberg, Priscilla Chan, and Alex Rives.
我们会聊聊Biohub，以及他们如何大规模应用人工智能，构建细胞和生物学各层级相互作用的全局模型。
We'll be talking about Biohub and all their various efforts to now start applying AI at scale to do world models of cells and different levels of interactions across biology.
马克、普莉西拉，谢谢你们参与这次对话。
Mark, Priscilla, thank you for doing this.
谢谢你们邀请我们。
Thanks for having
不客气。
It us
挺开心的。
was fun.
祝贺你们开启新的使命。
Congratulations on new missions.
谢谢。
Thank you.
你们把Biohub作为主要的慈善项目，还投入了五亿美元支持这个虚拟生物学计划。
You guys made Biohub your primary philanthropic effort and then committed $500,000,000 to this virtual biology initiative.
能跟我们说说，你们是怎么决定要这么做？从‘我们应该资助这个’到‘这简直太适合我们了’，中间经历了什么？
Can you tell us a little bit about, you know, why do that and how did you go from we should fund this to this is like
我们是谁？
who we are?
目前的Biohub，我们非常看好。
So Biohub in its current form, we're super excited about.
它和我们的理念、资源以及我们能共同实现的目标完全契合。
We feel like it's a really good fit for who we are and what we bring to the table and what we can achieve together.
但这项工作其实十年前就开始了，那时候我们在想，怎么才能回馈社会？
But this work started ten years ago when we were thinking about how can we give back?
我们希望创建一个组织，能在本世纪末彻底治愈、预防和控制所有疾病。
And wanted to build an organization that could cure, prevent, and manage all disease by the end of the century.
我们当时和一群科学家开了好几场会，那些可是拿过诺贝尔奖的顶尖学者，他们听完都笑得不行。
And we had a series of hilarious meetings with scientists that, like, famous Nobel Prize winning scientists were just laughing at us.
那是你们的起点吗？
Was that your starting line?
你们就打算根除所有疾病？
We're just gonna cure all disease?
不是。
No.
不是。
No.
要说明的是，我们并不认为自己会亲自治愈这些疾病。
And to be clear, we don't think that we're gonna be the ones curing the diseases.
我们的目标一直是打造能加速整个科学领域的工具，让科学界 collectively 能够治愈所有疾病。
Our goal is always to build tools that could accelerate the whole scientific field, so that way the scientific field collectively could cure all the diseases.
但当时人们还是觉得，到本世纪末实现这个目标太夸张了。
But still people thought that by the end of the century it was a stretch.
现在我觉得，这个目标反而太保守了。
Now I think it's like too conservative.
所以我们一直都在想，好吧，我们之前有过一系列挺搞笑、挺尴尬的教育性对话，当时我们就问，那到底为什么呢？
And so we kept being like, okay, well, we had these series of funny, awkward educational conversations where we were like, okay, but like, why?
为什么你觉得这不可能？
Like, why do you think it's impossible?
当你只是在场听着的时候，就会觉得，哎，我也不清楚为什么。
And like, you know, just being the person in the room is just like, oh, I don't know why.
你倒是说说看。
You tell me.
最后，人们终于松口了，说：行吧。
Finally, we got people to like, they're like, fine.
如果你真想知道的话。
If you really must know.
我们就说，当然想知道啊。
And we're like, you know, we do.
这事儿看起来挺重要的。
It seems important.
你知道吗，他们说，我们各自为政。
You know, they were like, well, we work in silos.
一旦发表成果，信息就没人共享了。
And when you publish, information doesn't get shared.
数据一锁就是好长时间。
It gets locked up for long periods of time.
我们连像样的工具都没有。
And we don't have tooling.
他们举了个例子，说有个博士后在实验室开发了一个特别棒的工具，结果就存在他自己的电脑上。
You know, they gave the example of like, we build a great tool by one postdoc in a lab and it lives on their computer.
等他毕业了，这工具也就没了。
And when they graduate, the tool is gone.
我们听到的是，想做个大家都能用的工具来加速科研，建个共享的知识库来更快推动科学进展，实在太难了。
And we heard was very hard to build shared tools to move science faster, build a shared knowledge base to quickly move science faster.
于是我们就开始想，既然这些问题都摆在这儿了，我们能做点什么？
And that's sort of where we began in thinking about, okay, like if those are the problems, like what can we contribute?
是的
Yeah.
说白了，最初的生物枢纽模式就是把来自多所大学的工程师和科学家聚在一起，专注长期工具的研发。
I mean, so the original Biohub model was basically focus on long term tool development by bringing together engineers and scientists across multiple universities to focus on long term tool development.
结果还真行得通。
And basically, it like worked.
一开始，我们和CZI一起做了不少事情。
And, you know, we started off with with CZI doing a number of different things.
后来我们慢慢觉得，嗯，这样下去挺好。
And I think over time, we just felt like, okay.
科学部分进展得很顺利，我们就不断加大投入，到现在，这已经成了我们最主要的工作。
The science piece is really working, and we just kept on investing more and more and more in it until now it is basically the primary and main thing that we're doing.
我们已经把原来旧金山的生物枢纽扩展到了好几个地方。
And we've expanded the original San Francisco Biohub to a handful now at this point.
比如纽约。
There's New York.
还有芝加哥。
There's Chicago.
目前的核心重点和统一主题是虚拟生物学计划，利用独特数据集来建模，从蛋白质的最小单元开始，逐步扩展到细胞乃至整个生物系统。
The real focus and the unifying theme at this point the virtual biology initiative around taking the unique datasets that are able to be generated in order to model effectively starting with the smallest pieces of proteins, but then eventually cells and whole biological systems.
我们的发展路径就是这样：有些人说这是个AI问题，想建一个前沿AI实验室，但必须同时搭配一个前沿生物学团队，专门负责获取和理解构建这些模型所需的数据。
But that's kind of how we've evolved, this idea that we talk about around that some of this is an AI problem, and you want to build a frontier AI lab, but you need to couple that with a frontier biology effort that can do the work of basically being able to understand and get the data that you need to actually be able to build these models.
因为跟语言模型不一样，互联网上根本找不到大量现成的生物数据。
Because unlike language models, there's just like a lot of data out there on the Internet.
生物学领域根本不是这么回事。
That's not really the case with biology.
当然，几十年来学术界和科学家们积累了不少数据集，但很多我们想用的数据，目前根本不存在。
I mean, there are obviously a bunch of different data sets that exist that academia and scientists have generated over the the decades, but a lot of the stuff that I think we wanna put into this, it doesn't exist.
对吧？
Right?
你希望看到以前没人看见过的东西，这正是我们一直在做成像研究的原因。
It's like you wanna be able to visualize things that people haven't been able to see before, which is why we were doing the the imaging work.
你想记录体内发生的事情，所以我们才做细胞工程这块的工作。
You wanna be able to record things that are going on inside the body, which is why we're doing the kind of cellular engineering work.
或者你想以以前做不到的方式测量炎症，这正是芝加哥生物枢纽专注于研发这类设备的原因。
Or you want to be able to measure things like inflammation in ways that haven't been possible, which is why the Chicago Biohub is focused on building those kind of devices and being able to do that.
这将从根本上催生全新的数据集，从而支持新型模型的建立。
And that will fundamentally create new types of datasets that will allow new types of models.
我觉得这特别令人兴奋，回到你刚才说的，如果科学界最需要的是工具开发，而这些工具现在能帮各个领域的科学家更快地开展工作，那这就是我们希望通过长期投入工具研发来提供的价值。
And I think it's just a very exciting thing that, going back to what you're saying, if the scientific field, it primarily needs kind of tool development that now is going to empower scientists across the field to be able to do their work faster, that's what we think we can provide through this kind of long term focus on tool development.
但我觉得从我们最初起步到现在，亚历克斯带领的这项工作其实有一条清晰的主线，那就是我们最早提交的申请项目，就是围绕单细胞测序展开的。
But I think there's a fun through line on where we started and, you know, bringing us to our work that Alex is driving now, is that our very first request for application, RFA here, was around single cell sequencing.
我们当时想研究单个细胞里转录出来的RNA。
And we wanted to look at sort of like the RNA that is transcribed in individual cells.
那时候虽然已经可以做了，但对不同细胞如何表达DNA的理解还很初级，所以我们最初只是资助方法研究，帮人们描述怎么做，好让其他人能共享这些技术。
And that was possible, but it was still pretty early on in understanding how different cells were expressing their DNA to the point where at the beginning we were just funding methods, like getting people to describe how to do it so that others could share that methodology.
后来我们就转而资助了人类细胞图谱项目，现在它已经是全球最大的单细胞转录组数据库之一，但科学家们开始发现，给这些数据做注释越来越难了。
And then that became us funding the Human Cell Atlas, which is now one of the largest databases of single cell transcriptomes, it was getting hard for scientists to annotate the data.
所以我们开发了CellByGene，这是一个非常简单的标注工具，供科学家用来使用这些数据。
So we built CellByGene, which was like a very simple annotation tool that scientists could use to make use of that data.
后来围绕CellByGene形成了一整个社区，他们自发贡献了越来越多的数据，这些跟我们完全没关系，我们既没创建也没资助过。
Then a community came around Cell by Gene, built around Cell by Gene, and started contributing more and more data that we had nothing to do with sort of creating or funding or making happen in the world.
现在，CellByGene已经成为大量转录组模型的基础数据源，科学界也在日常频繁使用它。
And now Cell by Gene is a corpus of knowledge that a lot of the transcriptomic based models are based off of and is used regularly by the scientific community.
但依然总有人提出批评。
But still there are always critiques.
说这不过是收集标签，没什么意义。
Like, this is just stamp collecting.
你们只是在收集一堆数据，根本没法从中提炼出真正的科学洞见。
Like, you're just gathering bits of knowledge sorry, bits of data, and we're not going to be able to pull scientific knowledge and wisdom and insights out of.
我们一时也答不上来。
And we're like, well, we didn't have an answer for a while.
直到后来，大语言模型突然成了热门话题，我们才惊喜地发现，原来它们正好能处理这种海量数据。
And then imagine our delight when large language models became a huge topic of conversation that could make sense of large amounts of data.
对我来说，我就在想，如果我们真能搞懂生物学的运作机制该多好。
And I just for me, it was like, what if we could actually understand how biology worked?
把生物学从发现型科学，变成工程型科学，系统性地理解生命体、活细胞是怎么运行的，弄清楚问题出在哪儿。
Move it from a discovery based science to an engineering based science where we could systematically understand how living beings, living cells, worked and be able to understand why things go wrong.
当我们看到那一刻，就知道，对，就是它了。
And so when we saw that moment, we're like, this is it.
这里真有可能发生一件大事。
Something really big could happen here.
亚历克斯，你最初在MetaFair起步，但你本来的计划是逐步组建团队，筹集风投，模型也在不断推进。
Alex, you started at MetaFair, but you were on the path to, you know, you'd assemble the team at evolutionary scale and you'd raise venture and you were making progress in your models.
马克和普莉西拉是怎么跟你说的，让你觉得这才是真正契合使命的方向？
What was the pitch from Mark and Priscilla where you said like, that's actually the right way to go after the mission?
对我来说，关键时刻是我意识到，他们真的把前沿人工智能和前沿生物学融合在了一起。
Well, I think for me it was really kind of the moment when I understood that, you know, they they really saw this as as an integration of frontier AI and frontier biology.
我逐渐确信，这正是科学的新纪元刚刚开启，人工智能能带来什么，现在才刚开始显现。
And I think I had developed conviction that, you know, this is really a new era of science that's just beginning, kind of what's gonna be possible with artificial intelligence.
我们现在处于大规模信息理论的时代，这些系统能预测下一个词元，还能从中学到世界的运行模型。
And, you know, we're in the age of information theory at scale, and we have these systems that can basically kind of predict the next token, and they can, learn world models from that.
它们能从数据里学习生物学的知识。
They can learn biology from the data.
所以很明显，要为下一个时代打造这样的新机构，就必须掌握前沿的人工智能。
And so, you know, I think that it was really clear that, you know, to build kind of that next kind of institution for the next era, you would really need to have frontier artificial intelligence.
还得有前沿的生物学研究。
You would have to have frontier biology.
必须把这两者结合起来，让模型真正从生物数据中学习。
You would need to start to put those things in feedback and really have models that are learning from the biology.
而且，还得有合适规模的顶尖人才。
And I think, you know, And you need the right scale on the right people.
所以我觉得，这条路就是最对的方向。
And so this just really felt, I think, like the way to do that.
你们一直在开发多种不同的模型。
There's a variety of different models that you all have been working on.
我觉得挺有意思的是，生物学领域最早的突破之一就是AlphaFold，当时谷歌的一个模型展示了大规模预测蛋白质折叠的可行性，而这一点以前没人觉得能实现。
And I think it's kind of interesting because some of the earliest breakthroughs in biology were things like AlphaFold, where there was a Google model that showed that you could do protein folding at scale in a really interesting way that people didn't realize was very tractable.
这还发生在后来那些大型Transformer浪潮之前。
And this was pre sort of the really big transformer waves that came later.
你们现在同时在好几个不同规模的项目上发力，对吧？
And then you're working on a variety of different things at different scale, right?
你们在做分子建模和蛋白质折叠的逐步改进。
You're doing incremental molecular modeling and protein folding.
也在做基于细胞的研究。
You're doing cell based stuff.
还在思考如何分析更大规模的生物系统。
You're thinking about interrogating larger scale systems in biology.
你觉得从微观到宏观，这种思路能走多远？
How well do you think that extends from sort of the micro to the macro?
你提到过从基础模块开始一层层搭建，但细胞行为的建模和蛋白质折叠的建模完全是两回事。
You mentioned almost starting with building blocks and building up, but modeling cellular behavior is very different from modeling protein folding.
数据完全不同，建模方式也不一样。
The data is very different, the modeling is different.
我只是好奇，你觉得这些本质上都一样吗？都是数据，然后训练模型？
I'm just curious, do you think it's all similar in terms of it's just data and you train stuff?
还是说，处理这些系统时，实际上存在一些根本性的差异？
Or do you think there's some differences in terms of how you actually have to deal with these systems?
肯定有些不一样。
Mean, are probably some differences.
关于这一点，你可能更清楚具体细节。
I mean, you can probably talk more to the specifics around this.
但我觉得每一层都会呈现出质的差别。
But I mean, I think each layer is gonna end up being somewhat qualitatively different.
要理解细胞的运作，就得先搞懂蛋白质之间的相互作用。
But you need to be able to understand the protein interactions in order to be able to understand how cells work.
所以，你不能跳过蛋白质建模，直接去研究细胞。
So you can't just go straight to cells in a way without understanding the protein modeling.
如果你想要理解免疫系统的工作方式，或者一堆细胞是怎么相互作用的，那在没搞懂细胞之前，这事很难做到。
And then if you're trying to understand something like the, you know, the way the immune system works or a bunch of cells interact together, then, you know, it's tough to do that without first understanding cells.
我的意思是，你或许能在非常抽象的层面模拟整个系统。
I mean, you might be able to, at, like, a very high level of abstraction, simulate a system.
但如果你想真正弄明白它到底是怎么运作的，最好还是从每一层开始，按层级逐步构建模拟。
But if you really wanna, like, understand how it's gonna work, you kind of wanna build the simulations at each level hierarchically.
所以我们现在的做法，就是从最基础的单元——蛋白质开始，一层一层往上搭。
So that's basically the approach that we're going through starting with the building blocks and the and the protein.
不过呢。
But yeah.
我觉得，针对每一层，你想要收集的数据类型也会不一样。
I mean, I think that there's gonna be different types of data that you wanna collect for each.
建模方法嘛，咱们拭目以待。
The modeling techniques, think we'll see.
我的意思是，这些方面整体上都会持续进步。
I mean, that'll all keep on advancing across the board.
但我认为，策略中很重要的一点就是得从底层开始，一层一层地构建起来。
But I do think that, like, a big part of the strategy is this view that you need to build it up hierarchically.
而且，在这个领域里，我们独特的地方在于，人工智能和湿实验团队从一开始就是作为一个整体在运作的。
And, you know, one of the things that's unique about us in this space is we are very intentional that the AI efforts and the wet lab efforts were a single effort.
我们花了很多功夫把这两部分真正融合在一起。
And we've done a lot of work to bring them together.
最棒的是，我们能主动收集那些帮助我们打通各个层级的数据。
And the really neat thing that we can do is really try to pull and gather data that helps us connect across sort of the hierarchy.
比如，你可以观察细胞内转录组的空间分布，看它们具体定位在哪儿。
You know, you can look at transcriptomics with space within a cell and look at where it's localizing.
我们还能用透明的斑马鱼，观察不同细胞在发育过程中的变化，尤其是大脑是怎么形成的。
We can look at translucent zebrafish and look at the development across different cells and when the brain develops.
我们还配备了传感器，可以监测细胞之间不同分子的交流情况。
We have sensors that allow us to look at cell cell communication in different molecules.
因此，我们可以有针对性地设计实验、收集数据，让这些信息在各个层级之间形成连接，为模型的构建提供关键支撑。
And so we can be strategic about the types of experiments and data we want to collect that helps us bridge across these, that makes it so that there's some connective tissue that helps drive the modeling magic that happens.
是的
Yeah.
我问这个问题的原因是，我以前是个生物学家。
The reason I asked the question, by the way, is I used to be a biologist.
我有生物学博士学位，曾经在湿实验室工作了将近十年。
I have a PhD in biology, and I worked in wet labs for almost a decade and everything else.
你们在找工作吗？
Are looking for a job.
这个我们待会儿再聊。
We can talk about that later.
这可不是拒绝。
It's not a no.
在
At
这个时候
this point
在我的职业生涯里，你知道的。
in my career, know?
听到你这么说，我。
Hearing, I
我特别喜欢自己这种干脆利落的反应。
love my aggressive reaction.
我就像是丹尼·格洛弗，还有丽莎·瓦芬那样。
I'm like Danny Glover, you know, and Lisa Waffen.
我快到退休年龄了。
I'm almost at retirement.
但我一直觉得，生物学各个层面之间缺少一种整合的视角——发育生物学家自己做自己的，分子生物学家也埋头干自己的，所以我才对这个特别好奇。
But I think one of the things that was always lacking was this integrative nature across the different layers of biology, and the developmental biologists would work on their own, the molecular biologists would be doing And different so that's what I was curious about.
生物学通常有两种视角：还原论的和系统论的，但这两拨人很少真正深入合作。
Typically, there's a reductionist view of biology and there's a systems view, and those people didn't really work together deeply.
所以你正在做的这件事，最让我兴奋的地方，就是你正在弥合这两者之间的鸿沟。
And so one of the exciting things about what you're doing actually is how you're bridging that.
所以这也就是我提出这个问题的由来。
And so that was kind of the basis for the question as well.
对。
Yeah.
如果我可以补充一点的话。
And if I could add something there.
我觉得我们现在正处在一个信息理论应用于生物学的时代。
You know, I think that we're in the age of this kind of information theory in biology.
生物学里存在不同层级的复杂性和等级结构，每一层都是由更低层级构成的。
And so, you know, there there are levels of of complexity and hierarchy in biology, and kind of each level is is made up of and, you know, constituted by the lower levels.
如果你想获得更完整的描述，想构建出能真正概括并数字化回答实验室里那些实验问题的系统，就必须在每一层都打下正确的建模基础。
And so as you want to have that kind of more complete description and you want to have systems that can really generalize and begin to actually answer, you know, experimental questions digitally that you could ask in the lab, you need to have kind of the right basis for modeling at every level.
我觉得我们独特的地方在于，就像普丽西拉和马克说的，要在每一层都构建信息，收集这些层级之间的连接点，同时还要以足够的规模来做，这样才能揭示出背后的信息架构。
And so I think what's really unique about what we can do is to, as Priscilla and Mark were saying, really build information at each of these different layers, collect them, collect kind of those connection points, but then also really kind of do it at the scale that will reveal that underlying information architecture.
这对于我们构建能够回答新实验问题的数字模型至关重要。
And that's gonna be really critical to actually be able to build digital representations that can answer new experimental questions.
让我最受启发的一点，是普里西拉说的——我们对生物学其实还有很多根本不懂的地方，要是能搞清楚呢？
One of the things that inspires me most about this effort is really what Priscilla said, which is like, well, there's so much we actually don't understand about biology and what if we could?
我觉得这跟其他很多很有趣、很有用的AI问题很不一样，那些问题我们通常是在试图复制人类行为。
Which I think is actually very different from lots of other incredibly interesting and useful AI problems we attack where we're trying to replicate human behavior.
很多那方面的数据，网上早就有了，或者已经被收集起来了。
I'm like, a lot of that data's on the internet or captured.
我们不需要完全理解人类的所有行为，也能预测很多东西。
And without pretending to understand all human behavior, can predict a lot of it.
我觉得你们发布的内容里，最有趣的一点其实是你们提到的机制可解释性——我们能不能从模型认为正在发生的事情里，提炼出新的知识？
I thought one of the most interesting things in your release was actually, you know, the mechanistic interpretability stuff you alluded to, which is, can we actually extract new knowledge from, you know, what the model believes is happening?
对吧？
Right?
你能稍微讲讲这方面吗？
Can you talk a little bit about that?
是的，我对这个特别感兴趣。
Yeah, I'm really excited about that.
我觉得，在机制可解释性方面，传统上它被用在大型语言模型上，目的是弄清楚这些模型的表征空间到底是什么样子。
So I think, you know, in mechanistic interpretability, kind of traditionally it's been applied to large language models with the goal of understanding, you know, kind of what is the representation space of a large language model?
它们是怎么处理信息的？这些过程和我们对世界的直觉理解之间，有没有真正的联系？
How does it compute things, and does that really connect to, you know, what we understand about our intuitive understanding of the world?
因此，现在已经发展出一套非常丰富的工具，可以帮助我们提出这些问题。
And so there's, I think, this really rich toolkit that has been developed to start to be able to ask those questions.
那这套方法对生物学意味着什么呢？
So kind of what does that mean for biology?
我们训练的一类模型是蛋白质语言模型。
One of the classes of models that we train are these protein language models.
它们主要是基于蛋白质的序列数据进行训练的。
So they're really, you know, trained on the codes of proteins.
所以，模型学到的任何生物学知识，都是在训练过程中自然涌现出来的。
And so anything they learn about biology is is kind of emergent.
我们发现，它们能学会像蛋白质结构和功能这样的东西，而这些都不是直接教的，纯粹是从预测下一个词的任务里自己冒出来的。
And we've seen that they can learn things like biological structure and biological function, and that's just kind of emergent from this token prediction training task.
当我们思考这些模型的机制可解释性时，实际上是在探索未知领域，因为这些模型是在数十亿条蛋白质序列上训练的。
So, you know, as we think about mechanistic interpretability in those models, we're really seeing the unknown because the models have been trained on billions of protein sequences.
它们的训练数据涵盖了已知和未知的生物学信息。
They've been trained on, you know, both known and unknown biology.
但它们却逐渐形成了某种表征，能够对应上几个世纪以来积累起来的还原式生物学图景。
And yet they're developing these representations that start to kind of capture things that we can really see correspond to that reductive picture of biology that's been built up over the centuries.
因此，你开始能把那些我们一无所知的蛋白质，和那些我们有一定了解的蛋白质联系起来，因为模型的表征空间里存在一种底层的结构语法把它们串在一起。
So kind of you can start to connect the dots between proteins where we kind of really don't know anything about them with proteins where we do know something because there's that kind of underlying structure grammar that's linking them in the representation space of the model.
极端情况下，我们或许能理解以前从未了解过的身体系统，或是某种新疗法的作用机制，因为我们能向模型提问，深入探究它的表征。
And at the extreme, it could be, you know, we're going to understand systems in the body that we didn't before or the mechanism of action for a new treatment because we can ask the model, right, interrogate that representation.
没错。
That's right.
我们的期望是，真正弄清楚模型做出预测的底层原理。
The hope is that you kind of really learn the underlying basis for how it's making the predictions.
这样一来，我们就打开了黑箱，真正看懂了模型所表征的生物学内容。
And so you open up the black box and you can actually understand kind of the biology that the model is representing.
所以我是替朋友问一下，你们是不是都觉得，靠风险投资支持的公司才能对世界产生影响？
So asking for a friend, you know, you guys all believe in venture backed companies as a way to have impact on the world.
是数据来自斑马鱼，还是数据范围够大，或者是实验操作，又或者是规模上的优势？
Was it like collecting data on zebrafish or the span of the data, or the wet lab work, or just the scale?
到底是什么让这个项目更适合放在大型非营利合作生态里，而不是做成一家风投支持的公司？
Like what makes this a better fit for this big nonprofit ecosystem effort versus a venture backed company?
我觉得我们只是想把工具提供给整个科学界。
Well, I think we just want to give tools to the whole scientific community.
我的意思是，要说真想赚钱，其实我们也不是不能把它做成商业项目。
I mean, like, so I I think in order to have the biggest impact, I mean, part of it is just we're I mean, it's not actually clear that we couldn't run it as a business if we wanted to.
但我觉得，通过开源的方式，能让更多科学家更快用上这些工具，影响力才会更大。
I just think that we'll have a bigger impact by getting this in more scientists' hands quicker by doing it as open source projects instead.
所以，嗯，我觉得这就是我们的思路。
So, yeah, I mean, I think that that's kind of the approach.
不过我也说不准。
But I don't know.
这是个挺有意思的问题。
It's an interesting question.
我不确定，毕竟你之前是作为一家营利性公司来做这些建模的，但后来就会遇到一些瓶颈。
I'm not sure that I mean, obviously, you were doing it as a for profit company, a bunch of the modeling before, then you run into certain issues.
你得筹集一大笔钱才能搭建起计算集群。
You have to raise a large amount of money in order to build the compute clusters.
我觉得在很多方面，数据反而是更大的限制因素。
I think in a lot of ways, the data is actually even more of a constraint.
因为如果你看看这些模型和语言模型相比，它们规模更小，但之所以小，是因为可用的数据本身就少。
Because if you look at the scale of these models compared to language models, they're smaller, but they're smaller because the amount of data is less.
要获取这些数据，可不是找家工厂付钱就能批量生产的。
In order to get the data, it's not just like there's some factory somewhere that you can pay to produce the data.
你得真正发明新的、原创的科学方法，才能完成像我们在纽约做的细胞工程，或者芝加哥那些设备的工作——这正是为什么我们谈‘前沿生物学’和‘前沿AI’时，前沿生物学意味着你必须做真正的科研，去推动新的生物技术，才能观测到那些能产生模型所需数据的现象。
You actually need to invent new, novel, scientific approaches to be able to do the, know, for example, the type of cellular engineering we're doing in New York or the types of devices in Chicago, which is why when we're talking about this concept of frontier biology and frontier AI, the frontier biology is you need to do real science to advance different biological methods in order to be able to observe the things that create the data that go into the model.
所以这根本不是能随便买来就用的现成东西。
So it's not just like an off the shelf thing that you can create.
这确实是个不小的工程。
Now that's a pretty big effort.
我不觉得生物科技领域里还有多少类似的事情在做。
I don't know that there are, like, that many things like that that are done as as biotechs.
我觉得关键在于我们目标的规模，以及我们愿意为此投入的时间跨度。
I think it's just the scale of the ambition of what we're doing, the time horizon over which we're committed to doing it.
我觉得其中一个想法是，如果你在打造这么复杂的工具，最好能有十年到十五年的长期规划。
I think part of the theory is, like, if you're building tools that are this complicated, you kinda wanna have a ten to fifteen year time horizon on on building out these efforts.
至于所需的资金规模，理论上也不是不能靠一个资金极其雄厚的初创公司来完成，但我觉得这样更合理。
And then the scale of capital required I mean, I guess there's no rule that said that you couldn't do it as an incredibly well funded startup, but I think that this just made more sense.
而且，从战略上讲，不纠结于怎么靠不同项目赚钱，反而更简单。
And then it also is is simplifying strategically to not have to think about you're gonna make money with the different things.
我们的目标就是让这些模型真正落到大家手里。
I mean, we just we wanna get the models in people's hands.
我们把它们全部开源发布。
We release them as open source.
我觉得这是一件非常有价值的事。
I think that that's a very valuable thing to do.
再说一遍，我们的目标并不是要治愈这些疾病。
And again, I mean, the theory isn't that we're gonna cure the diseases.
我们做不到。
We're not.
我们的目标是帮助加速整个科学领域的进展速度。
It's that we wanna help accelerate the pace of progress for the whole scientific field.
作为这里最不擅长赚钱的人，我觉得我们这种中立的非营利性质，反而更能吸引大家参与进来。
As the person least experienced with making money here, I would say that there the sort of neutral nonprofit nature of our work actually helps harness more people to enter this effort.
要真正实现理解整个人类生物学、预防和控制所有疾病的目标，确实需要整个学术界和生物技术行业齐心协力，因为这背后有大量的人才。
And to actually achieve the mission of like understanding the totality of human biology and to cure, prevent, manage all disease, you actually do need the entire academic biotech industry to come together and to work on this in a sort of unified way, in part because there's a lot of talent out there.
我们不能让任何人才被排除在外。
And it's not helpful to leave any talent, exclude any talent from the effort.
还有大量罕见疾病需要关注。
And there's a super long tail of diseases.
有一些常见的疾病。
There are the common ones.
就连常见的病，如果你把心脏病、癌症、神经退行性疾病拆开来看，甚至把痴呆或抑郁症再细分，也会发现有无数子类，越来越专门化。
And even the common ones, I think if you unbundle heart disease, cancer, neurodegenerative diseases, even if you unbundle like dementia or depression, there are many, many, many subcategories that become more and more niche.
这还没算上那些极其罕见的罕见病。
And that's not even looking at the long, long tail of rare diseases.
这些病常常被忽视，因为在考虑如何高效改善多数人生活时，它们容易被排在后面。
Those often get orphaned and don't get brought along when we're sort of looking at what the most efficient way to impact the lives of many.
但如果你把努力分散开来，把工具交给更多人，就会有人开始说：其实吧……
But if you sort of decentralize the effort and put the tools in many people's hands, you start getting people who are like, you know what?
我对脊髓性肌萎缩症特别感兴趣，这对我来说是件特别重要的事。
I am super interested in spinal muscular atrophy, and that's something I care deeply about.
只要你把工具交到这样的人手里，他们就能推动进展。
And if you put the tools in that person's hands, they're gonna be able to make progress.
从某种角度说，如果你非要集中资源、下大注，可能根本不会选这种病——毕竟它只影响一小群人。但如果我们能搞懂这种病的机制，反而能解锁更多关于人体运作的通用知识。
In a way, if you had to focus your efforts and make big bets, you probably wouldn't because it's just a niche individual small group disease that actually will in turn, if we can understand that disease process, helps us unlock knowledge about a lot more about how the human body works.
你
Do you
觉得这项工作会最先影响哪些疾病领域吗？
have any thoughts or predictions in terms of what disease areas this work will impact first?
我知道要预测这些事情真的很难。
I know it's very hard to be predictive about these things.
但基于这项工作的特点和模型的性质，短期内或中期你最看好的方向是哪些？
But just given the nature of the work and the nature of the models, are there areas you're most optimistic about in the short to medium term?
至少我并不是这么想的。
That's actually not how I think about it at least.
我的想法是，我们真正想做的是理解生物学的运作机制。
The way I think about it is like we want to understand how biology works.
理想情况下，你应该能说：我了解了这个人的基因组。
The ideal world is you would say, I understand the genetics of this person.
所以我更关注个体层面的人。
So I want to think about people at the individual level.
我想弄清楚这个人的基因情况。
I want to understand the genetics of this person.
我想了解他们患各种疾病的风险。
I want to understand the risks they have to different illnesses.
我想搞明白基因变异、蛋白质和疾病机制之间的内在联系。
I want to understand the mechanistic connection between, say, a gene variant, a protein, and a disease process.
因为只要理清了这个链条，就能为他们量身设计蛋白质、药物，真正实现精准干预。
Because if you understand that through chain, then you can design a protein, design a drug, bespoke to them, and actually make an intervention.
现在，我相信我们都经历过生病的时刻。
And right now, I'm sure we've all had experiences being sick.
一旦遇到稍微有点不寻常的情况，就得去PubMed找论文，看补充材料，然后一头扎进方法部分。
And if you have something that's even remotely nonstandard, you go into PubMed, you look up a paper, you look up the supplement, and then you start going through the methods.
你心里会想：我这种状况在论文里有被涵盖吗？
And you're like, am I represented in this paper?
我们全都在瞎猜。
And we're just making guesses.
我们根本缺乏机制上的理解。
We really have no mechanistic understanding.
我们只是说，嗯，你跟我们研究过的那些人有点像，这个药可能影响了我们认为有关的通路。
We're saying like, okay, you're kind of like these people that we studied, and this drug kind of impacts the pathway that we think is implicated.
先试试看会不会有反应吧。
Let's try and see if anything happens.
然后时间就过去了。
And time passes.
有时候有效，有时候没用。
And sometimes it works and sometimes it doesn't.
所以我的目标是把每个人当独立个体来对待，搞清楚背后的机制，然后精准干预。
So my goal is to be able to treat the individual as an individual, understand the mechanisms, and be able to intervene.
不同的疾病在完整这条因果链条上所处的阶段也不一样。
And there are different diseases that are at different stages of filling out that whole through line.
因此对某些疾病来说，我们首先只想弄清楚哪些基因变异真正致病，哪些不是。
And so for some diseases, you just want to understand which gene variants actually cause disease and which don't.
这本身就能让患者感到特别有力量。
And that in itself can be super empowering to patients.
而且在某些疾病上，如果我们已经弄清了整个机制链条。
And if beyond that, there are some diseases where we understand the chain.
只是还无法干预并改变某个特定蛋白质的功能。
We just can't intervene and change a specific protein function.
这同样让人非常兴奋。
That's super exciting too.
比如，如果我们能设计出一种蛋白质来真正改变生理状态，那就真能治愈患者了。
Like, if we could design a protein to actually change the physiology, then we can actually cure someone.
但对我来说，能帮助我们理解一个人为什么会生病，这种贡献同样令人激动。
But to me, like, that is just as exciting as understanding contributing to our understanding of like how someone gets sick in
第一次。
the first time.
这是一个非常令人振奋的愿景，因为你实际上是在说，可以用通用的工具为每个人提供高度个性化的治疗。
And that's a very exciting vision because you're basically saying you can bring generalizable tools to provide very personalized things for each individual person.
是的。
Yes.
这种做法的厉害之处就在于，你构建的这些大型模型可以应用到任何地方。
And that's the power of the approach, is you have these big models that you build that can then apply anywhere.
我记得你之前提到过，你打算在一百年内预防所有疾病。
I know that you mentioned earlier that you were gonna try and cure prevent all diseases within one hundred years.
你还说，考虑到人工智能的飞速发展，这个目标可能实现得更快。
And you mentioned, hey, could actually be sooner now given all the advances in AI.
你有没有想过，我们大概什么时候能更接近这个目标？
Do you have some thought of when we think we'll be closer to that goal or something?
我觉得会更快，我挺乐观的。
Mean, I'm optimistic it'll be sooner.
我觉得复杂的地方在于，这其实是个动态系统。
I mean, I think the the thing that's complicated is that it's a dynamic system.
对吧？
Right?
如果你解决了一个问题，后面肯定还会有新的问题冒出来。
So if you fix something, there will obviously be future things that you need to work on.
我觉得我们现在知道的这些问题，不会是唯一需要解决的。
So I don't think that the current set of things that we're aware of are gonna be the only things that need to get worked out.
不过我也说不准。
But I don't know.
我觉得AI在这方面的进展真的非常令人振奋。
I think that the progress with AI is really is obviously very exciting on this.
另外，我想补充一下你刚才说的，我们更关注的是整个系统，而不是具体的某种疾病。
The other thing that I'd say, just adding to what you were saying a second ago, is we really look at more systems than specific diseases.
比如，有一个特别重要的领域是炎症，特别值得深入理解。
So for example, one area that seems really important to understand is inflammation.
我们之前已经聊过很多次了。
We talked about this a bunch.
这正是芝加哥生物枢纽的重点研究方向。
This is a big focus of the Chicago Biohub.
关于这个，已经有大量数据了。
There's a lot of data on that.
很明显，炎症和多种疾病都有关联，但我们觉得，与其一个个研究具体疾病，不如先更全面地理解炎症本身，这样其他公司就能用这些工具开发针对性的疗法。
It seems quite clear that it's connected to a bunch of different diseases, but rather than studying the specific diseases, we think that by trying to understand inflammation more broadly, that will make it so that other companies that can then use these tools can work on specific therapies.
另一个例子是免疫系统，我觉得它特别适合用来研究我们正在做的细胞工程工作——从蛋白质到细胞，再到体内复杂的动态系统，免疫系统确实是个很好的切入点。
Another example is I think that the immune system, I think, is a very good case to study for some of the work that we're doing in cellular engineering and When we're kind of ladder up from proteins to cells to whole dynamic systems within the body, I think that that one makes sense.
它本身就很有优势。
It's sort of privileged.
免疫细胞能在全身自由移动，等等。
The cells can travel around through the body, all that.
这在应对各种疾病方面显然起着关键作用。
Obviously, that has a big part in addressing different diseases.
我们该怎么让免疫系统运作得更好？
How do you make the immune system function better?
但具体怎么打通最后一公里，我觉得还是得靠生物科技公司或者独立研究的学者来推进。
But exactly how do you connect that last mile I think is gonna be more something that biotech or other academics individually studying things will be better suited to do.
所以我们一直在思考如何打造这套工具，帮助其他人加速进展。
So this is kind of how we think about building out the toolset that just helps accelerate all these other folks.
不管最终需要十年还是希望别超过一百年，我觉得这对普通医生、患者，甚至每个人——毕竟人人都可能是患者——来说都很重要，让大家能看到当前取得的进展。
Whether the timeline is ten years, hopefully less than one hundred now, I think this is useful for maybe your average doctor or patient, human being, everybody's a patient, to think about like what's externally visible in the progress here.
你以前在加州大学旧金山分校长期接触过病人。
You worked with patients for a long time at UCSF.
医生们该关注哪些迹象？
Like, what should doctors look out for?
如果我们真在加速进展，普通人又该留意什么？
What should people look out for if you're actually accelerating progress?
这部分我特别兴奋，尤其是亚历克斯和他的团队推出的这项成果。
This is the part You know, I'm super excited about the progress, especially with this launch that Alex and his team have put forth.
我觉得科学的发展马上就要提速了。
And I think it's very clear that science is gonna start moving pretty quickly.
但我还不太清楚的是，这些成果到底该怎么落地到临床，具体会是什么样子。
And I think the thing that's less clear to me is exactly how we translate to the clinic and what that looks like.
我觉得真正需要改变的，是我们做临床研究的方式。
And I think what has to change is actually the way we do clinical research.
我希望我们能大大缩短基础研究和患者受益之间的距离。
And my hope is that we're really shortening the distance between bench research and patient impact.
但中间还有很多环节，需要一线医护人员开动脑筋，思考如何安全地落地应用。
But there's a lot of steps there that we need people who actually take care of patients to think creatively and think about how to deploy safely.
这方面我们还有很多工作要做。
And that's a gap that we have some work in.
我们在UCSF的CRISPR治愈项目中，和詹妮弗·唐尼合作。
We partner with Jennifer Downey on our CRISPR Cures program at UCSF.
所以我们正在试着摸清，随着研究进展越来越快，成果落地的方式该怎么调整。
So we're dipping our toe in understanding how the deployment of research needs to change given how quickly research will be progressing.
不过这一点还在摸索中，还没完全成型。
But that one is still, I think, still shaping up.
也许我可以谈谈我们最近的这次发布。
Maybe I could say something about our most recent launch
因为我想
because I
觉得应该
think it
还有，呃，麻烦你了。
also Oh, yeah, kind please.
我们得直接问问你的看法。
We should ask you explicitly about it.
对。
Yeah.
其实，就在一周前，差不多这个时候，我们发布了新的ESMFold。
So, you know, because I guess it was just a week ago, about now, so we announced the new ESMFold.
这其实是一个面向蛋白质生物学研究的开放平台。
And so this is basically an open system for scientific discovery in protein biology.
它是一个经过训练的蛋白质生物学全局模型。
It's a world model of protein biology that's been trained.
这是一个基于语言模型的系统，训练数据包含了数十亿条蛋白质序列，能够自动学习蛋白质生物学中的深层规律。
It's a language model based, so it's been trained on billions of protein sequences, kind of learns these emergent representations of protein biology.
我们可以用它来预测原子级别的蛋白质结构，而且速度非常快。
And then we can use it to make predictions of atomic resolution protein structure, and we can use it to and it's really fast.
它的运行速度简直快得惊人。
So it's blazing fast.
它很好地展现了在结构预测中速度与精度之间的最优平衡。
So it's kind of you know, illustrating this Pareto optimal frontier of kind of speed and accuracy in structure prediction.
这让我们能够探索浩瀚无边的蛋白质世界，全面刻画其中的各个角落。
And so this allows us to kind of characterize, you know, really vast kind of stretches of the protein universe.
我们已经预测了超过十亿个蛋白质的结构，还通过机制可解释性找到了它们之间的关联特征。
So we folded over 1,100,000,000 proteins and predicted their structures and identified kind of features connecting all of them through mechanistic interpretability.
但我认为这个模型最令人兴奋的地方，在于它提供了一个通用的蛋白质生物学模型。
But I think the thing that I thought was most exciting about this model is this really general model of kind of protein biology.
因此，你可以把它当作一个理解蛋白质世界的通用框架。
And so you can use it as a world model.
你可以真正开始在这个世界模型的空间里搜索，设计出全新的蛋白质。
You can actually really start to search the space of the world model to design new proteins.
它在几乎所有结构预测基准上都达到了顶尖水平，尤其是在蛋白质-蛋白质相互作用和蛋白质-抗体相互作用方面，这对治疗设计至关重要。
And it's really hitting state of the art across pretty much every structure prediction benchmark, and especially on protein protein interactions and protein antibody interactions, which is really critical for therapeutic design.
我们发现，现在可以直接用这个模型来设计蛋白质，甚至设计单链抗体。
And so what we found is you can actually now use the model to design proteins and to design actually single chain antibodies.
所有这些都可以在数字环境中完成，然后只需少量实验尝试——比如用一个96孔板，从成千上万条数字模拟路径中筛选，实际合成96种蛋白质，在实验室里快速完成一轮测试。
And so you can do all of this digitally and then, you know, really in a small number of experimental trials, basically like a 96 well plate, you know, select from hundreds of thousands of trajectories digitally, actually synthesize, you know, 96 proteins, test them in the lab in a really kind of short, easy experimental cycle.
我们在那里找到了纳摩尔级别的结合剂。
And we found nanomolar binders there.
这正是治疗活性所需要的水平。
And so, you know, that's really the level for therapeutic activity.
所以我认为，这清楚地表明，这类通用模型真的可以做到——我们并没有专门针对抗体去设计这个模型。
So it's really, I think, showing that you can have these kind of general purpose models that can you know, we didn't design a model for antibodies.
我们也没打算让它专门结合某一个特定靶点。
We didn't design a model to, you know, to be able to bind one particular target.
展开剩余字幕（还有 282 条）
我们只是设计了一个能理解蛋白质的模型，结果蛋白质设计就自然而然地出现了。
You know, we just designed a model that could understand proteins, and you kind of get protein design as an emergent property.
这也体现了开放科学和开源的力量，因为我们把这个工具当作一个开放的发现平台发布出去了。
And then I also think it illustrates kind of the power of open science and open source because, you know, we release this as basically an open discovery engine.
所以任何人都可以在此基础上继续开发。
And so really anyone can build on it.
以往那些耗时耗力的实验室实验，得在高通量筛选中测试成千上万甚至上百万种抗体，现在完全不一样了。
And so it takes what are these really intensive laboratory experiments where, you know, you have to screen through hundreds of thousands or millions of antibodies and high throughput screens in the lab.
你只需启动一个计算实例，跑一跑就能生成抗体了。
And, you know, you can really just kind of spin up an instance and compute and now, you know, be able to generate antibodies.
你们在做抗体筛选时，其实还用到了那些数据，之后在细胞里验证了PDL的表达。
You should say more about sort of like we took that data when we looked at an antibody screen, and then we validated we looked at PDL in cells.
然后又用冷冻电镜观察了结构，这些实验结果和模型预测完全相互印证。
And then we looked at it under the cryo EM and sort of how all that complemented validated what you were seeing in the models.
没错。
That's right.
是的
Yeah.
所以我认为，真正去实验室里对这些分子进行表征至关重要。
So, I mean, I think it's really critical, you know, to actually go and characterize these molecules in the lab.
我们这里有一个结构生物学中心。
We have a structural biology center here.
我们拥有极其强大的冷冻电镜设备。
We have incredibly powerful cryo EM microscopes.
因此我们能够从生物物理和功能层面深入观察这些蛋白质。
And so we're really able to kind of look at these proteins biophysically and functionally.
我们为多个具有治疗意义的靶点设计了蛋白质，并且成功验证了它们的功能，这
And so, you know, we designed proteins for several therapeutically relevant targets, and we're able to confirm their function and It's
当一切按预期运行时，效果真是相当惊人。
some pretty light when it works the way it's supposed to.
对，这确实非常了不起。
Yeah, it's very amazing.
我们可以看到结合位点的原子级结构。
We're to look at the structure So you can see atomic resolution kind of at the binding interfaces.
没错。
Correct.
我知道你们的很多工作都集中在基础研究，搭建基本框架上。
I know a lot of your work is really focused on basic research and kind of building out the fundamentals.
如果看实际转化成药物或药物开发，一个临床试验通常要十五年。
If I look at actual translation into drugs or drug development, often a clinical trial will be fifteen years.
成本大约是十五亿美元，其中约五千万用于前期分子研究，花上几年时间。
It'll cost $1,500,000,000 About 50,000,000 of that often is the molecule in preclinical work, and it's a few years of work.
剩下的十四亿五千万和十多年时间，都花在药物开发这一块了。
And then the other 1,450,000,000 and decade plus is actually the drug development side of it.
这其中很多环节被监管问题卡住了，还有受试者招募，各种因素都有。
A lot of that seems to be gated on some regulatory issues, some of it's recruitment, it's a variety of things.
但很多失败其实是因为药物在吸收、毒性等方面出了问题。
But a lot of it also has to do with the failure of drugs and trials around things like absorption or toxicity or things like that.
你有没有考虑过去解决分子设计和构思这一环节的问题？还是说你的主要精力还是放在基础生物学和最初的分子研究上？
Have you considered at all tackling that other chain of sort of molecular design and thinking, or is the primary focus more on the basic biology and sort of the initial sort of molecules?
我的想法是，至少我希望通过构建一个全面的模型来理解细胞是如何工作的，这样就能预测出脱靶效应。
I mean, at least my hope in building this like comprehensive model of how, you know, cells work is actually also being able to predict off target effects.
我觉得
I think
你可以做到
you can do some of
这一点
that
其实跟生物模型有关。
actually with biological models.
因为目前有些脱靶效应，是我们之前根本不知道的，比如你的肾细胞其实也表达了这种受体。
Because right now some of the off target effects are we just didn't know, you know, your kidney cell also expressed this receptor.
然后我们在人体测试时，就发现确实出现了这种情况，并观察到肾毒性。
And then when we test it in human, like we see it happening and we see renal toxicity.
如果你有一个涵盖所有不同类型细胞的单细胞图谱，其中一些细胞类型甚至在建模之前都没被预测到，那你就能找出哪些细胞实际上表达了你原本以为只针对特定目标的受体，从而在进入人体试验前就预测到这些下游效应。
And so being in If you have a single cell atlas that looks at all the different cell types, some of which actually were not predicted before we modeled them, you can start looking at which cells actually do have receptors for the target you thought you were exclusively targeting and be able to predict some of these downstream effects before we get into the human trials.
我认为，这正是转录组模型最令人兴奋的应用之一——它能帮助我们真正理解，当你进行干预时，不同细胞会如何反应。
And I think that that's actually one of the more exciting applications of like a transcriptomic model to understand actually how the different cells will react when you intervene and do something.
但当你考虑给药方式和患者护理时，你就得开始思考：究竟该优先解决哪种疾病？
But I think when you think about delivery mechanisms and patient care, that's where you start having to be creative about when you ask like, what disease do you want to care first?
有些疾病更容易递送治疗药物，或者风险与收益的平衡更合理。
There are certain diseases that will be easier to like deliver a therapeutic to, or the risk reward makes more sense.
我想我们都受到了去年CHOP团队治疗小KJ的启发，他们成功用CRISPR疗法修正了他体内一个会导致严重神经毒性和终身残疾的基因突变。
And, you know, I think we were all inspired by baby KJ, I think last year now, when the team at CHOP was able to deliver a CRISPR therapeutic to edit a mutation that he had that would have inevitably led him to significant neurotoxicity and altered his life.
但这个疾病之所以被选中，是因为我们需要靶向他的肝细胞，而当时我们已经有办法轻松地让药物在肝脏中发挥作用。
But we were able to, that disease was very carefully chosen because we needed to target his liver cells and if we could easily deliver a product that would work in his liver.
正是在这种时候，选择合适应用场景的创造力和判断力，才能帮我们打开首批应用的大门。
And I think that's when the creativity, the wherewithal to choose the right applications can help us unlock the first applications.
也许我还想补充一点。
Maybe something just to add to that also.
你知道，你刚才描述的其实是传统的药物研发流程。
You know, because, I mean, kind of you described the conventional, you know, drug development process.
对吧？
Right?
我觉得这些工具真的有可能对整个流程产生巨大影响。
And I think, you know, these kind of tools have the potential to have a lot of impact on that process.
但更有趣的是，我们得开始思考那些全新的可能性。
But, you know, what's interesting is to really start to think about kind of the new paradigms that can open up.
如果开发药物、设计分子、走过所有这些阶段的门槛都大大降低了，那意味着什么？
And, you know, what does it mean if if you can you know, the barrier to develop a drug, to design a molecule, you know, to kind of get through all of those stages is so much lower.
有了可编程的生物学，我们真的可以为每个病人量身定制药物。
And so you have programmable biology, and you can, really start to create a medicine for every individual patient.
我觉得这将对药物研发方式和医学的未来产生深远影响。
I think that has enormous implications for how we do drug development and what the future of medicine looks like.
等到FDA接受虚拟临床试验，比如一期试验基于某个患者的个人数据时，那将是令人兴奋的一天。
It'll be an exciting day when the FDA accepts like a virtual clinical trial for the phase one or something, or, you know, it's based on some person's view of that person.
是的。
Yeah.
或者更进一步，想想那些能加速研发的具体机制——比如，如果研究人员能更准确地预测药物对肾细胞的影响，或者因为有了更全面的理解而对毒性有更强的判断，他们就会更愿意尝试更多项目。
Or even short of that, like thinking about the specific mechanisms where you see this acceleration, like I imagine if people feel like they can predict impact in kidney cells or have a stronger perspective on tox because they have this broader understanding, they'll be willing to try many more programs.
对。
Yeah.
患者招募的方式也可能发生变化。
The recruitment could also change.
我们有个项目叫‘Rare as one’，核心想法是，大多数人只关注最常见的疾病，但还有一大类罕见病被忽视了。
We have this program, Rare as one, and the basic idea is that a lot of people focus on the most common diseases, but there's this long tail.
这些罕见病的商业回报对药企来说不太划算，但如果你能让患者群体联合起来，主动说‘我们愿意试一试这种实验性药物’，那么情况就不同了——因为你提到的那些成本，占了整个研发支出的很大一部分，如果能把这部分成本降下来，经济模型就完全不一样了：你能更容易地开发出药物，再匹配上一群有共同需求的患者，整个模式就变得可行了。
And the economics don't quite work out for companies to focus on those diseases, but if you can make it so that the groups of patients can kind of come together and organize and say, hey, we would take an experimental drug on this, then it actually, because of the cost that you're talking about and how that's a huge amount of the overall cost, if you can flip that, then it actually makes it so the economics make a lot more sense to then if you can generate something more easily and you can pair it with a group of people.
从科学和工程的角度看，一个有趣的现象是，人们常常在常见问题上撞得头破血流，疾病领域也是如此。
I think one of the interesting things from science and engineering is that often can hit your head against the wall on the common problems, and in this case diseases.
但很多时候，真正让你理解整个系统的，反而是那些罕见的、奇怪的、个案中出现的异常现象。
But a lot of times you learn a lot more about a system from finding some kind of rare or weird side thing that's happening in each case.
我不确定。
I don't know.
我觉得这一直是这个领域中一个很有趣的部分，而且和现在的情况特别契合——你将能推动大量新的想法得到尝试，并让它们更容易被验证。
I think that that's always been kind of an interesting part of this that actually connects pretty well to this because now you're gonna be able to enable a long tail of new kind of ideas to get tried and enable them to potentially get tested more easily.
是的。
Yeah.
关于罕见病，你这点说得特别好。
That's a really good point on rare.
在我们的罕见病患者群体中，首先，他们非常鼓舞人心，也很有力量。
In our rare disease cohorts, first of all, they're incredibly inspiring and powerful.
患者团体自己组织起了患者登记系统、自然病史数据库和生物样本库。
But patient groups are self organizing patient registries, natural history registries, biobanks.
他们甚至在自行组织临床试验。
They're organizing their own clinical trials.
有一个疾病群体在短短三到五年内就推动了基因疗法的进展，而以往这类工作通常要花上几十年。
There's gene therapy that one disease group has moved forward over the course of like, I want to say like three to five years rather than decades.
而且速度这么快，是因为患者自己已经组织好了科学家或医生需要的资源。
And the speed is so fast because the patients themselves have organized the resources that a scientist or a clinician might need.
这真是太了不起了。
And it's incredible.
但我认为，在某种程度上，你确实需要这样的机制，因为未来会出现更多全新的东西。
But I think to some degree you're going to need something like this because there are gonna be many more new things that can get created.
但这并不意味着，对于普通人群，我们就该降低过去那种严格的审核标准。
But that doesn't mean that for the general population that you're not gonna want the same level of vetting that we've had historically.
但让那些想走在前沿的人有能力去尝试，我觉得也会很有帮助。
But making it so that people who wanna be on more of the frontier have the ability to do that is I think also going be pretty helpful.
是的，让参与者自愿加入临床试验，这正是一场重要的转变，有望整体加速生物学的发展。
Yeah, letting people opt in to be part of trials I think is one of the big shifts that is starting to happen but could really help accelerate biology in general.
你们三位在不同时候都提到了开放生态系统在这么广阔领域中的力量。
All three of you have mentioned at different points like the power of open ecosystems in such a large space.
我觉得你们刚才谈到的开源理念，以及数据收集的广度和多样性，同样适用于语言模型和多模态人工智能领域。
I think some of that logic around open source and the breadth or diversity of data collection that you guys were describing, it should also apply in the language model world and the multimodal AI world.
你觉得这是对的吗？
Do you think that's right?
你们在这里做的工作，有没有改变你对人工智能和元宇宙的看法？
Does any of the work you're doing here change how you think about AI and meta?
我觉得整体上理念是相似的。
I mean, I think it's sort of a similar philosophy overall.
普莉西拉刚才提到，我们很多精力都放在开发工具上，帮助个人去实现目标。
And, you know, Priscilla was talking about this, that a lot of our focus is building tools that empower individuals to do things.
这其实是我所从事的很多工作的共同主题：把技术直接交给普通人使用。
And that's sort of a common theme across a lot of the things that that I work on is just kind of putting the technology in individuals' hands.
我们不认同那种高度集中的未来，即只有少数机构在推动所有进展。
We don't believe in this, like, very centralized future where there should be a small number of institutions that basically are advancing all of this stuff.
我们的愿景并不是会出现一个中央超级智能来解决所有科学问题。
Our vision is not that there's gonna be some central superintelligence that solves all of science.
我觉得人始终至关重要，未来会更加重要。给人们更多工具来提升效率，是任何积极未来的关键，这也是历史上进步一直的方式。
I think people are really important, and I think will be more important in the future, and giving people more tools to be more productive is gonna be a critical part of any kind of positive future that both and that's how progress has always been made historically.
对吧？
Right?
不是靠集中化。
It's not through centralization.
而是通过赋能个人去尝试那些主流之外的想法，别人觉得不行，可他们自己却认为这些想法值得去做。
It's through empowering individuals to try things that are somewhat out of the mainstream that other people didn't think were good ideas because they thought they were good ideas that already had been done.
所以我觉得这正是整个理念的核心——某种程度上，这也正是你为什么要创建社交媒体，给人们一个发声的机会。
So I think that that's very central to the whole ethos of I mean, to some degree, it's like why you create something like social media, right, to give people a voice.
我觉得，我特别关注的，是用个人化AI来赋能每个人。
It's, you know, I think a lot of the the stuff that we that I care about in terms of empowering people with individual AI.
开源是其中一种方式。
Open source is one instantiation of it.
但这不是唯一的方式。
It's not the only way to do it.
它至少代表了一种态度：我们要把这项技术交到每个人手里。
It certainly is one way that you basically are saying, we're going to take this technology and put it in everyone's hands.
在科学领域，我觉得这非常合理，我们也坚定地支持开源。
In terms of science, I think it really makes sense, and we're deeply committed to open source.
当然，这里面还有很多值得深思的问题，比如生物安全之类的，我们需要仔细权衡，找到妥善的应对方式。
There are obviously interesting considerations on this that are important too because there's a lot of considerations around biosafety and things like that that we're going to need to balance and think through how to handle.
但总的来说，这深深植根于我们在生物枢纽所做的工作的理念，也可能是我所有工作的共同主题——我们相信，一个美好的未来，是把技术当成工具，交到每个人手里，社会才能真正进步。
But I think overall, this is very deep in the ethos of the work that we're doing both at Biohub and probably a theme for a lot of the stuff that I do is just we believe that a positive future is one where you build a technology as a tool, you put it in individuals' hands, and that's kind of how society makes progress.
你在生物枢纽的使命，我觉得非常宏大。
You have this, I think, incredibly ambitious mission at Biohub.
但你知道，这里做人工智能的科学家们，其实也可以去商业公司工作。
And yet, you know, the AI scientists that work here could also go work in commercial enterprises.
你是怎么考虑人才问题的？怎么吸引人们来生物枢纽？
How do you think about the talent and, like, how to bring people to Biohub?
那咱们从哪儿说起呢？
I mean, where do want to start?
说实话，现在人工智能研究者的市场确实非常火热。
I think, you know, yeah, I mean, it's a very hot market for AI researchers.
但我觉得这意味着需求很大，他们非常抢手，可以自由选择自己想做的项目。
But I think that part of what that means is that there's a lot of demand, and they're very in demand and can work on the things that they wanna work on.
这又回到了我们之前说的关于前沿人工智能和前沿生物学的问题。
And I think this gets back to this point again about frontier AI and frontier biology.
所以，来这里的人工智能研究者完全可以去各大实验室做语言模型之类的工作，但那些地方都没有前沿生物学的板块。
So yeah, I think the AI researchers who work here could go work on language models or things at any of the main labs, But those labs don't have the frontier biology part attached to it.
因此，这里还有一个非常重要的使命因素——你能在这里做其他地方根本无法实现的独特研究。
So I think that there's also a just very large mission component of this, which is there's an ability to do this unique work here that you just can't really do at the other places.
如果你的关注点就在这里，那我真的觉得，世界上没有其他机构能同时做前沿生物学和前沿人工智能。
So if that's what your focus is, then I don't actually think that there's any other organization in the world that's doing both the frontier biology and the frontier AI.
对。
Yeah.
亚历克斯，你为什么来这里？
Why are you here, Alex?
我觉得原因很简单。
I mean, I think it's it's really simple.
是的。
Yeah.
我们的使命是预防疾病。
Our our mission is to take care of prevent disease.
我觉得，这真的就是
And and I think, you know, there's it's it's just such a
在不到一百年的时间里实现这一点，力量太强大了。
powerful with a straight face in a less than hundred year timeline.
认真的。
Serious now.
没有别的了。
There's no more.
没错。
That's yeah.
是的。
Yeah.
这真是一个非常有力量的使命。
It's it's a really powerful mission.
我觉得，说实话，确实如此。
And I I think, you know, you yeah.
我的意思是，科学家们，我认为，都
I mean, it's it's just, you know, scientists, I think, are
非常受科学精神的驱动。
very motivated by science.
没错。
Yes.
是的。
Yeah.
这是让人深深投入的事情。
It's it's something people are deeply motivated by.
我觉得，我们现在正处在一个关键时刻，这件事似乎真的有可能实现。
And I think, you know, we're at this moment in time where that actually seems like something that can be achieved.
而且我觉得，我们正在打造一个非常独特的地方，专门来解决这个问题。
And I think, you know, we're building a really unique place where where we're tackling that problem.
我们手头有资源，我觉得该有的东西都齐了，真的能去推动这件事。
And, you know, we have the resources, I think, kind of the right things to actually really go after that and do that.
对。
Yeah.
作为经常接触和招聘科研人员的人，我特别有共鸣。
I mean, that resonates with me as somebody who, you know, talks to and hires a lot of research scientists.
他们想知道你有没有数据、有没有工具、有没有算力、有没有人才，然后才是你的使命是什么。
They wanna know if you have the data, if you have the tools, if you have the compute, if you have the talent, then what the mission is.
所以我真的觉得，这一点特别有竞争力。
And so I actually think that's super competitive.
另一点是，你不需要一个特别大的团队。
The other thing is that you don't need a very large team.
所以我觉得，这个世界有意思的地方就在于，人们关心不同的使命，这其实是好事。
So I think it's like an interesting thing about the world is that people care about different missions, and that's good.
我觉得这正是关键所在——打造这些工具，让人能去探索自己真正关心的问题，不管是科学领域还是其他方面，之所以能推动社会进步，正是因为每个人关注的东西都不一样。
I think that's part of the whole I mean, part of why building these tools and giving people the ability to explore what they care about, whether it's across science or just across everything, is such a powerful way to make progress in society is that people care about different things.
要在人工智能领域取得进展，其实不需要成百上千名AI研究人员，甚至不需要那么多。
And in order to make progress in AI, you don't need many, many hundreds of AI researchers or thousands or anything like that.
我觉得，只要有一小群顶尖人才，十几二十来个人，就足以取得重大突破。
I think you can really make progress with a very strong group of a dozen or a couple dozen people.
而且，找到真正认同这个使命的人，并不是什么难事。
And, yeah, mean, finding people who care about this mission is not a particularly hard thing.
这可是世界上极其重要的事情之一。
I mean, this is like a super important thing in the world.
所以我觉得，这确实没错。
So I think that that's Yeah.
世界本来就是这样，人们自然会被不同的使命吸引。
It's just kind of a cool thing about the world is that people obviously are drawn to different missions.
所以我觉得，哪怕只是对这个领域稍有关注的人，脑子里最简单的认知模型大概就是：用于蛋白质的结构预测模型，以及蛋白质相互作用模型。
So I think the simplest mental models that folks have, even if they're paying attention to the space, are essentially like, okay, you know, structured prediction models for proteins and protein protein interaction models.
还有一块是基础性理解。
And then, so there's this one piece, is fundamental understanding.
然后还有一种设想，就是将来我们能直接零样本地把东西推到临床，而且成功率会高得多。
And then there's this like theory of someday we're just going to be able to like zero shot things into either the clinic or the clinic with much, much better hit rate.
从ESMFold2到那个目标，我们需要实现哪些突破？
What needs to happen for us to go from ESMFold2 to this other piece?
这真的可能实现吗？
Is that feasible?
我觉得这是个很好的问题。
I think that's a great question.
说实话，我对这一点非常乐观。
I mean, I would say that I'm really optimistic on that.
一方面，这些问题在过去可是能让人花上整个职业生涯去钻研的。
So think, you know, on the one hand, you know, these are problems that historically, you know, people could spend kind of an entire career working on.
比如，怎么才能有效优化一种药物？
Like, how do you how do you figure out how to effectively optimize a drug?
你怎么把它推进到临床前阶段？
How do you get it, you know, get it through preclinical?
早期安全性怎么评估？
How do you do the early safety?
我觉得，当一个新的科学范式出现时，那些曾经很难的问题，往往会变得简单起来。
I think that, you know, when you have a new scientific paradigm, kind of, you know, questions that were once hard kind of become simplified through the new paradigm.
所以我非常乐观，认为这些核心问题会通过这些模型以一种自然涌现的方式得到解决。
And so I'm very optimistic that kind of many of these core problems will be solved kind of in an emergent way through these models.
对。
Mhmm.
我觉得一个很好的例子就是毒性问题。
And I think one great example of that is is toxicity.
如果你能真正通过数字模拟，预测药物在人体内的分布和结合位点，
Whereas if if you can kind of really digitally digitally kind of simulate everything and be able to predict, you know, where a drug is going to distribute and bind across the human body.
那这个问题就已经有了初步的解决方向。
You kind of have the beginning of a solution to that kind of problem.
所以我认为，一旦在分子层面有了这些精准的模型，我们在很多核心问题上就会看到飞速的进展。
So I think that once you have these kind of accurate representations at the molecular level, we're going to start to see really rapid progress on a lot of these core problems.
过去一周发布后，你看到的最令人兴奋的模型应用或实验是什么？
What is the most exciting use or experimentation with the models you've seen in the last week since release?
是的。
Yeah.
能看到它被用在各种各样的地方，真的特别棒。
I mean, it's just been great to kind of see it get integrated in all kinds of things.
我们最近发现一个特别有意思的现象：有人把它和智能代理系统结合起来，实现自动化设计，把整个流程都自动化了。
I think one of the really interesting things that we've been seeing is people kind of connecting it with Agentic Systems to just kind of do automated design and kind of just automate that whole process.
这再次说明，把智能代理和前沿AI结合起来，再配上对生物学的全局模型，真正地理解并推理生物学，就能开始全面自动化整个设计流程。
So it's really, I think, another example of how you can kind of see bringing together Agentic and Frontier AI with, you know, the ability to have a world model for biology and actually reason about biology and, you know, really kind of start to automate the entire design process.
你们怎么决定下一步的研究方向？
How do you decide what the next step in the research agenda is?
比如生物学的全局模型，我这儿说得比较粗略。
It's like world model for biology, and then I could I'm just gonna be very coarse here.
比如我可以扩大规模，增加更多数据，或者加入新的方法和领域——加数据可不是件简单的事。
Like I could scale it up, I could add more data, I could add- like adding data is a nontrivial thing in terms of new methods and domains.
你们会关注整个生态系统的反馈吗？比如别人怎么用它，哪些地方还能改进？
Like what is- do you take input from the larger ecosystem about, you know, how people are using it and what would make it more useful?
还是说你们心里已经有明确的方向了，比如下一步该覆盖哪些结构或领域？
Or is it really like we we understand, like, the next step of structures or coverage that we're looking for?
我觉得有两方面要考虑。
I mean, I think there's two things.
我们对下一个重大挑战其实有个清晰的看法，那就是虚拟细胞。
So, like, we have a view on kind of the next big challenge, which I think is, you know, the the virtual cell.
对。
Mhmm.
关键是能把生物复杂性的各个层次，一层层往上推，直到整个细胞的层面。
And, you know, really being able to kind of ladder up the hierarchy of biological complexity to the cell.
抱歉。
Sorry.
这是个很基础的问题。
Very basic question.
对。
Yes.
虚拟细胞模型。
Virtual cell model.
那我该期待什么样的输入和输出呢？
Like, what is the input and output I should expect?
是的。
Yeah.
我觉得对此有不同的看法。
I mean, I think there's different views on that.
但归根结底，你希望的是一个能模拟各个复杂层级的系统。
But I think kind of what you ultimately want is a system that can really model each of the levels of complexity.
比如蛋白质组层面、基因层面、转录组层面，再把这些和表型联系起来。
So, you know, the the proteomic layer, the genetic layer, the transcriptomic layer, and connect that to the phenotype.
你需要足够的通用性，这样即使面对模型没训练过的新干预手段，也能向它提问并得到回应。
And you need enough generality so that you can ask the model questions about a new intervention in a context that it hasn't been trained on and kind of get an answer from it.
而我们这个领域需要弥补的差距，就是真正做出能泛化到新场景的预测。
And, you know, the gap that we need to close as a field is being able to really make those predictions that can generalize.
这就需要投入巨大的努力来生成数据。
So that's gonna require an enormous effort to generate data.
对。
Yeah.
至于接下来该怎么做，我觉得这其实是个很常见的资源权衡过程。
And then, I mean, in terms of what you decide to do next, I think this is like you know, a pretty normal process of constraint management.
对吧？
Right?
我觉得，全球每个领域的实验室，大概都面临算力不足的问题。
I mean, it's like like, I think every lab in every field across the world probably feels compute constrained.
这里应该也不例外。
I think that that's probably true here too.
对吧？
Right?
我的意思是，我知道总会有一些问题。
It's like so, I mean, I know, like, you know, there's always questions.
好吧。
It's like, okay.
我们是不是该更专注于推进蛋白质方面的研究？
Should we double down more on advancing the protein piece?
还是该多做点细胞层面的工作？
Should we do more of the cellular stuff?
我觉得这些在如何安排优先级上，一直都是大家在讨论的问题。
I think those are kind of ongoing debates in terms of how you sequence that.
而在这些选择里，我们其实也处在帕累托前沿上，得权衡每个模型该训练到什么程度，而且模型的规模也取决于你手头数据的量，毕竟如此。
And then, yeah, within that, there's kind of being at the Pareto frontier about how much you wanna train the different models in order to like, and and the size of the models is also dependent on the scale of the data that you have because you have yeah.
这都是显而易见的原因。
For for obvious reasons.
所以，我觉得这其实取决于你希望在这些曲线上处于什么位置，再加上一些现实限制，但我觉得这跟任何研究机构都会经历的过程是一样的——你总想往多个方向探索，但又得在有限资源下做权衡，集中精力把一件事做到世界级水平，同时埋下一些种子，让它们在未来两三年里慢慢开花结果。
So, yeah, I mean, I think it's there's some of that as just where you wanna be on the curves and then normal constraints, but I think that this is, like, probably the same process that, like, any research organization goes through of, like, you wanna go in all these different directions, and you're just trying to constraint optimize and make enough progress to do world class work at one thing at a time while planting some seeds that can blossom over the next couple years as well.
对。
Yeah.
这可能是我职业生涯中最具变革性的时候了。
This has been the most dynamic
至少在我所经历的科技领域里，这是最充满活力的时期。
period of technology, at least, I've seen my career.
AI领域现在发生的一切都让人无比兴奋。
It's so exciting in terms of everything that's happening with AI.
每周都有新的突破出现。
Every week there's something new that's changed.
你是觉得疲惫，还是充满干劲？
Are you tired or invigorated?
我两种感觉都有。
I'm both.
我觉得每个人都处于亢奋状态。
I feel like everybody's in manic phase.
没错。
Yes.
这是一种既振奋又疲惫的混合感受。
It's a combination of invigorated and exhausting.
确实如此。
Yeah.
这太棒了。
It's wonderful.
所以，眼下很多事情都很难预料。
And so I guess, know, things are very unpredictable right now.
真的很难说接下来会发生什么。
It's really hard to know what's coming.
我们在模型方面已经看到一些早期的指数级增长迹象，尤其是在智能体流程上，表现得特别有意思。
We have this almost like early signs of exponentation on the model side with agentic flows that we're starting to see in really interesting ways.
模型现在能帮的忙越来越多了，不过这方面的进展还是非常早期。
Models starting to help more and more with models, but that's still very very early days for that.
如果你回看五年后，想想当时怎么定义你这些努力的成功，我知道现在变化太快了，情况早就大不一样了。
If you're thinking back five years from now and you were to define what success was relative to your efforts, and I know things are very dynamic, things changed a lot.
但有一点始终没变，就是为生物枢纽打造工具体系。
But you have this common thread of tooling for the Biohub.
还有一个核心目标，就是大规模地赋能科学家。
You have a common thread of empowering scientists at scale.
再过五年回头看，你最希望确保自己实现的是什么？有没有一个特别想达成的主要目标？
You're looking back five years from now, is there a specific thing that you really want to make sure that you've accomplished or achieved or a primary goal?
我觉得我们对这套围绕生物学构建的多层次世界模型，已经有比较清晰的方向了。
Well, I mean, I think we have a pretty clear view of this hierarchical set of world models that we wanna build around biology.
另外，我们还希望做到全球顶尖的科研质量。
And the other part of that is that we wanna do the highest quality work in the world.
对吧？
Right?
我的意思是，我们现在已经具备了实现这一目标的条件，既有世界级的AI研究团队，又汇聚了生物枢纽和一批顶尖的生命科学研究机构。
I mean, I and I think we're basically set up to do that between having a world class AI research team and this collection of of Biohub's or world class life sciences research organizations.
我认为，这种组合在全世界范围内是独一无二的，没有任何其他机构能复制。
I think that that's, like, fundamentally a setup that no other organization in the world has.
但话说回来，就算有再多优秀的资源，也不等于一定能成功。
But, you know, you can have a lot of great ingredients, and that doesn't guarantee that you succeed.
所以，五年后回头看，我确信其他实验室或项目也会尝试做出类似我们正在做的事。
And so, I mean, to me, five years from now looking back, I think, you know, I'm sure other labs or efforts will try to produce things that approximate what we're trying to do.
但我相信，我们一定能做出真正更出色、对世界有独特智力贡献的成果。
And I just think that we should be able to do something that is meaningfully better and a unique intellectual contribution to the world.
其实，做任何研究，本质上都是为了追求这一点。
I think that that's kind of what you, whenever you do any kind of research, that's what you're trying to do.
对吧？
Right?
所以，如果我们真做到了，我相信大家都会觉得特别值得。
So, yeah, so if we do that, think we'll all feel very good.
我还认为，迟早会看到更多使用模型的人自己提出新想法。
I would also expect that at some point we'll just start seeing a lot more idea generation from the people using the models.
但我足够相信这部分会自然发生，所以对我来说，关键还是把工作做到顶尖水平。
But I have enough faith that that part will materialize that for me, it's more just about, like, making sure that we do world class work.
只要我们把核心部分做好了，其他事情差不多就会水到渠成。
And I think if we do, like, the rest almost will take care of itself.
最后一个问题想问你。
Very last question for you.
想象一下，现在是2026年年中。
Snapshot of it's mid-twenty twenty six.
过去一年里，你对Biohub或者这个领域最重大的想法转变是什么？
What's the biggest update in your own thinking about Biohub or the domain from the last year?
过去这一年的话。
Well, from the last year.
你去年才加入的。
I mean, you joined in the last year.
我的意思是，过去一年里，我们最重要的调整其实是正式确立了Biohub作为我们慈善事业的核心重点。
I mean, I think the biggest thing that that we basically rotated and and I think in the last year, we basically kind of formalized that Biohub is the main focus of our philanthropy.
所以这确实是一个很大的转变。
So I think this has been a very big shift.
但亚历克斯和团队加入后，我觉得特别有意思，不仅因为他们是一流的团队。
But Alex and the team coming in, I think, been interesting, not only because it's a world class group.
对吧？
Right?
你们其实已经一起共事很久了。
I mean, you guys have worked together for a while.
你也提到过，这个领域的变化实在太快了。
I think also, I mean, you talked about how stuff is changing so much in the field.
但有一件事常被忽略——这是一群极其优秀的人，彼此熟悉，配合默契，团队稳定，素质过硬。
I think one thing that's underrated is this is an extremely talented group of people who also know each other and work well together and are stable and good.
而且我觉得，人们长期在稳定环境中高效协作所带来的复利效应，这一点也常常被低估了。
And I think that that also is underestimated in terms of the compounding benefit of people being able to work well in a stable environment over time.
所以我觉得这一点非常关键。
So I think that that's a really important piece.
但在亚历克斯接手之前，生物枢纽的前任负责人主要是对技术感兴趣的生物学家。
But part of what we wanted to do was, prior to Alex leading the effort, the previous leaders of the Biohub were basically primarily biologists who were interested in technology.
现在我们真正实现了角色的反转——当然你也懂生物学，但你本质上是一位有生物学背景的AI研究者。
Now I think this is the point where we really flip that, where, I mean, obviously you have a background in biology as well, but you are primarily an AI researcher who has a background in biology.
这反映出我们对未来创造更大价值的期待。
I think that that's a deep reflection on the way that we expect that this is going to drive more value in the future.
所以过去一年里，我们在工作上的最大变化大概就是这些了。
So those are probably the biggest updates in the last year in terms of the work that we're doing.
不只是换了个新领导，而是一整个团队，我觉得都非常出色。
I mean, it's a new leader, not just the leader, but a team, that I think has been is really good.
然后，确实，我的意思是，
And then, yeah, I mean,
至于整个行业其他方面，进展还算顺利。
I think on the rest of the industry, it's like it's on track.
我的意思是，这事儿其实挺疯狂的，因为当你面对一条指数增长曲线时，那种感觉就是它涨得太快了，快到让人觉得根本不可能一直这样下去。
I mean, I think, like, every it's it's kind of this crazy thing because, like, when you have an exponentially growing curve, I think the way that an exponential curve feels is it's growing so quickly that the the kind of emotional feeling is it can't possibly keep going.
对啊。
Mhmm.
对吧？
Right?
但问题就在于，指数曲线的本质本来就不会一直线性往上走。
Because, like, it's because it's just like but but, I mean, the nature of an exponential curve is it it doesn't just keep going.
它是在不断加速。
It keeps accelerating.
对吧？
Right?
指数增长就是不断加速的过程。
Exponential growth is accelerating.
所以我觉得，这里面牵扯到很多情绪和心理层面的东西。
So I think that that has all of these emotions and psychology attached to it.
但我觉得，从整个行业的趋势来看，本质上它仍然走在正确的轨道上。
But I think fundamentally when you look at the curve in the industry, the kind of fundamental thing is it is on track.
它一直沿着这条曲线发展，这在我看来对各个领域都有深远的影响。
It has remained on that curve, which I think has all these very profound implications for all of these domains.
这无疑证实了我们的判断，也让人觉得在那些能带来长期回报的领域大举投入是值得的，而目前看来，我们确实还在这条轨道上。
But certainly, it validates and makes one feel very good about making a very big investment in in the things that that will play out if that if you stay on that track, and it seems like we are.
所以，我觉得这是个非常好的消息。
So that, I think, is very good news.
我觉得你做的最重要的事，是把实际的生物学数据和系统真正连起来了。
I think the most important aspect of what you're doing there is you're actually closing the loop with the actual biology.
对。
Mhmm.
因为代码和研究本身就依赖闭环系统。
Because with code and research, it's closed loop systems.
所以迭代起来特别快。
And so they're very fast to iterate.
这是一个开放循环的系统，而你们正在把它闭合起来。
This is an open loop system, so you're closing a loop.
这对推进进展至关重要。
And that's really crucial to progress.
是的。
Yeah.
对我而言，我们现在推行的策略最大的变化之一，就是在亚历克斯的带领下，我们不再只是各自团队大致朝着同一个方向努力、理解彼此潜在的合作与关联，而是真正地紧密协作。
For me, one of the biggest changes with the strategy we're driving now, and Alex at the helm is, you know, before we had amazing teams moving generally in the same direction and understanding, like, the potential collaborations and interconnectedness of our work.
现在我们手挽着手齐头并进，目标非常明确。
But now we are arms linked moving together, which It's is very very directed.
这让人非常兴奋。
And it's very exciting.
虽然也有点让人紧张，但真的像一支团队在相互配合，共同朝着目标前进。
It's a little bit scary, but it's like truly a team playing off each other and trying to make progress towards this goal.
这背后付出了很多努力，也得益于团队的工作已经成熟到足以无缝衔接的程度。
And that has taken a lot of work, but also the maturity, our teams being able to have their work at a level of maturation where it actually does make sense to interlock.
太棒了。
Amazing.
感谢各位团队的努力，你们做得很好。
Well, to teams being on the curve, thank you guys for doing this.
感谢你的参与。
Thank you for joining Yeah.
谢谢。
Thank you.
谢谢。
Thank you.
在Twitter上关注我们：NoPriersPod。
Find us on Twitter NoPriersPod.
如果你想看到我们的脸，记得订阅我们的YouTube频道。
Subscribe to our YouTube channel if you wanna see our faces.
在Apple Podcasts、Spotify或你常用的平台关注本节目。
Follow the show on Apple Podcasts, Spotify, or wherever you listen.
这样你每周都能收到一集新内容。
That way, you get a new episode every week.
还可以在 nopriors.com 订阅邮件或查看每集的文字稿。
And sign up for emails or find transcripts for every episode at nopriors.com.
关于 Bayt 播客
Bayt 提供中文+原文双语音频和字幕，帮助你打破语言障碍，轻松听懂全球优质播客。