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。
我在后台有点吓坏了，当时帕默说，屏幕上有些东西，说我不该说这个。
I was a little terrified backstage when Palmer's like, there's something on the screen that says I'm not supposed to say this.
我当时就想，老天爷，不管是什么，千万别是……
I was like, for the love of God, whatever it is, please not be...
千万别是关于《巨兽》的。
Please don't make it about Colossal.
拜托别说出来。
Please don't say it.
所以我还记得去年三月的Abundin 360大会上，就在你上台之前，埃隆也在场，他说，对，我想要一只毛茸茸的猛犸象当宠物。
So I remember at Abundin 360 last March, just before you came on, Elon was there, and he said, yeah, I want a pet woolly mammoth.
那个项目进展怎么样了？
How are you doing on that project?
我们经常收到这种请求。
We get that request.
但确实有一集，我觉得是《美国老爸》或者《恶搞之家》里的，他们解释过这个，你看了吗？
But there was actually an amazing, I think it was either American Dad or Family Guy episode, where they explained, did you see this?
是的，我不知道你们有没有看到这个。
Yes, I don't know I you guys
你们看到这个了吗？
saw this.
这个到处都在播。
It's all over the air.
它解释了CRISPR，真的很棒。
Explained CRISPR, and it was amazing.
真的很棒，因为最后他们还说，威廉·阿尼玛斯出来了，大概有狗那么大。
It was amazing, because then at the end, they were like, yeah, and the William Animas came out and they were about the size of a dog.
实际上它们就是那么大。
That's actually how big they were.
化石记录是错的，对吧？
The fossils were just wrong, right?
所以，这真的很棒，因为它解释了CRISPR，但我就想，你知道吗，我们收到的最多的请求之一就是茶杯猛犸象。
So, it was like amazing because it explains CRISPR, but I was like, you know, we get the request, like the number two request we get is teacup mammoths.
茶杯？
Teacup?
茶杯猛犸象。
Teacup mammoth.
就像，就像，
Like, like,
对，大家都想要一个茶杯大小的。
Yeah, everybody wants a teacup.
他们想把猛犸象放进你的包里，对吧？
Like they wanna take mammoth and put it in your purse, right?
所以，当我们能改造出一种新的黑色素，让它变成粉色，然后巴黎可能就想要了。
And so, when, I mean we can engineer a new melanin, make it pink, and then Paris may want to.
所以，我们目前还没在做这个。
So, we're not working on that currently.
但我们在猛犸象项目上进展不错，所以。
But we're making good progress on the Mammoth project, so.
太厉害了。
Amazing.
那，我就直接开始问了。
So, let me kick this off.
在创立Colossal之前，你已经成功创办了四家公司，涉及游戏、国防、人工智能和移动领域。
So, you've started like four successful companies before Colossal, gaming, in defense, in AI, in mobile.
而你对合成生物学一无所知。
And so, you knew nothing about synthetic biology.
现在这家公司市值超过一百亿美元。
Before you started, now a company worth over $10,000,000,000.
那你为什么把“灭绝”作为你的登月计划呢？
So, why made you make the extinction your moonshot?
而且没有合成生物学的背景，这对你来说是障碍还是优势？
And having no background in synthetic biology, was that a hindrance or an advantage?
哦，我觉得这是巨大的优势。
Oh, think it's a massive advantage.
我觉得这让我有机会走进那些房间，你知道，帕尔默也提到过一点，就是招人来取代你，招比你聪明得多的人，对吧？
I think that it affords me the opportunity to go into rooms of, you know Palmer talked a little bit about this too, of like hiring people to replace you and hiring people that are much smarter than you, right?
所以，我就能……
And so, I get...
从我的角度看，好处在于，我还是要处理很多同样的破事，但同时，我可以坐下来跟贝丝·夏皮罗学习古DNA提取，还有怎么避免犯错。
What's great about this from my vantage point is like I get to deal with a lot of the same bullshit, but at the same time, I can go in and like sit down with like Beth Shapiro and learn about ancient DNA extraction and like how not to do things.
但下一个会议，我又能去了解我们在多重编辑方面推进的边界，以及我们一次能做多少次编辑，对吧？
But then the next meeting I can go learn about where we're pushing the boundaries of multiplex editing and how many edits that we can make at once, right?
所以，我想说，我知道怎么——而且我觉得在我的整个职业生涯里，我一直知道怎么问对问题，因为我真的很好奇。
And so, I would say that I knew how to, and I guess in all of my career I've known how to ask the right questions, because I'm really curious.
但我确实很认同那句老话，就是把最聪明的男女放在你身边，然后就是问他们问题，对吧？
But I really do kind of subscribe to that old adage of putting the top smartest women and men around you, just asking them questions, right?
所以，我开会就是去问问题，90%的时候，我觉得人们会想，好吧，如果你更懂生物学，这个会能开得快一点。
And so, I get to go to meetings to just ask questions, and 90% of the time I think people are like, okay, if you knew more about biology, this meeting would go faster.
但剩下10%的时候，他们会说，我们从来没这么想过，对吧？
But then 10% of the time, they're like, we never thought of it that way, right?
顺便
By
说一下，这对我们所有创业者来说都是一个非常重要的教训，对吧？
the way, this is such an important lesson for all of us as entrepreneurs, right?
仅仅因为你在某个领域不是专家，并不意味着你就不应该进入那个领域，只要你热爱它，并且能吸引最优秀的人才围绕在你身边。
Just because you're not an expert in an area, doesn't make it an area that you shouldn't go into if you love, if you attract the best talent around you.
乔治·丘奇，这是我唯一会炫耀的事情，我为此感到非常自豪。
George Church, and like this is the only brag that I will say, which I'm very proud of.
乔治·丘奇说我是他教过的最好的学生，因为我真的会不停地问他问题，比如，我一年中最喜欢的时光就是假期，因为那时候没人工作，我可以和乔治·丘奇打上好几个小时的电话，就聊合成生物学的各种可能性。
George Church says that I am the best student he's ever had, because I will literally just pepper him and be like, hey, my favorite times of the year are during holiday seasons when like no one's working, because I will just get on calls for hours and hours and hours with George Church, and just talk about the possibilities with synthetic biology.
乔治·丘奇是谁，Fervent？
Who's George Church, Fervent?
哦，如果你不认识乔治，乔治可以说是合成生物学之父。
Oh, if you don't know George, George is the father of, arguably the father of synthetic biology.
他是哈佛大学遗传学系的主任，很多下一代读写技术都是出自丘奇实验室。
He's the head of genetics at Harvard, and a lot of the next gen readwrite technologies that were invented came out of the church lab.
他的实验室成果丰硕。
And his lab's prolific.
从那个实验室已经孵化出了很多市值数十亿美元的公司。
There's been numerous multi billion dollar companies that have spun out of the lab.
这可能是最活跃的创业生物学实验室，而且他身高六英尺七英寸，患有嗜睡症，还特别风趣。
It's probably the most active startup biology lab in And the he's also six'seven with narcolepsy and hilarious.
而且他人特别好。
And such a sweetheart guy.
他是最乐于合作的人，对吧？
And he is the most collaborative person ever, right?
虽然他并非软件出身，但他非常热衷于开源，以及技术的民主化，比如基因组测序。
And so he's also hardcore, even though he doesn't come from software, he's hardcore into open source and just like the democratization of technologies, like genome sequencing.
他希望把成本从数十亿降到一百美元，他在这方面非常积极。
He wanted that to go from billions to a $100, which he was very active in that category.
所以，他真的是我合作过的最有协作精神的联合创始人。
And so, he is literally just the most collaborative co founder I've ever had the pleasure of working with.
戴夫？
Dave?
对，我对生物学的商业模式真的很好奇。
Yeah, really curious about the business model of biology.
我觉得大家都知道，在AI即将实现的所有应用场景中，解决所有疾病、治愈所有痛苦是优先级最高的。
I think we all know that of all the use cases of AI that are imminent, solving all disease, curing all pain is just highest on the priority list.
生物学的商业模式总是……
The business model in biology is just always...
而且作为一位从外部进入生物领域的连续创业者，听听你的观点真是太好了。
And it's great to get your perspective as a serial entrepreneur who came into biology from the outside.
所以我的第一个关于商业模式的问题是基础模型。
So my first question on like the business model is the foundation models.
就像我们刚才看到的，Astra开箱就能开车。
Like we just saw that out of the box Astra can drive a car.
它会不会开箱就能做生物学，还是你会说，不不不，我们需要自己构建？
Is it gonna do biology out of the box, or do you say no, no, no, we need to build our own?
还是得这样，所以现在的基座模型，我们很幸运能和OpenAI还有Anthropic合作，提前接触到一些这样的东西，这真的很棒。
Still have to, so the foundation model's now, and we've been very fortunate to work with both Open AI and Anthropic on getting early access to some of that stuff, which has been great.
很长一段时间里，我们都是从软件的角度，把它当作一种智能中间件来用，对吧。
For a long time, we were leveraging it for kind of like thoughtful middleware, right, coming from a software perspective.
我们把实验室笔记本和Jira连接起来。
Where we were connecting lab notebooks and Jira.
我觉得我们在公司做过最难的事情之一，就是让科学家们学会用Jira。
I mean, of the hardest things that we did at the company was retrain scientists to work in Jira.
有时候这比重新编程还难。
So, that's harder than some reprogramming at times.
所以，实际上很长一段时间，你知道，这些在前沿模型之前的LLM，真的很擅长写学期论文，对吧？
And so, actually, for a long time, you know, these, the LLMs before the frontier models, were really good at writing term papers, right?
它们在做祖先状态重建或者比较基因组学方面，其实不太行，对吧？
They weren't really great at doing like ancestral state reconstruction or comparative genomics, right?
就像那些，它们就是搞不定，要是能搞定那就太棒了。
Like that, that's where they just didn't work, which would have been amazing.
所以很长一段时间，我们都在用它们做基础设施的中间层，这样你就不必去请德勤或埃森哲那类公司来搭建所有这些系统和报告了，对吧？
So for a long time, we were leveraging them for kind of like middleware layers of infrastructure, so you don't have to go hire the Deloitte's or Accenture's of the world to build all these systems and reporting, right?
所以我们做了个……
So we did a...
我觉得我们在这方面一直做得非常成功，持续了很长时间。
I think we were very successful at deploying that for a long, long time.
现在它们已经进步很多了，可以运行模拟实验了。
Now they're getting pretty far, where you can run simulation experiments.
现在有很多公司冒出来，比如Lila和其他一些，都在尝试围绕这个做实验室自动化。
There's a bunch of companies out there that are standing up, Lila and others, that are trying to do lab automation around it.
我们在投资这个领域时，一直很谨慎地考虑怎么投、什么时候投。
Been pretty thoughtful about how and when we invest in that category.
但我觉得你会发现，真正有趣的洞察往往出现在那些本质上还是语言问题的连接之间。
But I think that what you're gonna find is you're gonna find interesting insights between connections that are really still language problems.
举个例子，如果你想为人类建一家新的治疗公司，去搜一下所有关于老鼠的文献，你会发现很多已发表的论文里，同一个基因会有不同的叫法。
So, example, if you go search the literature on everything on mice, if you just wanna go build a new therapy company or whatever for humans, and you go look at all of the published papers on mice, many of them, they'll call a gene a different thing.
他们会用不同的方式分类。
They'll classify it differently.
他们会稍微改变实验方法。
They'll run that experiment slightly differently.
所以，如果你尝试完全重复同样的实验，你会发现已发表的研究中有百分之四十到六十的失败率，这太糟糕了。
And so, if you go try to run the exact same experiment, you're gonna get somewhere between forty and sixty percent failure rate on published work, which is terrible.
所以，我认为在下一波浪潮中，除了小分子药物发现之外，AI真正能帮上忙的地方，就是消除这种差距，弥合不同研究之间的术语差异。
So where I think AI is gonna be really helpful in the next wave outside of small molecule drug discovery, is I think it'll be very, very helpful eliminating that gap and bridging that nomenclature between work.
有时候我们会做研究，后来才发现有一篇同行评审的论文已经解决了我们试图解决的问题，但我们当时查找的方式在PubMember上根本找不到。
There'll be times where we'll do work, and we'll find out later that there was a peer reviewed published paper one of the things that we were trying to solve, but the way we were looking for it wasn't findable on PubMember '1.
所以我认为短期内，AI会弥合沟通层面的差距，但我确实认为你将会……
So I think in the short term, it's gonna bridge the communication layer, but I do think you're gonna...
你看到了，我想Anthropic几天前宣布了这个，对吧，他们……
And you saw this, I think Anthropic announced this a couple days ago, right, where they've...
而且现在有Ginkgo等公司准备进行测试。
And you have Ginkgo and others that are now going to test it.
你仍然需要湿实验室实验来测试和验证这些。
You're still gonna need to have a wet lab experiment to test and validate those.
但是，在大量不同的实验中运行模拟设计，并帮助创造性地提出下一个实验，我认为这就是未来三到五年AI和生物学结合的方向。
But running simulation design across a myriad of different experiments and helping creatively come up with the next experiment, I think that's where we're gonna see AI and biology for the next three to five years.
但这样想，如果我看文本数据和研究结果，翻阅成千上万的先前测试，那完全是LLM的领域。
But what about this, if I think about text data and research results and going through thousands and thousands of prior tests, that's all LLM city.
但如果我的输入向量只是一个基因序列呢？
But what if my input vector is just a gene sequence?
比如我现在把它扔给Anthropic，我敢打赌不会有什么好结果。
Like if I dump that right now into Anthropic, pretty sure nothing good is gonna come out the
对，但在规模上。
other Yeah, but at scale.
所以，谢谢你帮我引出这个话题。
So, thank you for the tee up.
所以，这正是我坚信我们正在推出的全球可行系统的重要性所在，对吧？
So that is what I fundamentally believe is the importance of our global viable system that we're rolling out, right?
所以，我们正在努力推出这个类似“诺亚方舟2.0”的模型。
So, we're trying to roll out this like Noah's Ark 2.0 model.
不是说诺亚在第一阶段做得不好，但至少我们的模型有点不同。
Not that Noah didn't get it really right in the first stage, but at least ours is slightly different.
我们做的是ATAC测序以及其他他没做过的事情。
We're doing ATAC sequencing and other things that he didn't do.
所以，我们正在与世界各国政府合作，建立本地化的生物库，收集所有这些数据，以便进行T到T测序。
And so, we're going out and trying to work with governments around the world to stand up localized bio vaults to get all of that so that we can do T to T sequencing.
所以，针对你的观点，我不认为单一基因组就能，比如，你把它输入到神话中，它就会说，哦，如果你做这六个改变，它就能永生。
And so, to your point, I do not think that a single genome is going to, like, you're not gonna feed it into mythos, and it's gonna be like, oh, well if you make these six changes, it's immortal.
但我认为，如果你去看，比如，你知道，鸟类物种中所有进化路径上的数千个基因组，发现它们对许多疾病不易感，你会看到这一点，对吧？
But I think that if you go look at, like, you know, you know, thousands of genomes across all these evolutionary lanes in avian species, and see that they are not susceptible to many of these diseases, you'll look at that, right?
我们正在小规模地做这件事，比如研究p53、永生水母等，观察物种的已知结果，然后尝试沿着生命之树回溯。
We're doing that on a small scale with things like p53, immortal jellyfishes, other things, looking at kind of like known outcomes of the species, and then try to backtrack it down the tree of life.
所以，我们目前正在我们的一家公司里做这些工作，对此我们非常兴奋，早期迹象是积极的。
So we're doing some of that right now at one of our companies, which we're pretty excited about, and early indications are positive.
但你说得对，你不能只是把基因组扔进去，然后它就会告诉你该怎么修复，让它变得完美。
But to your point, you're not just gonna be all throw a genome in, it's gonna be like, here's how you fix it, and it makes it perfect.
但我确实认为，从比较基因组学的角度来看，规模化的数据量最终会让你达到那个目标。
But I do think that that amount of data, from a comparative genomics perspective, at scale, will get you there.
但我也认为，我会争辩说，数据集比模型更有价值。
But I also think that, so I would make the argument that the data set is more valuable than the model.
因为我有数据集。
Because I have the data set.
我们已经多次强调过这一点了。
We've made that point so many times.
亚历克斯。
Alex.
是的，我很想谈谈关于“去灭绝”的话题。
Yeah, I'd love to talk a bit about de extinction.
我知道你现在已经有四个以上的衍生公司了，而且……
I'm cognizant that you now have four plus spin offs, and...
不过话说回来，我觉得灭绝物种复活这件事，还是你和Colossal最出名的成就。
But nonetheless, de extinction I think is still what you and Colossal are perhaps best
我们最出名的成就……
known for....
那些我们不该提的事。
Things we're not supposed to talk about.
不过那些我们就不聊了。
We won't talk about those though.
所以……
So
说到灭绝物种复活，有一位俄罗斯哲学家，尼古拉·费奥多罗夫，十九世纪末，他是俄罗斯宇宙主义这一哲学流派的先驱。他认为，人类发展科学技术的最终目标，本质上就是开发出技术，让每一个曾经活过的人都能复活。
de extinction, there was a a Russian philosopher, Nikolai Fyodorov, late nineteenth century parent of a strain of philosophy called Russian Cosmism that argued that the ultimate trajectory of humanity in developing science and technology would be essentially to develop the technology to revive every human who's ever lived.
俄罗斯宇宙主义，你可以去查查。
Russian Cosmism, you can look it up.
所以从某种意义上说，你们做的灭绝物种复活，就是……
So de extinction, in some sense you are
我们现在还没在做这个。
We are not doing that currently.
那就好。
Good know.
所以……
So in...
但从某种意义上说，通过复活各种模式物种，包括猛犸象，你是我所知道的世界上第一家至少有一个可行的商业模式或至少有一条技术路径去尝试覆盖整个历史生物圈的公司。
But in some sense, by de extinguishing, de extincting various model species, including the woolly mammoth, you're the first company to my knowledge on earth that at least has a plausible business model or at least technical trajectory to try to go after the entire historical biosphere.
所以也许不仅仅是每一个曾经活过的人，而是每一个曾经活过的非人类生物。
So maybe not even just every human who's ever lived, but every non human organism that's ever lived.
在超级智能时代，重大挑战接踵而至，你认为人类共同的任务，就像十九世纪俄国宇宙主义者所认为的那样，最终会拥有技术来复活每一个曾经活过的生物，或者至少是每一个人，你觉得这有可能实现吗？
In an era of super intelligence where grand challenges are falling left and right, do you think that humanity's common task, as the nineteenth century Russian cosmos thought that will ultimately have the technology to revive every organism that's ever lived, or at least every human, do you foresee that becoming possible?
我不这么认为。
I do not.
我认为通过DNA合成、预测模型，以及，你知道，某种程度……
I think through DNA synthesis, prediction models, and, you know, some level...
通过合成生物学的某些组成部分，我们能够做到非常接近的近似。
Some components of synthetic biology, we will be able to get pretty close approximates.
不过我不认为……
I don't think though...
但有两个重要的前提。
With two big caveats.
第一，顺便说一下，Colossal 在人类方面不使用我们任何技术。
Number one, just a second for those other, Colossal does not use any of our technologies at Colossal for humans.
所以，我们不会那样做，尽管我认为我们的……
So, we won't do that, even though I think our...
那你们需要为此成立一个衍生公司吗？
You'll need a spin off for that?
你是说……
Is I that what you're
我认为当你开始涉及人类时，你需要不同的投资者，有时甚至需要完全不同的治理结构，对吧？
think that when you start working with humans, you need different, sometimes different investors, sometimes indefinitely different governance, right?
而且你会经历不同的流程——
And you go through different processes-
还有更多的病人。
And more patients.
对，还有更多的病人，对吧？
And more patients, right?
这其实是鲍勃·纳尔逊给我的一个非常好的建议。
That was actually really good advice I got from Bob Nelson.
鲍勃·纳尔逊早期就跟我说，嘿，暂时别把这些技术用在人身上，因为一旦跟FDA打交道，你就等于进入了代码冻结状态，然后就会陷入那种无休止的单调循环。
Bob Nelson in the early days was like, hey, don't apply any of this to humans for a while, because you effectively go into code freeze with the FDA, and it's just a, you're just gonna be in this monotony forever.
先把技术做到成熟稳定，然后再拿出来用。
Go get to the technologies until they plateau before you then take them out.
但回到你的问题，我觉得从根本上说，大多数生物……
But back to your question, I think that fundamentally, most organisms...
记住，这些是我们已知的大多数生物。
Remember, these are the most organisms that we know.
很少有人意识到这一点，但历史上发现的霸王龙不到100只。
Very few people realize this, but there's less than 100 T.
曾经存在过的霸王龙……
Rexes that have ever been...
我就想顺着这个话题说下去，可能会吓到你。
I just want to go down the path, it'll scare you.
被发现过的霸王龙不到100只，对吧？
There's less than like 100 T rexes that have ever been found, right?
但据说在地球上不同时期，霸王龙有数十亿只，对吧？
But with that, there's billions that allegedly were on Earth at different points in time, right?
所以，能留下化石记录的生物非常少。
And so, very few things leave a fossil record.
所以，我觉得我们更有可能从可编程生命的角度，按照我们的意愿去设计生命，而不是把曾经存在过的一切都带回来，因为我觉得我们甚至都不完全了解，对吧？
So, I think it's more likely that we will be able to engineer life from a programmable life perspective in the way that we want it, than just bring back everything that ever existed, because I don't even think we know, right?
我觉得在某个时间点，基于超级人工智能，我们很可能会达到这样的程度，我们会说，哦，我们要设计这个东西。
I think it's probably highly likely based on AI superintelligence at some point, where we get to the point where we're saying, oh, we're gonna engineer this thing.
回到你的问题，也许它确实存在过，但我们永远不会知道，因为它没有留下化石记录，对吧？
And to your question, maybe it did exist, but we just never know because it left no fossil records, right?
所以，我认为更有可能的是，我们会为了自己的利益而改造生命，试图让一些东西复活。
So, I think it's more likely that we will be engineering life to our advantage in trying to bring back things.
即使Colossal或者我们的一些技术，无论是克隆还是基因组工程，能让一些东西复活，你也知道，环境因素、表观遗传学，还有其他所有这些因素。
And even if Colossal or a subset of our technologies, both cloning and genome engineering, can bring back things, as you know, environmental factors, epigenetics, all these other things.
如果我们能克隆出莫扎特，那也不意味着他会出来说，哦，我要解决音乐糟糕的发展方向。
If we could clone Mozart, that doesn't mean that he's gonna come out and be like, oh, I'm gonna solve where the terrible trajectory of music has gone.
嗯，也许我有个后续问题。
Well, just maybe a follow-up question.
所以，具体到莫扎特，我们确实对他的生平了解很多。
So, for Mozart specifically, we do know quite a bit about Mozart's life.
所以，如果我们真的想复活莫扎特，我们完全没问题重建他的童年环境。
So, our arduando, if we did want to resurrect Mozart, we'd have no problem at all with reconstructing his childhood environment.
在一定程度上，对吧？
To a point, right?
所以，我并不是在……
And so, I'm not in...
顺便说一句，我要说清楚，我绝对不是在鼓励这种提问方向，但我们确实是在培育人类。
And by the way, just to be very clear, I am not encouraging in any way this line of questioning, but we are taking humans and growing them.
我还以为你要说迈克尔·乔丹和勒布朗·詹姆斯的超级篮球队，吓我一跳。
Was like, you're gonna go down this Michael Jordan LeBron super basketball team for a second and scare me.
所以，我确实认为，如果我们能做到——或者说，不是可能，而是百分之百准确地说，我们了解他们的基因倾向和在这些特质上的天赋，并且在合适的环境下，如果你想对他们进行模拟设计，把他们放在合适的环境里，那么他们可能……
So, I do think it's highly likely that if we were able, or it's not highly, it's 100% accurate to say that we understand their genetic disposition and aptitude towards these certain traits, and under the right environments, if you wanna go like, you know, if if you if you wanna go all simulation design on them and put them in the right environments, then they could probably...
那么你的莫扎特2.0可能会比原版莫扎特更出色。
Then then your Mozart 2.o could probably be better than the Mozart
但是，Ben，让我们回到动物王国，因为现在的核心论点是，AI系统可以设计出反映表型的基因组序列。
But one Ben, let's bring it back to the animal kingdom because right now the whole thesis is that the AI systems can design the genome sequence that reflect a phenotype.
所以，如果你想要一个动物，它正在变得更大……
So if you want an animal that's It's getting that, bigger
虽然还没完全达到，但……
it's a 100% there yet, but
但这就是目标，对吧？
But it's, that's the objective, right?
如果你想要一个鼻子更长的动物，或者有翅膀的动物。
If you want an animal that's got a longer snout, or an animal that has wings.
所以，就像我在FII舞台上问你的，你知道，你能做出一个皮卡丘吗？
So, like, I asked you on stage at FII, you know, could you make a Pikachu?
那似乎是个奇怪的热门话题。
That that seems to be a weird, like, fan favorite.
你知道吗？
Know?
我觉得人们更容易接受皮卡丘，而不是你想象的那些大型基因人类群体。
People are more accepting, I think, of Pikachu than these large genetic human camps that you're thinking Branded species.
是的，没错，非常对。
Yeah, yes, Very, yes.
对，我觉得人们更接受那种基因组工程模式，而不是……
Yeah, think people are more open to that model of genome engineering than
但听起来，也许总结一下，这似乎不是技术上的反对，更多的是对社会、政治、监管方面的担忧。
But it sounds, just maybe to tie a bow on this, sounds like isn't a technical objection per se, it's more worries of social, political, regulatory concerns.
对，对，是的。
Correct, correct, yes.
还有伦理方面的。
In ethical.
这就是我想插话的地方。
So this is where I'd like to jump in.
你知道，斯图尔特·布兰德说过一句著名的话，彼得，你写了一本著名的书，他说，我们是神，我们不妨开始像神一样行事。
You know, in Stuart Brand made this famous comment, Peter you wrote a famous title of your book, he said, we are as gods, we might as well start acting like that.
他是在1968年说的。
He said that in 1968.
我们现在处于一个可以灭绝或复活任何物种的时代。
We are at a point where you can de extinct or bring back any species.
你如何看待伦理问题，我们应该复活什么？
How do you think through the ethics of what should we bring back?
哪些该复活，哪些不该复活？
Which ones, which ones shouldn't we bring back?
所以，我是
So, I'm a
斯图尔特的超级粉丝。
huge Stewart fan.
我很喜欢他，也喜欢瑞安。
I love him, love Ryan.
那句名言流传很广，就像那部著名的恐龙电影里的台词一样。
That quote has been out there, kind of like, know, many from a famous dinosaur movie.
但我不确定我会像斯图尔特那样，直接坐上飞机去行动。
But I don't know if I would characterize it like Stewart would then get on a plane.
但在我看来，我想说的是，我们花了很多时间，但从合成生物学的角度来看，目前还不可能复活所有物种或改造一切。
But from my perspective, I would say that we spend a lot of time, It is not plausible yet to bring back everything or engineer everything from a synthetic biology perspective.
所以我们尽量深思熟虑，人们会问我们，有没有一个清单？
So, we try to be very thoughtful, and people ask us, is there a checklist?
你们到底是怎么选择物种的？
How exactly do you go about selecting your species?
因为有些物种我们已经公开了，还有一些我们还没有公开。
Because there's a species that we've been very public about, then and there's species that we have not Hold yet
对。
on.
我不是在说你们具体在做什么。
I'm not talking about what you're doing specifically.
哦，你说的是哲学层面上的。
Oh, you're saying philosophically.
从哲学层面来说，我们可以复活任何物种，那谁来决定呢？
Generally philosophically, we could bring back any species, who gets to decide?
你们用什么依据来决定，哦，复活这个或者那个？
What evidence do you use to say, oh, bring this back or that?
是的。
Yep.
从社会伦理的角度来看，我们应该怎么考虑？
Just from a societal ethics perspective, how should
我们该怎么想，对吧？
we be thinking Right?
所以，这是个很好的问题。
And so, it's a great question.
因为我们在这个领域算是先行者，我们的思考方式是，它们对环境有什么贡献？
And because we're kind of the guinea pig in this world, way that we think about it is, what was their contribution to the environment?
它们对食物链有什么贡献？
What was their contribution to the food web?
它们当时……
What was their...
它们为什么会灭绝？
Why did they go extinct?
还有……
What...
原住民对此有什么看法？
How do indigenous people feel about it?
比如我们研究的一些物种，对原住民来说有很深的精神联系。
Like, some of the species that we work on, they have a deep spiritual connection to the indigenous people side of it.
所以我觉得你不可能直接跟他们引用斯图尔特·布兰德的话。
So I don't think you're gonna walk in and quote Stuart Brand to them.
所以我认为把这些因素都考虑进去并权衡轻重是非常重要的。
So I think that it's very important to factor all those and weight those.
我们可能会犯错，社会也可能会犯错，但我们会继续尝试。
And we're probably gonna get it wrong, and I think society's gonna get it wrong, but we're gonna continue to try.
但我们也会考虑，这有没有教育意义？
But then we also look at, like, is there an educational benefit to it?
比如我们研究恐狼的时候，在跟原住民群体和红狼联盟团队合作之后，我们也考虑了流行文化的影响。
Like, one of the things that we did when we did the Dire Wolves, right, after working with the indigenous people groups and the Red Wolf Coalition teams and all these different components is we did weigh in the pop culture nature to it.
我们想，有没有办法把那些关注科幻、权游和《The Gathering》的人吸引过来，让他们参与到狼的保护中呢？
We thought, is there a way that we can bring all these people that focus on sci fi and Game of Thrones and The Gathering, can we bring them to wolf conservation?
那我们能不能把他们吸引过来……
And can we bring them...
我们能不能教他们基因组工程，因为我们带回了一种他们原本以为只存在于幻想世界里的神话生物？
Can we teach them about genome engineering because we brought something back that they thought was only a mythological creature in their fantasy universe?
所以，我们会尽量权衡所有这些不同的因素。
So, we try to weight all of these things differently.
我们没有一个完美的内部算法来处理这个问题，但我确实认为，随着技术的普及和更多政府采用我们的技术，他们必须权衡这些因素。是的，每个个体……
There's not a perfect kind of internal algorithm for how we look at it, but I do think that as the technologies proliferate and more governments are deploying our technologies, they will have to wait that on an Yes, individual
他们会的。
they will.
那么，本……
So Ben
而且我相信我们所有人都会在某个时候犯错。
And I'm sure that all of us are gonna get it wrong at some point.
嗯，因为涉及的因素太多了，对吧？
Well, because there's so many different factors involved, right?
比如我是个土著部落的人，我崇拜袋獾。
Like I'm some indigenous tribe and I worship the Tasmanian devil.
但这并不意味着你应该或不应该把它带回来。
That doesn't mean you should or shouldn't bring it back.
还有很多其他因素需要考虑。
There's all sorts of other factors.
我们得非常仔细地思考这个问题……
We're gonna have to think through that at a very...
因为你带来的这种力量，是人类历史上从未见过的。
Because the power that you're bringing to the table here is something that we've never seen in the history of humanity.
确实，你就像一个行走的奇点。
Mean, is quite You're like a walking singularity.
是的。
Are.
我们确实……
We do...
我觉得社会并没有完全理解这一点，我的意思是，也提到了帕克，虽然他们都听过伊恩·马尔科姆谈论过，但我确实认为他说得非常对。
I think that society doesn't fully understand, I mean, also address Park, and I think that they've all heard Ian Malcolm talk about it, but I do think he was really right.
比如说，我认为正在发展的技术，结合了合成生物学、计算、人工智能，最终还有量子技术，会比历史上任何武器系统都更强大。
Like, do think that the technologies that are being developed with the kind of intersection of synthetic biology, compute, AI, and then eventually quantum will be more powerful than any weapon system that's ever been created.
绝对是的。
By far.
本，我们谈论的是解决一切问题。
Ben, we talk about solving everything.
我们谈论的是，你知道，数学已经搞定了，接下来是物理，然后是化学和生物学。
We talk about, you know, math is cooked, physics is next, chemistry and biology.
是否存在一个拐点，一个生物学上的奇点，突然之间，所有已知的DNA序列、所有已知的表型都形成了一个完整的知识库，你可以直接设计并操作。
Is there an inflection point, a singularity in biology where all of a sudden there is a complete knowledge base of all the ingested DNA, all the ingested phenotypes and you can literally design and work
就像用CAD软件来做生物学一样。
with You like take CAD software for biology.
你可以设计出那样的东西。
You can design that.
你可以输入指令，
You can prompt,
给我一个能做这个的动物。
give me an animal that does this.
对，我觉得DNA合成还没完全到位。
Yeah, I think that DNA synthesis isn't quite there yet.
不过项目
Multiple But project
对我来说。
for me here.
对，对，所以我觉得那个世界不到十年就会到来。
Yeah, yeah, So I think that world is less than 10 out.
好的。
Okay.
行。
Alright.
所以，你可以用10来提示你最喜欢的皮卡丘或动物，因为它是10。
So, you can prompt your favorite Pikachu or animal in 10 because it's 10
谁拥有那些技术，就应该为你做这件事，但我确实认为，能够在不同分支的基础生物体上设计关键表型，并能够合成它们或进行多重工程改造，然后在体外培育它们，这些技术将在十年内实现。
Whoever owns those technologies should do it for you, but I do think that the technologies of being able to engineer key phenotypes on base level organisms in different clades and be able to synthesize them or multiplex engineer them and then grow them ex utero is within a decade.
好的，体外培育。
Okay ex utero.
十年之内。
Within the decade.
说说你的，那是什么？
Talk about your, what's that?
十年之内。
Within a decade.
十年之内。
Within a decade.
我的天哪。
Holy shit.
你能做……
Can you make...
这太不可思议了。
This is unbelievable.
好吧，让我给你看一条当前的曲线，但这显然不意味着它会一直沿着这条曲线走下去。
Well, let me let me give you a current curve and this doesn't obviously mean that it's gonna continue on this curve.
但是，你知道，我们当时在完成20次编辑时还在庆祝呢。
But, you know, we were taking victory laps at 20 edits delivered.
不。
No.
而且我觉得今天99%的生物制药公司和学术界都会这么做。
And I think that 99% of biopharma and academia would do that today.
我们现在能稳定地完成300多次编辑，效率超过90%。
We are delivering 300 plus at 90 plus percent efficiency consistently.
那还是一年前的事了。
That was that was a year ago.
哇。
Wow.
我们现在正在测试一千...
We're now testing thousand...
所以这是一个指数增长曲线。
So that's an exponential growing curve.
这并不意味着它会一直持续下去。
That doesn't mean it's gonna continue.
我们正在测试一千条曲线。
We're testing thousand curves.
我们正在努力...
We're working...
我们现在已经交付了一千次编辑。
We're a thousand edit deliveries right now.
这并不意味着它会成功。
It doesn't mean it's gonna work.
我一直在做这个。
I've been working on it.
每年翻三倍的基础设施？
Tripling year over year base setups?
我们不知道这能不能成，但我们正在……
We don't know if it's gonna work, but we are...
但你觉得有可能。
But you think it might.
有可能……
It could...
早期指标效率不高，但确实在起作用。
Early indicators have low efficiency, but it's working.
但到了一定程度，大规模货物交换用DNA合成就是更好的选择。
But there is a point where large cargo swaps with DNA synthesis is just better.
不幸的是，达到一定规模后，这类人其实没有商业动力去合成DNA。
Unfortunately, the people in that category don't really have a business driver to synthesize DNA after a certain scale.
所以我们刚开始在内部做这件事。
So we just started doing that internally.
那么算一下，每年翻三倍，你现在已经有一千次碱基编辑了。
So doing the math, tripling year over year, you're at a thousand base edits right now.
那给你十二……
That gives you twelve...
十二年。
Twelve years.
我们
We are
一直稳定在三百以上。
consistently north of 300.
我们正在测试一千。
We're testing a thousand.
我不想……
I don't want...
我觉得我们很有可能会搞定这个。
I think it's highly likely we will get that working.
什么时候才能真正爆发，让你想做什么就做什么？
What's the point at which it kinda really goes crazy and you can do whatever you want?
嗯，我觉得合成技术会……
Well, I think that synthesis is gonna...
我觉得合成技术会比多重编辑更快地取代它。
I think synthesis will replace multiplex editing faster.
所以基本上，就是一台机器，能生成你想要的，你知道的，那个那个
So basically, a machine that generates the, you know, the the
所有的
All the
新代码。
new code you want.
我觉得这个可能性很高，而且成功的几率更大，速度也更快。
I think that's high, I think that has a higher likelihood of success faster.
好的。
Okay.
那么，本，你提出了人造蛋的概念。
So Ben, you've come up with the artificial egg.
不是吃的那种蛋，而是能孕育生命的蛋。那么，我们离现场观众中某位女士——无论是年轻的还是年长的——在人工子宫里生孩子还有多远？
Not the kind you eat, but the kind that gives birth to an How avian far are we from a lady here in the audience, a young lady or an older lady in the audience having a baby in an artificial womb?
人类……
A human...
嗯，从技术角度来看，我认为这和从社会接受度、伦理以及监管角度来看，答案是非常不同的。
Well, from a technology perspective, think it's a very different answer than a societal and acceptance and ethics and regulation.
好的，好的。
Okay, okay.
那么，我们来谈谈……是的，
So, let's talk about Yeah,
人工子宫。我认为在二十四个月内，我们将能够完全在体外孕育哺乳动物。
artificial womb for a I think within twenty four months, we will colossal birth mammals fully ex utero.
从怀孕到分娩。
From gestation through delivery.
对，完全没经过代孕。
Yeah, that never went into a surrogate.
哇。
Wow.
哇。
Wow.
这太厉害了。
That's pretty amazing.
这太不可思议了。
That's wild.
希望更快，但我觉得二十四个月内实现的可能性很大。
Hopefully sooner, but I think that twenty four months is highly likely.
哇。
Wow.
我想问个问题。
I need to ask a quick question.
现场观众，举手示意一下，你们是不是都惊呆了？
Just show of hands in the audience, is your mind blown?
好的，就是确认一下。
Like okay, just checking.
好，明白了。
Okay, got it.
确认一下我不是一个人这么觉得。
Make sure I'm not alone here.
那Colossal的“iPhone时刻”是什么？
So what's the iPhone moment in Colossal?
是威廉猛犸象登上舞台的那一刻吗？
Is the William Mammoth stepping onto the stage?
还是有别的什么，比如……
Or is there something else that like...
展开剩余字幕（还有 334 条）
我觉得……
I think that...
我觉得是……
I think it's...
我觉得……
I think...
一个让人震惊的时刻。
A WTF moment.
想想……
Think...
嗯，我是说，我们已经有过一些这样的时刻了，但我觉得……
Well, I mean, I think we've had a little bit of those already, but I think that...
你们确实有。
You have.
我觉得下一个重大的转折点是我们向世界展示下一个灭绝物种的时候，对吧？
I think the next major inflection points are when we show the world the next extinct species, right?
我觉得那种从零到一的突破，就是下周的事。
Like I think always that kind of zero to one That's next week,
什么时候？
when is that?
开玩笑的。
Just kidding.
快了。
Coming soon.
所以，第一，通过精准基因编辑让另一种灭绝物种复活。
So, I that showing another extinct species back through precision gene editing, number one.
第二，我觉得哺乳动物的人工培育和孕育是第二件事。
Number two, I think that mammalian artificial development and gestation is number two.
然后，你知道，我们还在做一些还没公开的事情，我觉得那些和恐狼一样有趣。
And then, you know, we are working on things that we haven't shared yet that I think are equally as interesting to like dire wolves.
如果那让你感到惊喜和开心的话，我们还有更多惊喜和乐趣。
We have some more surprise and delights if if that surprised you and delighted you.
如果那让你既惊讶又害怕，那我们也有让你更惊讶更害怕的东西。
If that surprised and scared, well then we have that for you too.
我能问一下
Can I ask
你关于数据的事情吗？你知道，蛋白质折叠真的让所有人都措手不及。
you about the data So set so you know, protein folding just really snuck up on everybody?
感觉一夜之间就解决了。
It's like solved overnight.
对整个生物技术领域来说，这简直是一座金矿，我女儿每天都在用。
Been just a total gold mine of change for all of biotech, and my daughter uses it every single day.
是的，我们也在用那个，
Yeah, we use that,
太厉害了。
Massive.
我的意思是，所以相当于基因型到表型的映射问题，你说，好的，这个序列产生了，哦，那是亚历克斯·维斯纳-格罗斯。
I mean, So the equivalent, the genotype to phenotype mapping problem where you say, okay, this sequence produced, oh, that's Alex Wissner- Gross.
好的，这个序列，哦，产生了萨利姆，太好了。
Okay, this sequence, oh, that produced Salim, great.
这个，哦，是dota，没关系。
This is, oh that's a dota It's Okay.
我喜欢这最后的跳跃。
I love this final jump there.
几个
A couple
我们很喜欢那个跳跃，顺便说一下。
of We love that jump by the way.
那么，有没有一个点，在你积累的数据集基础上，你可以进行插值，然后说，好了，现在我不需要去创造它，我确切知道会产生什么。
So is there a point where given the data set you're accumulating, you can interpolate and you can say, okay, now I don't have to create it, I know exactly what would come out.
所以我们从产品角度来看，目前是按物种进行的，或者试图外推到大的动物类群。
So we're doing it on a, from a product perspective, we're doing it on a species basis currently, or trying to extrapolate to large clades of animals.
所以，尺寸是一个大问题，对吧？
So, sizing's a big one, right?
所以，如果你拿我们研究的袋狼模型物种来说，比如袋狼，我们用的是肥尾袋鼩，从有袋类小鼠到有袋类狼，体型差了1500倍，对吧？
So if you go, if you're taking, like our model species we're working with for the Tasmanian Tiger, or thylestine, is a fat tailed Dunart and it's a 1500x fold from a marsupial mouse to a marsupial wolf, right?
对。
Yep.
所以理解这一点，并推断它，以及它是如何被调控的——不只是基因，还有在发育过程中调控的方式和时机——那么，这能不能转化到比如，你能不能让一头虎鲸变得和你养的宠物金鱼一样大？
So understanding that and extrapolating that and how that, whether, what regulates that, not just the genes, but how and when it regulates in development, how does that transfer to like, can you make a killer whale the size of your pet goldfish?
可能不行，但你也许能在某些功能水平范围内进行缩放。
Probably not, but you could probably scale within certain levels of function.
如果你看某些物种，比如狗，还有某些鸟类物种群，它们有巨大的体型缩放功能，对吧，甚至超过1500倍。
If you look at certain species like dogs and also certain certain species groups of birds, they have tremendous scale functions, right, which are larger than 1,500.
所以我们正在……
And so so we're...
或者说1500倍。
Or 1,500 x.
所以我们正在从每个……的角度来看这个问题。
So so we're looking at it on a per...
从非性状工程的角度，而是从纯粹的灭绝物种复活角度来看，专门研究驱动X、Y、Z的基因。
From a non trait engineering perspective, but from a purist perspective in de extinction to look specifically at the genes that drove x y and z.
但另外，我们也在尝试在更基础的层面上进行推断，以便我们能说，好吧，我们如何影响体型，哪怕是在其他物种中一些边缘的50%分支里。
But separately, we are then trying to extrapolate that on a clay basis so that we can say, okay, how can we affect sizing even within some marginal 50% offshoot within other species?
所以我认为，皮毛、毛色、体型，还有皮肤、鳞片、羽毛这些形成的东西，都会是高度可测量的。
So I think that it's likely that coat, coat color, sizing, know, placket, like things that form skin, scales, feathers, all of that will be highly measurable.
而且能够很快地被诱导出来。
And be able to be inducible very quickly.
有意思。
Interesting.
所以你可以……
So you can...
好的。
Alright.
我不确定所有方面，但你知道，我们有一个专门研究图案和条纹的AI团队。
I don't know about everything but like, you know, we have a whole AI team that's just working on patterning and stripes.
这其实是个非常难的问题。
It's actually a really hard problem.
花纹、獠牙、条纹、鼻子的长度。
Patterns, tusks, stripes, length of snout
毛发。
Hair.
毛发。
Hair.
对。
Yes.
毛发。
Hair.
是的。
Yep.
顺便说一句，还挺多的。
By the way, the quite a bit.
你知道吗，我……
You know, I...
当我带……
When I took...
当我和克里斯带着孩子们去达拉斯参观时，看到那些毛茸茸的老鼠，真是又惊喜又开心。
When Chris and I took our boys down to Dallas to visit, it was a real surprise and delight to see the woolly mice there.
对。
Yep.
那需要多少次基因编辑？
How many gene edits did that take?
我们公开展示的第一代是八次。
The first generation, is what we've shown in the public, was eight.
太厉害了。
Amazing.
一次就完成了。
In one delivery.
八个碱基对。
Eight base pairs.
八次编辑。
Eight edits.
八次编辑。
Eight edits.
八次。
Eight.
一次递送完成八次碱基编辑。
Base edits in one delivery.
太厉害了。
Amazing.
我们来看看观众的一些问题。
Let's go to some of the audience questions here.
这是布鲁诺提出的。
This is from Bruno.
我们以后可能会有另一个版本。
We may have another version at some point
好的。
that Okay.
更有意思。
More interesting.
我等不及了。
I can't wait.
那很有意思。
That's interesting.
我们会请你回来上《Moonshots》节目聊聊这个。
We'll have you back on moonshots to talk about it.
太好了。
Great.
好的。
Alright.
所以布鲁诺问，登月计划是不是必须解决一个巨大的问题？
So Bruno asked, must a moonshot tackle one enormous problem?
前几天沙纳来了，我当时就想，天哪。
We had we had Shatner there the other day and I was like, fuck.
我们应该搞个三连发。
We should A triple.
哦，对。
Oh, yeah.
就像我们用哈维尔和塔利斯做出了Puffing一样，我们本可以让沙纳对三连发特别兴奋的。
Like like we made Puffing out of Harrier and Talis, we could have like gotten like, you know, Shatner very excited about a triple.
三连发。
Triple.
没错。
That's right.
我们需要三连发。
We need triples.
你可以把三连星带回来。
You can bring triples back.
只是没有……
Just without...
回想这个词不太对，但我
Think back is the wrong word, but I
我觉得我们可以……
think we can...
往前看。
Forward here.
把它们往前带。
Bring them forward.
把它们往前带。
Bring them forward.
好的。
Alright.
所以布鲁诺向你提出了以下问题。
So Bruno asked the following of you.
登月计划必须解决单一垂直领域内的一个巨大问题，还是可以横向发展，解决跨越多个垂直领域的看似平凡的日常需求？
Must a moonshot tackle one enormous problem within a single vertical, Or can it be horizontal, addressing a seemingly mundane everyday need that cuts across many verticals?
我们稍后也会向阿斯特罗和穆恩肖特船长提出这个问题。
And we'll ask that of Astro as well, Captain Moonshot shortly.
我觉得它可以跨越多个垂直领域。但记住，我有注意力缺陷多动症，所以我觉得缺乏专注反而让我们在多个类别中拥有更广泛的智慧，对吧？
I think it can go across verticals, Like, but remember, I have ADD, and so, like, I think that the lack of focus gives us a larger amount of wisdom across multiple categories, right?
所以我们把去灭绝看作一个系统问题，但同样的系统建模，既能用于保护物种、让物种复活，也能用于人类医疗保健领域的各种工作。
And so we look at de extinction as a systems problem, but that same system modeling that can be used to preserve species, bring back species, can also be used to do all kinds of work specifically in human healthcare.
所以我认为，如果你把基础建立在你想构建的东西上，你就可以把它应用到许多用例中，而不必那么狭窄，以至于如果错过那个窗口，它就没有其他更广泛的适用性了。
So I think that if you ground your fundamentals in what you're trying to build, I think you can apply it to many use cases and it doesn't have to be so narrow that like if you miss that window it doesn't have other broader applicability.
这里有一个来自特蕾莎的好问题。
Here's a great question from Theresa.
你们打算如何处理在体外创造哺乳动物所涉及的伦理问题？
How are you planning the ethical issues in creating mammals ex utero?
哺乳动物有情感需求，仅仅创造一个动物并不能免除你对一个活生生的生物所承担的情感责任。
A mammal has emotional needs and just creating an animal doesn't relieve you of the emotional burden of a living creature.
这其实是一回事。
That's the same thing.
比如，如果它是体外出生的，你会给它一个家庭一起生活吗？
Like if it's born ex utero, you give it a family to live with?
所以我们做了很多工作。
So we do a lot.
我觉得大多数人不知道这一点，因为你知道，媒体并不总是报道我们所做的一切。
I think most people don't know this because like, you know, the media doesn't always cover all of the stuff that we do.
我们有一个基金会。
We have a foundation.
我们把所有用于保护的技术都开源了。
We open source all of our technologies for conservation.
所以任何人都可以免费使用我们的任何技术来保护动物。
So anybody can use any of our technologies for conservation for free.
我们在全球有75个合作伙伴。
We have 75 global partners.
我们非常感激他们。
We're very grateful for them.
但我们也专门为此资助了一些项目，对吧？
But we've also funded projects specifically around this, right?
比如，猛犸象和大象是高度群居的动物，对吧？
So like, mammoths and elephants are highly social animals, right?
所以我们不会只带回一头猛犸象。
So we're not gonna bring back a mammoth.
我们要带回的是整个象群。
We're bringing back herds of them.
我们同时在进行16个不同的品系。
We have 16 different lines being worked on at the same time.
多少个？
How many?
我们有16条不同的生产线在同时运作。
We have 16 different lines working.
不会吧。
No.
你总共想带回多少头猛犸象？
How how many total mammoths do you wanna bring back?
我想带回多少？
Do I wanna bring back?
好几万头。
Tens of thousands.
好几万头猛犸象？
Tens of thousands of mammoths?
哇。
Wow.
在洛杉矶？
In LA?
这里得用哥伦比亚猛犸象，或者侏儒猛犸象……
It has to be Colombian mammoths for here or pygmy...
我要偷走……
I'm gonna steal...
我要偷走你的“房间里的大象”——问题是猛犸象放哪儿？
I'm steal your The mammoth in the room is where do the mammoths go?
但回到刚才说的……
But going back to...
等等。
Wait.
等等。
Wait.
等等。
Wait.
等等。
Wait.
我想回答那个……
I wanna answer the...
因为……
Because it's...
所以我觉得这是一个非常重要且值得深思的问题。
So I think it's a really important thoughtful question.
你看，加州秃鹫，还有其他各种濒临灭绝的物种，人们都在努力繁育和照顾它们。
And so you've had California condors, you've had all these different close to extinct species that people work on and the rearing of it.
我们发现，虽然这对今天的大象有积极的光环效应，但对Colossal正在做的事情也有更广泛的影响。
And so we find, while there's a halo effect of the positivity of this for today, for elephants, it also has a broader implication for what Colossal's trying to do.
我们在博茨瓦纳发现了一个了不起的组织，叫“大象庇护所”，他们专门照顾那些已经出生但被遗弃的孤儿象，不是因为机器人，而是因为各种原因被遗弃的。
We find in Botswana an incredible group called Elephant Havens, which is working with orphaned elephants, right, that have already been born, not actually a robot, been born and been abandoned for whatever reason.
他们正在努力使用……
And they are working to use...
他们用人工智能，用各种不同的工具，来研究如何……
They use AI, they use a lot of different tools to figure out how do we...
他们是怎么在母系社会的大象中建立合成象群的？因为大象……
How do they create synthetic herds from a very matriarchal society of elephants where elephants...
它们现在得不到那种母系社会的环境。
Where they don't currently get that.
他们怎么养育这些大象，训练它们像大象一样生活，同时还能在象群里协作呢？
How do they rear those elephants and train them to be elephants and also work together in a herd, right?
因为大象就是这样的行为方式。
Because that's how elephants behave.
另外，我们还在资助和研究从卫星成像到无人机再到人工智能的各种技术，开发不同大象迁徙路径和走廊的项目，以便了解它们移动时的社会动态和等级结构，对吧？
Separately, we're funding and doing research in everything from satellite imaging to drones to AI, building programs in different elephant migratory patterns in corridors so that we can understand the social dynamics and hierarchy of moving, right?
所以，很好。
So, Nice.
所有这些技术和数据都对今天的大象保护工作有影响，对吧？
All of that technology in that data impacts elephant conservation work today, right?
对，所以你不一定需要……孤儿象就能重新野化整个象群。
Right, so you don't have like elephant, orphaned elephants, can rewild entire herds.
但所有这些数据也告诉我们，如果这些动物是在体外出生的，我们该如何养育它们，让它们在正确的社会等级中成长？
But all of that data also informs us how we are going to rear these animals in a way where if they are born ex utero, how do they grow up in a social dynamic with the right hierarchy?
戴夫，你想说什么？
Dave, you were gonna say?
这是个真实的故事。
It's a true story.
你那位乔治·丘奇，在我们在麻省理工学院的一次演讲上，主题是全球变暖。
Your guy, George Church, was at a presentation that we had at MIT, and the topic was global warming.
他说，嗯，我有那个，你知道，研究。
And he said, well, I have the, you know, study.
我有解决全球变暖的办法。
I have the cure for global warming.
我们要带回……
We're gonna bring back...
我们有好几个，但没错。
We have several, but yes.
我们要把猛犸象带回来。
We're gonna bring back the woolly mammoth.
猛犸象的原生栖息地是西伯利亚和加拿大北部。
The woolly mammoth's native habitat is Siberia and Northern Canada.
就是那些冻土地带。
The tundras.
冻土地带。
The tundras.
在它们还活着的时候，那里没有树，因为猛犸象——对。
And back when they were around, there were no trees because the woolly mammoths Yes.
它们走来走去，把树都撞倒了。
Walk around and knock down all the trees.
其实大象在非洲也是这么做的。
And elephants actually do this in Africa too.
它们在这方面特别厉害。
They're incredible at this.
所以我们就在想，这跟全球变暖到底有什么关系？
So so we're like, well what the hell is the connection to global warming?
对。
Yep.
因为，如果没有树，草就会长起来。
Because well, without the trees, the grass grows.
草实际上比树吸收更多的碳。
The grass actually sequesters more carbon than the trees do.
草的效率大约是树的六倍，而且还有两到三倍的反照率效应，能把光反射回太空。
It's about six times more efficient and a two to three x albedo effect for light reflection to space.
所以我觉得……
So I think...
我觉得你们没跟加拿大人商量过，就直接把它们放出去了。
I don't think you guys checked in with the Canadians to see if it's okay, but you turn them loose.
把它们放出去，然后……
Turn them loose and...
我又不住在冻原上。
I don't live in the tundra.
其实我来自印度。
I'm from India actually.
说真的，我们该在哪里建立信任呢？
Seriously though, where do we build trust?
这个
This
问题提得非常好。
a really good point.
所以，在早期，我是个大数据爱好者，所以我……
So, in the early days, and I'm a big data guy, so I...
我的背景主要是软件，还有一点太空硬件，但大多数时候，我只是想跟着数据走，对吧？
Most of my background is in software and a little space hardware, but for the most part, I just want to go where the data takes us, right?
所以，有像乔治这样非常聪明、信念坚定的科学家，他们会说，如果你在冻原的这些地方有这么高的猛犸象密度，它们会对永久冻土产生影响，温度会降低六到八度。
So, you have high conviction, really smart scientists like George that will say, if you have this mammoth density at these places in the tundra, they'll have this impact on the permafrost, a lowering of six to eight degrees.
在夏季，它只会融化到一定程度。
In the summer months, it only melts so far.
所以你可以据此进行推算。
So you can extrapolate that out.
我们确实做过这个推算，结果还挺有意思的。
We have actually done that exercise, and it's quite interesting.
然后，这又回到了自上而下还是自下而上的方法，看它们对环境的影响。
Then, but it goes back down to a top down versus a bottom up approach of how they affect the environment.
科学界里，包括Colossal内部，也有人认为它们不会有那么大的影响，对吧？
There's other people in the scientific community, including at Colossal, that think that they will not have that level of impact, right?
但好消息是，总的来说，不管他们怎么看待用一万多头猛犸象在北极解决气候变化的问题，这其实不重要，因为它们在帮助恢复生态系统方面对环境有净正面的好处。
But the good news is that generally speaking, whether they fall on the, how do you solve climate change with mammoths at 10,000 plus mammoths in the Arctic, doesn't really matter because they have a net positive benefit on the environment in terms of helping restore that ecosystem.
所以我试着做的，帕默上次也提到过，就是怎么让两个固执的人坐到一起，达成一致。
So what I try to do, Palmer mentioned this in the last thing, is like how do you get two assholes in a room and get them to agree?
嗯，当这两个人都是博士的时候，这个问题就变得复杂十倍了。
Well, it gets extrapolated to like the tenth degree when they're both PhDs.
所以，以我的经验来看，我没有博士学位，有时候人们觉得我跟学术界过不去，但说实话，最难搞的人往往就是那些博士。我发现，当你把他们拉到一起，告诉他们其实双方都有道理，然后帮他们理清思路，这招真的管用。
And so, which is like, in my experience, I don't have a PhD and people think that I have a war on academia at times, but like the pretty hard to deal with people out there or PhDs in my experience, they actually have a model where I have found those when you sit them down and tell them that they're both right, and you help them walk through that, it actually works.
所以我说，也许乔治是对的。
And so what I've said is maybe George is right.
就是说，在这个经纬度上，这种密度的猛犸象会产生这样的影响。
That this level of density of mammoths at this lat long, latitude longitude will have this level of impact.
但也可能其他人是对的，说这对动植物有好处，但并不能解决碳汇问题。
But maybe others are right saying that it will have a positive benefit on the flora and fauna, but it won't cure carbon trees.
不管怎样，只要对今天的象群和已经严重退化的冻原生态系统有净正面影响，其实结果并不重要。
Either way, it doesn't really matter if it's having a net positive benefit on elephants today, as well as the ecosystem of the tundra, which is highly degraded.
所以大家都同意，现在的生态系统很糟糕，我们需要改善它，对吧？
So everyone can agree that the ecosystem sucks and we need to make it better, right?
所以，我已经这么做了，你知道，我也不确定。
And so, so I've done that and you know, I don't know.
我不想说乔治一定是对的，但我只能说，我看了数据，觉得乔治挺聪明的。
I don't wanna say that George is right, but I'll just say, I've looked at the math, and I think George is pretty smart.
好的。
Okay.
本，你刚创立了Astromech，一家一开始就是数十亿美元的公司。
Ben, you just spun out Astromech, a multi billion dollar company from the start.
Astromech在做什么？
What is Astromech doing?
所以，我们正在研究……
So, we're looking at...
这算是你刚才那个问题的引子。
So this was kind of a tee up from your question.
基础模型和动物模型并没有完全覆盖整个生命之树，它们也不会一夜之间神奇地给出一个基因组然后给我们答案。
The foundation models and animal models aren't quite looking at the entire tree of life, and they're not they're not gonna magically overnight give them a genome and give us an answer.
所以我们试图构建，就像那种……
So we are trying to build, like what are kind of inflect...
我们内部称之为“转折模型”。
We call them internally inflection models.
什么是可以接入那些基础模型的转折模型？它们可以说，好的，我们研究了并理解了关于这个基因组序列的一切，包括它是如何进化的，更重要的是，它是什么时候进化的，以及为什么在相关的支系中没有进化。
What are inflection models that can plug into those foundational models that can say, okay, we studied and we understand everything about this genome sequence across how it's evolved, and more importantly, when it evolved and why it didn't evolve in related clades.
他们观察了从气候到各种因素，是什么推动了那个变化？
They were looking at everything from climate, like what spurred that?
因为我们想构建一个预测模型，来回答：好的，那个变化去了哪里，为什么会发生？
Because we want to build essentially a prediction model to say, okay, where did that go, and why did it go?
因为我不认为Astromech会有那种神奇的、像Anthropic那样的神话，你知道，四万亿美元或者不管最新一轮融资是多少，就能解决所有问题。
Because I don't think that astromech's gonna have the magic, you know, anthropic mythos, you know, $4,000,000,000,000, or whatever the latest round is, that solves all things always.
但我确实认为，它会有足够独特的数据集，以及如何分类和理解这些数据集的方法，这样它就可以接入那些基础模型。所以当你有了全球生物库，有数百万个样本可以输入到一个神话模型或者类似神话模型的竞争对手中时，这个模型可以充当你的交通管制员，告诉你该去哪里，该关注什么。
But what I do think is I think it'll have enough of the unique data sets in how to classify and understand that data set, that it can plug into those, so that when you do have global bio vaults and you have millions of samples that you can feed into a mythos, this can be acting like kind of the, or a mythos like competitor, this can be acting as like your traffic control cop of where to go and where to focus.
让我综合一下几个问题。
Let me synthesize a couple of questions here.
你希望人们正在解决的最大问题是什么？
What's the biggest problem you wish people were working on?
最大的登月计划是什么，人们还没有着手做的？
The biggest moonshot that people I are not right
我认为，未来二十五年内我们会失去一半的生物多样性。
think that the, we are going to lose half of biodiversity in the next twenty five years.
不管是喜欢我们还是讨厌我们的人，都同意这一点。
And everyone who loves us and hates us agrees with that.
所以，我们必须采取行动。
So, we need to do something about it.
政府方面，这个问题不是靠动物园或非营利组织就能解决的。
Governments, like this is not gonna be solved by a zoo or a non profit.
我们需要数十亿美元的联邦资金，多个政府合作，至少为生命提供备份。
We have to have billions of dollars of federal funding across multiple governments working together to at least back up life.
我们为其他一切做了备份。
We back up everything else.
我们备份了照片。
We back up our, you know, we back up our photos.
我们备份了短信和邮件。
We back up our texts, our emails.
我们会备份所有东西。
We back up everything.
但说到底，我觉得大多数人备份的是短信和邮件。
But ultimately, I think most people back up their texts and emails.
所以有些人用Signal。
So some people do signal.
但总的来说，我真的觉得我们现在必须投资基础设施来备份那些物种。
But like, for the most part though, I I I really think that we we have got to right now invest in infrastructure to back those species up.
因为如果你不喜欢……
Because if you don't like...
当我们失去那些物种时，会对生态系统产生负面影响。
When we lose those species, we're gonna have negative impacts on ecosystems.
当我们失去那些物种时，会对动物的食物链产生负面影响。
When we lose those species, we're gonna have negative impacts in the food web for the animals.
所以，如果你喜欢生态系统，就应该备份它。
But so, if you like ecosystems, you should back it up.
如果你不喜欢生态系统，讨厌环境，但你喜欢动物，那你也应该备份。
If you don't like ecosystems and you hate the environment, but you like animals, you should back it up.
如果你讨厌环境和动物，但你喜欢人类，那你也应该备份。
If you hate the environment and animals, but you like fucking humans, you should back it up.
你知道，那里面的数据将来能帮到人类。
You know, there's data in there that will help humanity.
我，我就是……
I I just...
顺便说一句，我有一家公司，CEO是鲍勃·胡里，叫Singularity，我们有个项目叫Life Bank USA，就是宝宝出生时，把胎盘细胞都存起来。
As a quick aside, you know, one of my companies with amazing CEO, Bob Houri called Singularity, we have something called Life Bank USA that when your baby is born, you store all the placental cells.
对。
Yep.
基本上你就有了原始启动盘。
And you've got basically the original boot desk.
你孩子的干细胞、T细胞、自然杀伤细胞，全都在里面。
You've got your kid's stem cells, T cells, natural killer cells, everything.
这就像如果你的宝宝自带一套备用器官，你会扔掉它们吗？
And it's like if your baby came with an extra set of organs, would you throw them away?
大概不会吧，但为什么要扔掉呢？你知道，胎盘就是制造这些的3D打印机。所以，这种对安全和备份的设想真的很棒。
Probably not, but why throw away, you know, the placenta is the three d printer that creates the So, I mean, this kind of envisioning of safety, of backups is amazing.
所以，这就是你现在在迪拜做的事情吗？
So, this is what you're doing right now in Dubai?
我们正在做
We're doing it
在迪拜。
in Dubai.
刚刚宣布了与美国鱼类和野生动物管理局的合作，这是内政部长关于备份让美国伟大的自然资源的指令的一部分。
Just announced a partnership with US Fish and Wildlife here as part of Secretary of Interior's directive on backing up the natural resources that that that make America great.
所以我们现在在国内也在做这件事。
So we're doing that now here domestically.
我们还有两个政府尚未公布，等他们想公布的时候我们会宣布，他们也是我们框架的一部分。
We have two other governments we haven't announced yet that we'll announce when they want to announce that are part of our framework.
而且这也像我和乔治之前聊到的开源一样，完全是开放的。
And then it's also kind of like what George and I also talked about with Open Source, it's completely open.
所以我们……
So we...
任何非营利组织、任何学术机构、任何大型基金会、任何个人，甚至帕尔默那个秘密亿万富翁俱乐部里想加入的任何人，都可以参与。
Any non profit, any academic institution, any big foundation, any private individuals to anybody in Palmer's Secret Billionaire Boys Club that wants to throw a Everybody line wants to be boys.
你在那个亿万富翁俱乐部里吗？
You in the be boys group?
所以，谁都可以，不过这个我不评论。
So, anybody that like No comments.
我可不在那个秘密聊天群里。
I'm not in the secret chat group.
有问题要问。
Have question.
好了，
Alright,
那我们用你的快速提问来收尾吧。
let's close it out with your quick question.
我想成为像你一样有创意又疯狂想法的人，你觉得我该怎么做才能达到那种状态？
I wanna be Ben with the mind as creative and crazy as yours, How envision these do I go about doing that?
哦，我做了什么？
Oh, I do what?
我该怎么去
How do I go about
你是怎么变成这样的？
How did you become?
我该怎么去培养你那种思维方式，像你一样把技术应用得那么有创意？
How do I become, take on the mindsets that you have to apply technology as incredibly creatively as you have done?
嗯，你太客气了。
Well, very kind.
我觉得我这个人特别好奇，对吧？
I think that I'm very curious, right?
所以，我就是喜欢学新东西。
So, I like to just learn new things.
而且我觉得在这个世界，尤其是有了AI之后，每个人都能随时找到答案，我比较擅长告诉别人我不知道什么。
And I think in a world, especially with AI, where everyone's got every answer to their tool, at their fingertips, I'm pretty good at telling people what I don't know.
所以，我觉得我有点像带着孩子般的好奇心去看事情，就说，嘿，这个我不懂，但我肯定能找到答案。
So, I think I kinda take a childlike wonder to things and just say, hey, I don't know this, but I'm sure I could, go find the answer.
我还发现，大多数人……
And what I've also found, which most people...
这是我也想给所有人的一个重要建议。
This is big advice that I'd also give everyone.
人们会帮助你的。
People will help you.
比如，我是个乐观主义者。
Like, I am an optimist.
我相信科技，但我更相信人性，人们会帮助你的。
I believe in technology, but I believe in humanity first, and people will help you.
所以我觉得人们不太会主动寻求帮助。
So, I don't think people ask for help enough.
我觉得大家走在路上、坐地铁的时候，都低头看手机，但只要你稍微抬头看看，开口求助，人们是会帮你的。
I think everyone's, like, walking and on subways and shit, looking at their phones, but if you just look out for a second and ask for help, people will help you.
所以当我不明白某件事的时候，有时候我会打电话给这些人……
And so it's like, when I don't understand something, sometimes I will call people in this...
就是……就是……
In in in...
比如，Colossal 的一些顶级顾问之所以成为顾问，就是因为我直接给他们发了冷邮件。
Like, some of our top advisers at Colossal are advisers because I just cold emailed them.
就是，嘿。
Like, hey.
这个我不懂。
I don't understand this.
我的团队告诉我这个。
My teams are telling me this.
你是这个领域的全球顶级专家。
You're the world's expert in this.
能不能……
Can...
跟我开个会？
You have a meeting with me?
他们完全没有……
They have no...
这位女士或先生根本没有理由跟我说话。
This this woman or man has no reason to talk to me.
但他们还是会接这个电话。
But they'll take the call.
所以我觉得我就是带着这种普遍的好奇心，再加上……
And so I feel like I just have this general curiosity wrapped with
像你这样的人。
People like you.
他们想帮忙。
They wanna help.
不。
No.
我觉得他们会喜欢每个人的。
I think that they'll like everybody.
我真的这么觉得。
I really do.
我觉得你只要……
I think you can just...
如果有人求助，十有八九，我相信人性，他们会帮忙的。
If people ask for help, nine out of 10 times, I believe in humanity, they will help
你。
you.
关键是要开口问。
Big thing is will ask.
快速……
A quick...
大家给点掌声。
Give it up for that.
保持好奇心，以目标为导向，还有凯西·伍德在后台问的一个快问：昨天Anthropic宣布了他们的新湿实验室，能够进行类似CRISPR的技术操作。
A curiosity mindset, a purpose driven mindset, and a quick question from Kathy Wood backstage, who said yesterday Anthropic set up their, announced their wet labs that were able to create something like CRISPR like.
这有没有点燃你工作的热情？
Does this light a fire in your work?
还是会阻碍前进的动力吗？
Or does it kill forward momentum?
我觉得这特别能给人信心，对吧？
So I think it's massively validating, right?
这是个很好的问题。
Like, it's a great question.
凯西很棒。
Cathy is great.
我觉得这真的很重要，对吧？
I think it's really important, right?
你看，生物学里有太多问题需要解决了。
Like, there are so many problems to solve in biology.
这些技术还只是……
These technologies are...
我们其实连门都还没打开。
We are just like, we haven't even opened the door.
门只是刚刚裂开一条缝，真的只是裂开一点点。
Like, the door's barely cracked, barely, barely cracked open.
我觉得他们昨天宣布的那些东西，人们看到后会说，天哪，生物科技应该完蛋了，因为Anthropic已经渗透到他们生活的方方面面了。
I think that what they announced yesterday, like, people see that and say, my gosh, biotech should be dead because it should be anthropic in their in their lives.
但事实并非如此。
That's not true.
那真的不是事实。
That that's just not true.
所以我认为这对整个行业来说是一个巨大的转折点，它展示了那些懂AI的公司——是的，RR有生物学背景——那些AI公司明白这将是他们技术最被接受的用例和部署之一，而且人们希望拥有更健康的家庭，更健康、更长寿的生活，对吧？
And so I think that that was a huge watershed moment for the industry to show that like, you have the AI companies that understand, and yes, the RR has background in biology, you have the AI companies that understand that that will be one of the most accepted use cases and deployments of their technology, And that people want to have healthier families, healthier, longer lives, right?
所以这对社会来说是一件好事。
So it was a great thing for this society.
真的很短。
Really short.
你有p doom吗，还是你甚至不需要考虑这个问题？
Do you have a p doom or do you need to even think about the question?
我是个乐观主义者。
I'm an optimist.
谢谢。
Thank you.
我，我是个乐观主义者。
I I am an optimist.
比如，我不同意...
Like, I don't agree with...
你知道，我觉得我们会有一些吓人的时刻。
You know, I I think we're going to have some scary moments.
当然。
Of course.
而且我觉得这没关系。
And I think that's okay.
因为我确实相信人类的聪明才智能够解决这些问题。
Because I do believe in human ingenuity to work through those problems.
但我觉得你得有一个……
But I think that you've gotta have a con...
一个对话。
A conversation.
我确实觉得媒体现在有点过度渲染了，说得客气点。
I do think that the media is overselling that a little bit right now, to to to be kind.
但我确实认为，你知道，我们真的会达到那个目标。
But I do think that, like, you know, we we really will get there.
还有你……
And you...
我们会有一些惊险时刻，但就像飞机遇到气流一样，大家最终都能安全降落。
We're gonna have a couple scary moments, but it's like, you have turbulence on planes and everyone still lands.
这没什么大不了的。
Like, that's okay.
好的。
Alright.
欢迎来到乐观主义者的奥斯卡颁奖典礼。
Welcome to the Oscars of optimism.
让我们欢迎Ben Lamm！
Give it up for Ben Lamm.
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。