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。
我们生活在一个医疗需求迅速增长的世界里。
We live in a world where the demand for healthcare is just rising so quickly.
每天有上万人达到 Medicare 的年龄门槛，而我们根本无法快速培训出足够的医生来照顾这些人。
We have 10,000 people aging into Medicare every single day, and we just can't train doctors fast enough to take care of all these people.
在未来三到五年内，临床医生的工作方式将发生根本性的变化。
The practice surface area for the clinician will look fundamentally different in the next three to five years.
当我早年部署自己的软件时，医生们会说：‘你又搞了个新工具？’
When I was deploying my own software back in the day, doctors would grow and then they'd be like, you got another tool?
他们会觉得：‘你为什么又塞给我一个新东西？’如今，他们在日常生活中体验到的科技便利与工作中所使用的工具之间的差距，第一次真正缩小了，这彻底改变了他们对技术的看法。
Like, why are you stuffing this down my The delta between the magic of the tools that they're experiencing in their consumer lives and what they do in their work has for the first time narrowed just even a little bit to a point where it's just fundamentally changed the nature of how they view technology.
我认为，这是第一次我们看到了希望。
I think this is the first time where there's hope.
我们找到了用更少资源做更多事情的路径。
There is a pathway to doing more with less.
我们找到了让医生和护士这份工作变得更有成就感的路径。
There is a pathway for the job of being a clinician, being a nurse, to be a fulfilling one.
患者就医体验也有了一条路径，可以不再像有时那样令人困惑和充满绝望。
There is a pathway for the experience of a patient not being as confusing and full of despair as sometimes it is.
当尼基尔·布杜马还是斯坦福大学的博士生时，他的一位导师因医疗失误去世了。
When Nikhil Buduma was a PhD student at Stanford, he lost a mentor to a medical error.
他没有完成他的医学博士联合培养项目，而是选择退学，投身于用技术改善医疗体系的工作。
Instead of finishing his MDPhD, he dropped out to work on fixing healthcare with technology.
他在早期的深度学习领域度过了多年，曾与后来创立OpenAI的研究人员共事。
He spent years in the early deep learning community, including time with the researchers who would go on to found OpenAI.
他见证了2017年Transformer架构的诞生，并目睹了规模定律逐渐显现其威力。
He watched Transformers emerge in 2017 and saw the scaling laws start to click.
然后，他做了一件不同寻常的事。
Then he did something unusual.
他和联合创始人创办了一家医疗机构，不是为了永久提供医疗服务，而是为了真正理解运营一家医疗机构意味着什么，与医生合作、实施电子健康记录系统，并看清技术在哪些地方存在不足。
He and his co founders started a medical not to deliver care forever, but to understand what it actually feels like to run one, to work with doctors, implement an EHR, and see where technology falls short.
这段经历成为了Ambience公司的基础。
That experience became the foundation for Ambience.
如今，该公司与国内一些最大的学术医疗中心合作，超过75%的临床医生每天都会使用该产品，其中一个医疗系统预计通过该平台实现3000万美元的净新增利润。
Today, the company works with some of the largest academic medical centers in the country, over 75% of clinicians use the product daily, and one health system is projecting $30,000,000 in net new margin from the platform.
a16z普通合伙人朱莉·尤与Ambience Healthcare的首席执行官兼联合创始人尼赫尔·布杜马进行对话。
A16z general partner Julie Yoo talks with Nikhil Buduma, CEO and cofounder of Ambience Healthcare.
非常高兴你来到这里，尼赫尔·布杜马是Ambience的首席执行官，也是联合创始人之一。
Super excited to have you here, Nikhil Buduma, is the CEO of Ambience, one of the cofounders.
Ambience，我觉得你们是2020年成立的吧？
And Ambience, I feel like, you know, you guys were founded, what, twenty Twenty twenty.
2020年。
2020.
我想，你是第一个通过Zoom做出的投资。
I think we were the first investment you made on Zoom.
是的。
Yes.
通过Zoom。
On Zoom.
没错。
That's right.
就在新冠疫情爆发、所有人都陷入混乱的时候。
Right when COVID was hitting and everyone was going crazy.
但更疯狂的是过去五年多来人工智能领域发生的变化，你可以说是全程亲历了这一切。
But even crazier is what has happened in the last, what, you know, five plus years on the AI front, and you sort of had a front row seat to the whole thing.
所以你应该分享一下你的背景，你是如何进入这个疯狂的人工智能领域的，以及在市场过去几年如此迅速演变的过程中，作为临床人工智能领域最前沿公司之一的掌舵人，你的经历如何？
So you should share a little bit about your background, how you got involved in this whole crazy AI space, and then what's it been like to, you know, be at the helm of one of the most cutting edge companies in the clinical AI space as the market has evolved as rapidly as it has over the last few years?
是的。
Yeah.
过去十二年在人工智能领域简直不可思议。
I mean, the last twelve years in AI have been kind of insane.
这个故事几乎可以分为几个不同的阶段，嗯。
And there's almost like a few different arcs Mhmm.
故事的脉络。
To the story.
但你知道，我刚开始职业生涯时以为自己会成为一名医学博士兼哲学博士，但后来我退学了，主要是因为一位导师因医疗失误去世了。
But, you know, I started my career thinking I was gonna be an MD PhD, and then I ended up dropping out mostly because I'd lost a mentor to a medical error.
而且，我在思考，再培养一位医学博士兼哲学博士对世界带来的净收益， versus 如何系统性地解决这些问题。
And, you know, thinking about the net benefit to the world for another MD PhD versus thinking about how do you solve these problems systemically.
我在这里的硅谷长大，从小就与斯坦福大学紧密相连。
And I grew up here in the Bay Area and sort of was embedded at Stanford.
所以你可能记得，当安德鲁·吴正在构建深度信念网络并用GPU扩展这些模型时的情景。
And so you might remember when Andrew Ng was building deep belief nets and scaling up those models of GPUs.
那是大约2010年。
This is, like, 2010.
我想在那之后，我开始写一本关于深度学习的书。
I think after that, started writing a book on deep learning.
到了某个阶段，我被邀请到硅谷，与这个领域的早期研究者们交流。
And I think at a certain point, got called out to Silicon Valley to hang out with a lot of the early researchers in this space.
我们当时都聚集在格雷格的公寓里，这个小团体后来就成了OpenAI。
We were all hanging out in Greg's apartment for a little while, and that group became OpenAI.
那时，人们普遍认为无监督学习和强化学习将催生通用智能和通用推理模型，但没人真正清楚这会是什么样子。
In that time, I think there was a belief that in general, unsupervised learning and reinforcement learning was gonna give rise to general intelligence and general reasoning models, but no one really had a clear sense of what that was gonna look like.
有一群人正在研究OpenAI Gym，试图让强化学习代理在模拟环境中行走。
You had a group of people working on OpenAI Gym, trying to get RL agents to walk in simulated environments.
还有一群人正在研究变分自编码器。
You had a bunch of people working on variational autoencoders.
我认为，在那个时期做任何研究的难点在于，每六个月就会发生一件事，整个深度学习的某个分支就会突然崩溃。
And I think the hard part of doing any kind of work in that time is every six months, something would happen and an entire sort of branch of deep learning would just sort of, like, collapse.
对吧？
Right?
比如，我们从认为深度信念网络将是首选架构，转变为大约十二个月内没人再关心深度信念网络。
Like, we went from thinking deep belief nets were gonna be the architecture of choice to no one cares about deep belief nets in, like, about twelve months.
嗯。
Mhmm.
有趣的是，我和我的联合创始人迈克退后一步，说我们最感兴趣的是如何将这些技术应用于医疗保健。
And I I think what was interesting is, you know, my cofounder Mike and I took a step back, and we said, we're most excited about thinking about how do we take these technologies and apply them to health care.
于是我们做了件疯狂的事：我们创办了一家医疗交付机构，不仅运营这家诊所、与医生合作、实施电子健康记录系统，同时还把研究实验室里看到的所有技术应用到实践中。
And so we actually did the crazy thing, which is we started a care delivery asset, and we started to not only run that practice, work with doctors, implement an EHR, but at the same time, we were taking all of the techniques that we were seeing at the research labs and bringing them into the practice.
对吧？
Right?
2017年，Transformer模型问世了。
So 2017, transformer came out.
我们已经在生产环境中使用Transformer来处理理赔数据，预测住院风险。
We were using the transformer in production to ingest claims data, predict risk of hospitalization.
我认为，在Transformer问世之后，整个研究社区几乎都转向了这一架构，因为它清晰地解决了语言建模和推理中的许多难题，而之前的架构却未能做到这一点。
And I think during this time, sort of post transformer, we just saw the entire research community sort of just collapse on this architecture because it was so clear that it solved many of the challenges around language modeling and reasoning in a way that we didn't quite see the previous architectures work.
突然间，规模定律和RLHF出现了，我们在餐桌上就看到了GPT-2的演示。
And all of a sudden scaling laws and RLHF, we saw a demo of GPT two over the dinner table.
我认为，迈克和我逐渐形成了一种共识：我们正站在某种指数级突破的边缘。
And I think Mike and I were coming to this conviction that we were on the precipice of some sort of exponential here.
当你看到你最尊敬的人、你最亲密的朋友，甘愿冒险筹集数十亿美元来扩展这些架构时，你就知道，某些重大的变化正在发生。
And when you see the people you respect the most and some of your closest friends go risk their reputations to raise billions of dollars to scale up these architectures, they're like, something is happening here.
是的。
Mhmm.
在那些早期的日子里，当你看到这些早期Transformer的性能时，比如GPT-3刚发布时，大家都说它在医疗领域表现很差。
In those early days, when you saw the performance of those early transformers, like, today, when GPT-three first came out, everyone was like, oh, it sucked at health care.
对吧？
Right?
当时还存在大量的幻觉风险。
There there was still a ton of hallucination risk.
你知道，它并没有在大量关于医疗实践、指南等的专有数据上进行训练，这些数据可以说并不在互联网上。
You know, it wasn't trained on a lot of the proprietary data, sort of not on the Internet, so to speak, about medical practice and guidelines and all that.
而显然，过去几年里它的表现有了巨大的提升。
And then, obviously, it's improved, you know, drastically over the last few years.
但那时候，它在医疗环境中的表现真的很糟糕吗？实际表现如何？
But back then, like, was it super bad, or how did it actually perform in a health care setting?
当你把它应用到你的第一家初创公司时，是的。
And when you were applying it to your first start up Yeah.
我的意思是，为了让它真正适用于临床环境，你花了多长时间进行事后训练和微调？
Like, how long was the pull of, you know, sort of post hoc training and and fine tuning that you had to do to to get to get it to actually work in a clinical setting.
迭代周期简直疯狂。
The iteration cycles are insane.
对吧？
Right?
如今，人们会觉得，我可以直接拿出一个大语言模型，它在某些用例中就能直接工作。
So today, people are like, well, I could pull LLM out of the box, and it kind of works out of the box for certain use cases.
但一旦你进入真正专业的垂直领域，就必须大力投入事后训练。
And then the moment you move into really deep domain verticals is when you have to really invest in post training.
嗯。
Mhmm.
那时候，老实说，你得重新思考如何进行预训练和事后训练。
Back then, honestly, you had to rethink how you did pretraining, how you did post training.
当时的迭代周期长达一年。
The iteration cycles were like a year.
是的。
Mhmm.
你当时在构建深度定制的专属数据集，以及为自动驾驶汽车所必须搭建的整个机器学习运营基础设施。
And you're building deep custom bespoke datasets and all the sort of, like, ML operations infrastructure that you had to build for self driving cars you were pulling over.
是的。
Mhmm.
那时候要让这些架构运行起来，是完全不同的世界，而且规模也截然不同。
So it was a very different world to get these architectures to work and also very different scales.
对吧？
Right?
在2017年和2018年，我们讨论的是一些模型拥有几千万，甚至可能上亿的参数，而如今我们谈论的是万亿级参数的模型。
We were talking about tens of millions, if not maybe a 100,000,000 parameters for some of these models in 2017 and 2018 versus now we're talking about trillion plus parameter models.
是的。
Mhmm.
而你们则做了相反的事——许多初创公司一开始是构建技术，然后将其销售到医疗健康市场，尤其是这样。
And you guys kind of did the reverse thing where a lot of startups start by building technology that gets sold into the health care markets in particular
嗯。
Mhmm.
然后意识到向医疗提供方销售技术有多困难，比如，于是决定转向全栈模式，说我们就想自己先用自己的产品。
And then realize how hard it is to sell technology to providers, you know, for instance, and then decide to go full stack and say, we just wanna eat our own our own dog food.
我们想获取更多与技术平台之上所提供服务相关的价值。
We wanna capture more value associated with the services delivered on top of a technology platform.
你们则走了相反的路，一开始就是全栈模式，然后决定在下一轮业务中转型为平台公司。
You guys did the the opposite where you started full stack and and then, you know, decided that you wanted to do a platform company in your next play.
跟我们聊聊，你们是如何坚定地认为这是正确的方向的。
Talk to us about, like, how you got to conviction that that was the right move.
是的。
Yeah.
我认为当我们创办上一家公司Remedy时，我们非常清醒地意识到，我们对医疗系统运营者的真实处境几乎没有同理心。
I think when we started our previous company, Remedy, we were just extremely sober to recognize that we had very little empathy for what it's like to sit in the shoes of an operator of a health system.
为了真正理解整个背景、待完成的任务以及潜在机会，我们强烈认为，我们首先必须自己承担起这份责任。
And to truly understand the entirety of the context and the job to be done and the opportunities, we felt very strongly that we had to hold the responsibility ourselves first and foremost.
我认为，这不仅让我们在公司内部获得了快速迭代的灵活性，能够看清哪些有效、哪些无效，更重要的是，它为我们未来若真要创办一家平台公司奠定了基础——让我们能切身体会到，作为一位CEO，面对1%到3%的利润率、员工队伍的倦怠危机、复杂的IT系统架构，以及与Epic合作的种种挑战时，究竟该如何做决策？
And I think that not only did give us the flexibility to be able to have the rapid iteration cycles inside of the company to see what worked and what didn't work, but I think it also set the foundations for if we ever did start a platform company, what would it, like, feel like to sit in the shoes of the CEO looking at one to 3% profit margins, looking at a workforce and burnout crisis amongst your staff, having to navigate the complexity of an IT stack and having to work with Epic, and thinking about, like, how do you even make decisions in that world?
因此，在很多方面，运营医疗交付业务的经历为我们如今打造平台公司提供了丰富的直觉：如何真正成功地做到这一点？如何构建一个不仅能打动一线终端用户，还能赢得经济决策者——CEO、CFO、首席运营官——共鸣的产品？因为他们必须确保整个系统在财务上是可持续的。
And so in many ways, I think going through the experience of running a care delivery asset gave us a ton of intuition for now as we're building a platform company, how do you actually do that successfully, and how do you build something that can resonate not just with the end user on the ground, but also resonate with the economic buyer, the CEO, the CFO, the chief operating officer of an organization that has to make sure that this system is financially sustainable too.
嗯。
Mhmm.
让我们稍微聊聊Ambience的历史。
And, you know, just to get into a little bit of the the kind of the history of Ambience.
某种程度上，你们已经覆盖了医疗服务提供者市场的广阔领域，对吧？
Like, you guys, in some ways, have traversed a fairly large surface area of the provider market, right?
你们如今显然在与全球一些最顶尖的医疗系统合作。
You've worked with you're obviously today working with some of the most prominent health systems in the world.
你们也曾与规模较小的医疗团体，甚至早期的数字健康公司合作过。
You also have experience working with smaller medical groups, even digital health companies at the beginning.
谈谈你对市场细分的看法吧？
Talk to us about what's your view on the segmentation of the market?
此外，我们今天在人工智能采用方面处于哪个成熟阶段？
And also, what's the stage of maturity that we're at today with respect to AI adoption?
因为我认为过去几年最令人惊讶的一件事是，医疗提供方曾经是技术采用的最后堡垒。
Because I think that's one of the most surprising things that's occurred in the last few years is that providers used to be the last bastion of technology adoption.
他们一直是最落后的群体。
They were always the laggards.
他们总是落后于其他领域，比如支付方、生命科学公司等等。
They were always, like, behind every other segment, like the payers and life sciences and what have you.
但就在过去几年里，他们突然间成为人工智能采用最快的一批群体之一，不仅是在行政管理场景中，更是在像你们产品这样的临床诊疗点上，直接服务于医生。
And then all of a sudden, in the last couple of years, they are arguably amongst the fastest adopters of AI and not just in the administrative setting, but, like like, your product, like, at the point of care for doctors.
那么，你们是如何推动这一切的？在刚才我提到的这些医疗提供方中，你们看到哪些领域采纳率最高？
And so how do you sort of, like, rock all that, and where do you see the pockets of, like, sort of highest adoption rates amongst that set of providers that I described?
是的。
Yeah.
也许我们可以退一步，思考一下为什么这对与我们合作的组织来说如此有吸引力。
Maybe we would take a step back and we think about why this is so compelling to the organizations we work with.
现实是，你看看医生今天的一天，根本没什么乐趣，是的。
The reality is, like, you look at a doctor's workday today, there's not a lot of joy Yep.
医学实践中已经很少有快乐了。
In the practice of medicine anymore.
你无法直视你的病人。
You can't look your patient in the eye.
你总觉得自己落后于进度。
You're constantly feeling like you're running behind.
你觉得自己上学是为了照顾病人，真正面对诊室里的临床问题，但你大部分时间却在做各种其他事情。
It feels like you went to school to take care of people and to actually grapple with the clinical medicine in the room, but most of your time is spent doing all sorts of other stuff.
对吧？
Right?
你得在电子病历里到处翻找信息。
You're, like, searching through the electronic medical record to find information.
你得写病历。
You're writing notes.
你得应对成千上万条编码和计费规则，这些规则因付款方类型和地区而异。
You're navigating these thousands and thousands of coding and billing rules that are different by type of payer, different by region.
它们每年都在变化。
They change year over year.
你会想，我上学可不是为了做这个的。
And you're like, I didn't go to school to do this.
所以，在很多方面，市场本身早已存在强烈的内在需求。
And so I I think in many ways, the organic pull from the market was there for a very, very long time.
这根本不是一个问题：嘿，
And it wasn't a question of, hey.
有没有人需要能完成这项工作的技术？
Can is there a demand for technology that can do this?
长期以来，真正的问题其实是：技术本身能否足够好地满足这一需求？
The question really was for a long time, can a tech is can technology even serve this need well enough in the first place?
是的。
Mhmm.
我认为，随着市场逐渐发展，出现了一种明显的分化，一边是高复杂性、高价值的使用场景，另一边是低复杂性、低价值的使用场景。
And I think as as as the market sort of evolved, there's almost a a bifurcation between what I think are largely the high complexity, high value set of use cases and the low complexity, low value set of use cases.
因此，我们特别选择在一些最大的医疗网络和学术医学中心中扎根。
And so for us, we've we found a home specifically in some of the largest IDNs in the academic medical centers.
我们认为这是一个非常值得深耕的领域，因为这些环境中所涉及的医学范围极其广泛，且在广度之下的深度也极具挑战性。
And part of why we think that's a very interesting place to play is the just this the the vast breadth of medicine that's being practiced in those environments and the depth across that breadth is extremely challenging to go to go tackle.
所以，如果你在思考如何构建系统，我们的观点是，未来三到五年内，临床医生的工作界面将发生根本性的变化。
So you think about if you're trying to build you know, our view is that the practice surface area for the clinician will look fundamentally different in the next three to five years.
比如，今天你对比使用Ambience的体验和一两年前的体验，就已经明显感受到巨大的变革性差异。
Like, today, even if you compare the experience today using something like Ambience versus what it was a year or two years ago, it already looks so so transformatively different.
但挑战在于，当你身处这些高度复杂的临床环境中时，一名主要管理多种慢性病的全科医生，与一名专科肿瘤医生的需求截然不同。
But the challenge is that when you're operating in these really complex clinical settings, the job of a primary care physician who's primarily trying to manage multiple chronic diseases versus the needs of a subspecialized oncologist, they're so they're so different.
他们做出的决策也不同。
They're making different decisions.
他们在电子健康记录系统中的工作流程也完全不同。
They have different workflows inside of the EHR.
他们查看的是不同的数据来源。
They're looking at different data sources.
因此，要构建能够服务广泛医学领域的基础设施，真的非常困难。
And so to be able to build the kind of infrastructure that serves the the broad range of medicine is really, really hard.
另一方面，一旦你进入中型市场或三到五家、十家医生的小型诊所，复杂性就会大幅下降，因此服务起来容易得多。
On the other hand, the moment you move to sort of the the mid market or the small sort of three, five, 10 doc practices, the complexity drops dramatically, and so it's much easier to serve.
所以，我猜测，总体来看，你会看到大量玩家争相争夺中型市场。
And so my guess is, in general, you'll see a proliferation of lots of players trying to compete over the mid market.
例如，我认为，这个领域的电子病历系统正在努力重塑自己，成为以人工智能为先的公司。
For instance, I I think, you know, you've got the EHRs in that space that are trying to reinvent themselves to be AI first companies.
还有AI记事员在瞄准这个市场，我认为他们很快发现，要创造足够的价值，就必须为这些机构掌控更多技术栈。
You've got AI scribes going after that space, and I think they're quickly finding out that to create enough value, they have to own more and more of the stack for for for these organizations.
然后我认为，企业级市场现在开始明朗化了，事实上，只有少数几家公司在这一市场拥有真正的入场资格。
And then I think the enterprise segment of the market is starting to shake out now where I think the reality is there's only a couple of players that have even had a right to play in that market.
除了Ambience之外，大多数公司都难以真正满足这些机构在实践过程中所需的复杂性。
And most of them outside of Ambience have really, really struggled to actually meet these organizations, the complexity at which they they need to practice.
对吧？
Right?
比如，会有许多组织来找我们。
So for instance, we'll have so many organizations come to us.
他们部署了某个系统，但只有20%的医生真正使用它。
They've rolled out something, and only 20% of their doctors actually use it.
即使那些使用的人，也只是
And even those who do use it, they use
用得不够好。
it just not good enough.
用得不够好。
It's not good enough.
在说吗？
Speaking?
好的。
Okay.
是的。
Yep.
而且即使使用它的那些人，也只在百分之二十、三十、四十的就诊中使用。
And and even the ones that you use it, they're using it for twenty, thirty, 40% of their visits.
另一方面，我认为对我们来说，我们与几家大型学术医学中心合作，现在我们的规模已经达到：他们超过百分之七十五的临床医生每天都在诊室使用Ambience。
On the other hand, I think for us, we work with several large academic medical centers, and we're at the scale now where seventy five plus percent of all their clinicians use Ambience every single day in clinic.
他们将它用于超过百分之八十的所有就诊，这带来了完全不同的机遇和规模。
They're using it for eighty plus percent of all of their visits, which is a completely different sort of opportunity and scale.
因此，我认为这在很大程度上反映了市场的分化：有一部分市场复杂度极高，非常难以服务。
And so I think that's sort of how the market has has shook out in many ways, which is there's a high complexity part of the market that's really, really hard to serve.
但如果你能服务好它，其他人就很难与你竞争。
But if you can serve it, is it's hard for others to compete in.
对。
Yeah.
当人们谈论这个领域，比如所谓的临床智能、用于临床智能的AI时，其中一个常见的说法是。
One of the tropes when people, you know, talk about this space in general, like, sort of the let's call it, you know, clinical intelligence, AI for clinical intelligence in general.
你知道，其中一个常见的说法是，最终通用基础模型会变得如此强大，具备全方位的智能，以至于它们将在那些只需整合信息并给出建议的领域中胜出。
You know, one of the tropes is that eventually the generalist foundation models will just get so good at, like, a global set of intelligence that, you know, they're gonna win these categories where, you know, all you have to do is is basically, you know, synthesize information and then make a recommendation.
而真正关键的竞争将在于行动层，也就是实际由智能体执行的工作。
And then really where the game is gonna be at is is kind of the action layer, like, you know, the actual work that is done agentically.
一旦你获得了这些信息，你会怎么用它？
What do you do with that information once you have it?
你如何真正地为患者和医疗提供者创造临床价值？
How do you actually, you know, clinically create clinical utility basically for for the patient and the provider?
第一，你是否认同这个前提：临床智能的临界点即将到来，任何仅专注于智能层的公司很快就会变得高度同质化，因此真正的竞争将发生在行动层？
Number one, do you agree with that premise that, you know, the the sort of the the cliff that is clinical intelligence, like, that that that is coming up soon and that, like, any company that's only doing the intelligence layer will soon become, like, very commodified, and that and therefore, the real competition is gonna be on the action layer?
还是你认为，正如你所说，仅靠临床智能本身的发展周期如此之长，以至于在通用基础模型全面覆盖医疗实践的各个领域之前，我们仍然有许多年的时间？
Or do you actually think, like, to your point, that the the length of the pull on just clinical intelligence alone is so long that you still have, you know, many years before any of the generalist foundation models will be able to crack the code comprehensively across the full surface area of of of medical practice?
这是一个非常有趣的问题。
So it's a fascinating question.
我可能会从一个稍有不同的抽象角度来回答：根据我们的经验，AI的迭代速度与产品迭代速度本质上是不同的。
I might answer it a slightly different abstraction, which is that from our experience, AI clock speed is fundamentally different from product clock speed.
嗯嗯。
Mhmm.
我认为这样想的部分原因是，智能的多个方面随着每一代基础模型的演进而不断提升。
And and and part of the reason I think about it that way is there's several aspects to intelligence that do get better with every generation of foundation model.
在很多方面，我觉得在这个领域构建产品令人担忧的地方在于，能力进化得太快了，我们在Ambience内部经常使用这个词。
And in many ways, I think what's kind of frightening about building in this world is the capabilities are evolving so quickly that and we use this word constantly inside of Ambience.
我们正在构建一个地板是熔岩的世界。
We're building a world where the floor is lava.
是的。
Yeah.
你需要一种能够迅速响应并随着能力持续演进而彻底自我重塑的组织结构。
And you have to have the kind of organization that can respond to and, on a dime, be able to reinvent themselves itself as capabilities continue to evolve.
对吧？
Right?
因此，我们投入了大量时间预测未来十八个月内能力会是什么样子，然后为那个未来做准备，而不是现在的能力。
So we spend an insane amount of time predicting what the capabilities will look like over the next eighteen months and then building for that future as opposed to building for the capabilities now.
嗯。
Mhmm.
话虽如此，我认为我们发现，这些模型要在医疗领域真正有效，仍面临巨大的“最后一公里”问题。
That being said, I think what we find is that there is still a massive last mile problem for these models to be effective in health care.
你可以将这个问题分解为几个不同的类别。
And you can break it up into a couple of different categories.
首先，这些模型是否从一开始就具备正确的上下文？
And it's gonna start it's gonna start with, do the models even have the right context to begin with?
嗯。
Mhmm.
我想你一定深刻体会到，要构建出从隐藏在FHIR API和专有API背后的记录系统中提取上下文的基础设施，是多么混乱。
And I think you probably have a deep appreciation for this, just how messy it is to even build out the right infrastructure to be able to pull context out of systems of record hidden behind FHIR APIs and proprietary APIs.
数据模型太过混乱且不一致。
The the data models are so messy and inconsistent.
你会遇到一些具体的标准，它们更像是某种规范概念。
You'll get specific standards where it's sort of like a concept of a specification.
是的
Yeah.
所以你必须得
And so you have to, like
人们就是会把自由文本随便塞进某个字段里，诸如此类。
People just, like, stuff, like, free text into a random field and all that.
没错
Yep.
百分之百
A 100%.
所以，即使只是在不同电子健康记录系统之间从记录系统中提取上下文这个问题，在我们开始构建Ambience时还是个未解决的难题，而我们解决了它。
So you're like, even just the problem of across EHR instances being able to pull out the context from systems of record in and of itself was an unsolved problem when we started building Ambience that we solved.
嗯
Mhmm.
对吧
Right?
因此，能够从EHR实例的任何部分读取数据，包括其底层的数据仓库，并利用这些数据建立经过整理的层级以在此基础上构建智能，这曾经是一个未解决的问题。
And so the ability to read out of any part of an EHR instance, including the data warehouse underneath it, and then using that and having a groomed layer to be able to then build intelligence on top, that was an unsolved problem.
我认为，很多人没有充分认识到，对于人工智能公司来说，最有价值的是决策轨迹。
I think another thing that a lot of folks don't don't fully appreciate is that the most valuable thing for AI companies is decision traces.
大多数EHR系统基于可变数据结构，这意味着你本质上会破坏这些决策轨迹。
Most EHRs are built on mutable data structures, which means that you inherently destroy Yeah.
决策轨迹。
The decision traces.
因此，从根本上说，你必须重新思考如何收集这些数据的架构，才能让智能真正发挥作用。
And so, fundamentally, you have to rethink the architecture of how you actually collect this data in the first place to even make intelligence Mhmm.
要针对特定领域进行定制。
Be specific to the domain.
因此，数据层还存在另一个大问题，但你还需要解决一系列其他数据层问题。
And so that's another sort of big problem in the data layer, but there's a a whole sort of range of data layer problems that you have to solve.
然后在中间部分，我认为从临床智能的角度来看，我们面临一个巨大的挑战，那就是如何定义质量。
Then in the middle, I think we have a really big challenge from a clinical intelligence standpoint around defining quality.
我们用太多不同的方式来定义质量。
We define quality in way too many ways.
没错。
Exactly.
这尤其具有挑战性，因为大多数用例都是开放式的，最初甚至简单到：如果病历中存在多条相互矛盾的信息。
And it's especially challenging because most of the use cases are open ended use cases, and they start as as trivial as if you have multiple pieces of contradictory information in the chart.
你有一个病人，问题列表里没有甲状腺问题的任何迹象，但这个人六个月前就被开了甲状腺药物。
You got a patient where there's no indication of thyroid problems in the problem list, but this person is on thyroid medication that was prescribed six months ago.
这说明这位病人的状况如何？
What does that mean about the state of this patient?
就连在这一层面上确定真相都很困难。
That like, even just resolving truth at that level is is tricky.
那么，如果你在住院环境中，突然有四位不同的临床医生对这位病人进行体格检查，而其中一次非常具体的体格检查是48小时前通过专科会诊记录下来的呢？
Then what if you're in the inpatient setting and all of a sudden you've got four different clinicians doing a physical exam on this patient, and you've got one very specific physical exam that happened about forty eight hours ago that came in through a specialty consult?
你该多大程度上依赖那次专科检查，而不是两小时前由全科医生做的体格检查？
How much do you index on that versus the the physical exam that happened two hours ago by a generalist?
嗯嗯。
Mhmm.
这是一个困难的推理问题。
That's a hard reasoning problem.
非常重要。
Matters a ton.
慢性病管理和资质认定都很关键，而且我们还意识到，最终需要进入病历的大量内容在就诊过程中从未被口头解释过。
Chronicity and credentialing matters a And and then there's also the realization that so much of what ends up and needs to end up in the medical record is never verbally sort of explained in a visit.
对吧？
Right?
例如，肿瘤科医生可能在脑海中沿着一条治疗路径进行决策，然后用患者能理解的语言告诉他们下一步该怎么做。
So for instance, an oncologist may be walking through a decision tree on a care pathway in their brain, which is then expressing to the patient in words that they can understand what the next step is.
但进入临床记录的，是这个决策过程的痕迹，而不是对患者所说的话。
But what goes into the clinical note is that trace of the decision tree, not the words that were spoken to the patient.
嗯嗯。
Mhmm.
因此，将质量定义为那位专业肿瘤医生所期望的内容，是这样的，对吧。
And so defining quality as to what does that special subspecialized oncologist expect Mhmm.
这本质上非常复杂。
Is is fundamentally complicated.
然后你还会遇到一些特定场景，比如ICD-10编码。
And then you've got certain use cases where, for example, ICD 10 coding.
你把两位医生关在一间屋子里。
You put two doctors in a room.
他们只有60%的时间能达成一致。
They agree 60 of the time.
嗯。
Mhmm.
你让一位医生处理ICD-10编码的问题。
You put a doctor on an ICD 10 coding problem.
你向他提出一个问题。
You ask them a question.
你让一个编码员来处理这个问题。
You put a coder on that problem.
你问他们一个问题，而不是让他们在房间里有十分钟时间讨论后再问他们问题。
You ask them a question versus you have both of them in the room with ten minutes to debate before you ask them the question.
你会得到三个不同的答案。
You get three different answers.
嗯。
Mhmm.
所以，如何定义质量呢。
And so how yeah.
定义质量实际上是一个根本性的难题，必须通过智能层来解决，而如今的基础模型并未解决这个问题。
Defining quality is actually just a fundamentally hard problem and one where it has to be solved with the intelligence layer, and it's not being solved by the foundation models today.
嗯。
Mhmm.
然后我认为，最后一部分是，对于这些公司来说，要取得成功，随着能力的飞速发展，有些公司在垂直行业中，迭代速度本来就很迅速。
And then I think the the the last piece is that for these companies to be successful, as capabilities are evolving really, really quickly, you've got some companies in vertical industries where iteration speed is naturally fast.
对我们来说，必须建立非常深入的关系，才能从原型开发过渡到在TST中部署，再上线到生产环境，最终实现与终端用户的实时学习闭环。
For us, you have to have really, really deep relationships to be able to go from prototyping something to deploying something in TST to turning it on in prod to then being able to actually make live learning loops with end users.
我认为我们组织解决的一个关键问题是：如何与标志性企业建立极其深厚的关系，从而在不到三十天内完成从概念到实际部署并实现用户学习的全过程，这在我们这个领域是前所未有的。
And I think one of the things that we've solved as an organization is how do you build extremely deep relationships with marquee organizations to be able to go from concept to live and deploy learning with users within, like, less than thirty days, which is unheard of in our category.
所以我想
So I
我认为，要让公司在这一领域具备价值，必须解决的远不止临床智能层面的诸多问题，这说得通。
think there's, like, all these things that you have to solve for companies to be valuable in this category that are just far beyond clinical intelligence, so that makes sense.
所以
So the
你的意思是，地板会保持熔岩状态更长时间。
floor will stay lava for much longer is what you're saying.
是的。
Yeah.
完全正确。
100%.
100%。
100%.
或许我想问问你：假设你现在要重新构建Kairis，想象一下你生活在一个这样的世界里。
And maybe a question for you is, imagine you lived in a world like, if you were to rebuild Kairis today.
嗯。
Mhmm.
对吧？
Right?
那需要多长时间？
How long would that take?
如果你有一个环境，有人已经解决了集成问题、数据完整性问题，以及与医疗系统的合作关系问题，你觉得对你来说需要多长时间？
And then if you had an environment where someone had solved the integration problem, had solved the data integrity problem, had solved the sort of relationship with the health system problem, how long do you think that would that would take for for you?
我的意思是，这确实是我一直都在思考的问题，我也看到很多公司实际上正在做某种形式的我原本会做的事情。
I mean, this is definitely a question I think about all the time, and I see lots of companies that are actually doing, you know, forms of of what I would have done.
但这只是重新梳理了一下你刚才说的一些内容。
But it it kinda gets just to, you know, rehash some of what you said.
我觉得有两个方面是我经常思考的。
Think I there's two components that I think a lot about.
一个是我们将要构建的产品的形式是什么？
One is what is the form factor of the product that we would have built?
它会从根本上不同。
It would have been fundamentally different.
对吧？
Right?
我们当时是一个传统的企业级SaaS产品，部署在呼叫中心环境中，必须培训人类员工正确使用软件流程以获得理想结果。
Like, we were a legacy enterprise SaaS product that got deployed into a call center setting where we had to train the humans to use the software workflow in in the right way, to get the right outcome.
而我们知道，在这些环境中，员工的流失率天然就非常高。
And as we know in those environments, the churn rate of employees is just naturally super high.
所以一旦你培训完一批人，他们就离开了，你又得重新培训下一批，导致人们使用该软件的合规率存在巨大差异。
So as soon as you train one batch of people, they leave, and then you gotta retrain them again, and you just have a huge air sort of range of compliance rates, let's call it, with which people are using that software.
因此，如果你能让一个智能代理来完成这项工作，就能大大减少结果的偏差。
So if you can actually make an agent do the work instead, you know, that removes a ton of the potential drift in outcome.
因此，我们现在经常看到的一种情况是：与其仅仅给呼叫中心提供工具，为什么不直接成为呼叫中心，构建语音AI代理，根据你希望在整个医疗系统中标准化的规则来实际完成工作？
And so that's, you know, one thing that we see a lot of these days is like, instead of just giving tools to a call center, why not be the call center and build voice AI agents that are effectively doing the work based on the rules of the road that you want kind of systematized across the entire health care system?
我想到的第二件事正是你所说的，关于这一点我想向你提个问题，那就是数据层。
The second thing I think about is exactly what you said, and I have question back to you on this is, you know, the data layer.
对吧？
Right?
比如，能够高效安排预约的最大障碍之一，正是我们当初试图解决的问题，那就是缺乏一个权威数据源。
Like, one of the biggest impediments to being able to do scheduling efficiently, which is what we were trying to do, is that there is no source of truth.
没有一种标准化的方式来表示临床排班。
There's no standardized way for representing clinical schedules.
不仅如此，这还涉及到每一位医生。
And not only that, it's individual physicians.
值得一提的是，他们显然都有各自不同的风格、偏好，以及不同的行医方式。
To their credit, they obviously all have very different styles, very different preferences, different ways they wanna practice.
而任何医疗系统都没有动力要求所有医生统一采用一种方式，因为这些医生非常稀缺。
And no health system has an incentive to tell every doctor to systematize it in one way because these doctors are very scarce.
他们希望留住这些医生。
They wanna retain them.
他们希望这些医生能吸引他们想看的患者。
They want those doctors to attract the patients that they wanna see.
因此，底层数据的异质性成为在系统层面实现目标的巨大障碍。
And so the heterogeneity of the underlying data was a huge impediment to getting it right at the system level.
如果我们今天来创建这家公司，或者像Ambience正在做的那样，实际上就是在建立一个新的数据记录系统。
And you know, if we were to build that company today or even, you know, what Ambience is doing, you're effectively creating the new system of record.
对吧？
Right?
你拥有了一种前所未有的、完整而细致的数据体系，在医疗系统中从未存在过。
Like, you're you have a a full texture of data that has never existed before in the health care system ever.
对吧？
Right?
医生与患者之间对话的高精度语义解析。
Conversation grade resolution on what's being said between a clinician and a patient.
然后，正如你所提到的，这如何转化为以某种专业方式记录下来的内容，并建立这些关联呢？
And then to your point, how does that translate to what actually gets documented in an esoteric fashion and creating those links?
这种事以前从未存在过。
Like, that's never existed before.
预约系统也是同样的道理，对吧？当预约被安排时，你听到患者的偏好，然后如何将这些偏好与医生的偏好进行比对，进而创建一种将两者关联起来的语义？这是一套全新的信息，以前从未存在过。
Same thing with scheduling, right, where, you know, as as appointments are being booked and as you hear what was the preference of the patient and then how do you bump that against the preferences of the doctor and then, you know, create a semantic that links those two things together, like, that's a de novo set of information that, like, never existed before.
所以我认为，对于当今正在构建这类系统的公司来说，另一个机会在于：不仅要思考需要完成的工作，还要创建一套全新的数据，这套数据不仅能更好地训练你的模型，使其在未来表现更优，甚至可能本身成为一套全新的系统。
So I think therein is another opportunity set for any company that's building in this day and age is, you know, how do you not only think about the work that needs to be done, but, you know, creating an entirely new set of data that not only trains your models better to, like, you know, perform better in the future, but effectively could become an entirely new system itself.
对吧？
Right?
我的问题是，当你思考电子病历的未来时，你怎么看？
And, I mean, that would be the question to you is as you think about the future of the EHR.
正如你所说，电子病历系统正试图垂直整合到这些工作流程中。
Like, to your point, the EHRs are trying to vertically integrate into these workflows themselves.
我认为，显然，我不看好它们能赢得这场竞争。
You know, I think, obviously, I'm betting against them being being the ones who win that game.
但，你知道，现在就是关键时刻吗？
But, you know, is this the moment?
我们所有人都等待了几十年，就为了等到这样一个时刻，能够真正颠覆现有的主流电子病历系统，并有机会引入新工具，一方面把这些系统带入现代时代，另一方面也让初创公司有机会去掌控这一层。
Like, all of us have been waiting for decades for the moment where we can finally disrupt, you know, that incumbent EHR layer and have a shot at potentially introducing new tools that, you know, number one, bring those those systems into the modern era, but also enable startups to have a shot at, you know, kind of owning that that layer.
你认为现在就是关键时刻吗？或者说，我们需要满足哪些条件，才能开始逐步渗透到这些大型ERP厂商当前掌控的系统层级？
Do you think this is the moment, or, you know, what needs to be true for us to start to kind of eat into the layers of the stack that are owned today by these kind of monolithic ERP players?
我认为现在就是关键时刻，而且有两个原因。
I think this is the moment, and it's the moment for two reasons.
第一个原因与你刚才提到的密切相关：那些能够构建出未来临床实践界面和未来管理架构的组织，必须能够突破当前的瓶颈——因为我们如今所有的遗留系统都建立在不适合AI和产品迭代速度的旧架构上。
The first is deeply related to what you were talking about, which is the organizations that figure out how to build the practice surface area of the future and the administrative stack of the future have to be able to unlock the sort of level of AI to product clock speed that is bottlenecked by the fact that today, we've built all these legacy systems on a legacy architecture that's not actually optimized for AI to clock speed to product clock speed.
是的。
Mhmm.
因此，Ambience的一项创新在于，我们构建了一层位于电子病历之上的中间层，它能从各类数据源系统中提取所有数据，并将其转化为便于构建AI产品的格式，从而大幅降低开发新应用场景的边际成本。
And so one of the innovations of Ambience is we've actually built out a layer that sits on top of the EHR that pulls all of the data out of sort of the systems of record, puts it in a form that makes it easy to build AI products on top so that the incremental cost of building a net new use case dramatically drops.
因为你可以想象，每一个应用场景和功能领域都在使用相同的底层数据源系统，你没必要一次又一次地重复搭建同样的基础设施。
Because you can imagine every single use case and application area is leveraging the same underlying systems of record, and you don't wanna recreate that same infrastructure over and over and over again.
对吧？
Right?
所以，对我们来说，从两个产品发展到十二个产品，再到明年推出二十四个产品，我认为一旦你拥有了我们构建的这种基础设施层，整个组织的产品迭代速度就会发生根本性变化。
And so for us to go from, like, two products to 12 products to next year 24 products out in the market, I think once you've had that infrastructure layer that we've created, it fundamentally changes your your product clock speed as an organization.
所以，这可能是第一个洞察，而构建这样的基础设施绝非易事。
So that's probably the first insight, and it's not a trivial piece of infrastructure to create.
对于我们而言，为了实现这一点，花了好几年的深度研发才做到。
Like, it for us, it took several, several years of deep r and d to be able to even do this in the first place.
我认为第二个原因是这样的。
I think the second is is the following.
随着能力的提升，世界的运行逻辑已经发生了根本性变化。
The physics of the world have fundamentally changed with the set of capabilities.
所以你刚才提到，如果今天要重新打造一个Kairos，它的形态还会是一样的吗？
And so you were talking a little bit about how if you were to build a Kairos today, like, would the form factor even be the same?
对吧？
Right?
我认为这正在整个系统的每一个使用场景中发生。
And I think that's happening across every use case across the span of the entire system.
对吧？
Right?
循环周期可能是你和我们都非常深入的一个领域。
Rev cycle is probably one that you're really, really deep in as we are too.
其中一个问题是，如今当你面临循环周期问题时，系统该如何解决？
And one of the questions is, like, today, when you have a rev cycle problem, how does a system solve it?
这其实有两个方面。
And and it's really twofold.
要么他们投入培训，教那些不想学编程的临床医生去做正确的事；要么就得组建庞大的后台团队，尽可能多地审查并事后纠正临床医生的错误，而这显然也让临床医生反感，且成本极高、效率低下。
Either they invest in training to teach clinicians who don't wanna learn coding to try to get them to do the right thing, or you've got massive back office teams to try to, like, review as much as you possibly can and then try to correct the clinician afterwards, which obviously clinicians hate as well and is extremely in is extremely expensive and inefficient.
因此，对我们而言，如果能将这种专业知识提炼成一个模型——虽然这极其困难——但如果你能做到，并以软件的成本将其分发到每一个与循环周期相关的交互场景中，同时你掌控了用户面前的整个流程，那么你该如何利用这套能力，让做正确的事变得简单？
And so the question for us as well, if you could distill that expertise into a model, which is extremely hard to do, but if you could, then and you can distribute it at the cost of software across every interaction where rev cycle expertise is is relevant, and you then own the window in front of that user, how do you then leverage that set of capabilities to make doing the right thing easy?
你该如何让做正确的事变得显而易见？
How do you make doing the right thing obvious?
突然之间，这开始瓦解了收入周期之所以如此构建的许多基本假设。
And all of a sudden, that starts to disintegrate a lot of the underlying assumptions for why rev cycle is constructed the way that it is.
是的。
Mhmm.
对吧？
Right?
所以，比如，构建一个用于账单前阶段的产品。
So, like, building a product for pre bill.
在一个你可以与临床医生紧密合作，创建更准确的事实来源并实时自动裁定的世界里，这还有意义吗？
Does that make sense anymore in a version of the world where you can work closely with the clinician to create a more accurate source of truth and automatically adjudicate real time.
所有这些都实时进行？
All of it in real time?
是的。
Yeah.
我们所讨论的一切的下一个明显步骤，就是自主的AI医生。
The next obvious step from everything that we're talking about is, you know, autonomous AI doctor.
对吧？
Right?
所以你们今天就像是一个协作者。
So you guys are today a copilot.
我认为，目前所有仍在实际使用的用例，都需要医生来签署笔记或文档等。
Every use case, I believe, that is still being used in the wild requires a physician to basically sign off on the note or the documentation, etcetera.
那么，为什么你们的平台不能成为一个自主的AI医生呢？
What what's between why why shouldn't your platform be an autonomous AI doctor?
也就是说，从今天的能力到未来我们很可能实现的完全自动化场景之间，到底存在什么障碍？在这些场景中，临床工作将完全自动完成，甚至做出真正的临床判断？
Like, what's between, you know, the capabilities today and, you know, the future, you know, reality that we will likely have a lot of this being done, you know, in a fully automatic fashion doing clinical like, actually making clinical judgments?
我们生活在一个医疗需求迅速增长的世界里。
We live in a world where the demand for health care is just rising quick so quickly.
每天有10,000人进入医疗保险体系，而我们根本无法快速培训出足够的医生来照顾这些人。
We have 10,000 people aging into Medicare every single day, and we just can't train doctors fast enough to take care of all these people.
而且，如今为医生的患者名单增加一位新患者的成本高得令人痛苦。
And on top of that, the the sort of cost of adding one more patient to a doctor's panel today is so painful.
因此，我认为这些技术的一个前景在于，我们如何用更少的资源做更多的事？
And so I think one of the promises of these technologies is how do we do more with less?
我们如何让临床医生能够轻松地说：我确实想接诊更多患者？
How do we make it painless for a clinician to say, I do wanna see see more patients.
我确实希望扩大我所拥有的医疗资源和专业能力的可及性。
I do want to increase access to the care and expertise that I have.
以一种长期可持续的方式实现这一点。
Do it in a way that is long term sustainable.
我认为大多数机构都希望扩大对患者群体的服务覆盖。
I think most organizations want to increase access to patient populations.
问题是，如何在不增加临床医生负担的情况下实现这一点？
And the question is how do you do that without asking your clinicians to do more?
这其中很大一部分在于，我们如何开始将部分工作交由虚拟医疗团队成员来承担，让他们代替医生完成下一步工作。
Well, a big part of that is how do we start to offload some of that work to sort of virtual care team members who can sort of take the next step on behalf of the on behalf of the clinician.
因此，我们已经开始在就诊开始前进行相关实验。
And so, already, we're starting to experiment with before the visit even starts.
展开剩余字幕（还有 233 条）
所以在今天的就诊前，Ambience已经提前预判了医生在见到患者之前需要了解的所有信息。
So before before a visit today, Ambience has sort of anticipated everything this doctor would need to know before they even see this patient.
对吧？
Right?
想象一下你的一个分身。
So imagine a clone of yourself.
你是一名内分泌科医生。
You're you're an endocrinologist.
你的分身花了数小时仔细研究了所有数据，并整理出了一份关于这位患者你所需了解的所有信息的摘要。
A clone of yourself has pored through all the data for hours and hours and hours, and it's put together a summary that of everything you need to know for this patient.
现在，再进一步往前推一步：你有一个能够访问所有这些上下文信息的智能代理，它还能预测临床医生最可能关心的问题。
Now imagine you you move that one step further upstream, which is you've got an agent that has access to all this context that can also anticipate, well, what are the most likely questions that are top of mind for the clinician?
你在就诊前就提出这些问题，然后把这些内容也一并加载到医生的摘要中。
And you start asking those questions before the visit even happens, and then you've loaded that also up into the summary for the doctor.
反过来，你和你的医生进行了一次非常棒的对话。
And then on the flip side, you had a great conversation with your physician.
临床医生在就诊结束后会给你一个事后总结，并通过患者门户发送给你。
The clinician sort of wrapped up with you with an after business summary that gets sent back to you through the patient portal.
如果你把这个过程变成一个能与你持续对话的智能代理，会发生什么？
What happens if you turn that into an agent that can have a continuous conversation with you?
回答你的问题。
Answer questions.
再次确认。
Double check.
你拿到药了吗？
Did you pick up your medication?
确保你在Quest Diagnostics完成了那项实验室检查。
Actually, make sure that you got that sort of that lab test done at Quest Diagnostics.
你拿到治疗幽闭恐惧症的阿普唑仑了吗？因为你在见肿瘤科医生之前要做CT扫描，而你对这个CT扫描感到焦虑。
Did you pick up your Xanax for your claustrophobia because you have to get a CT scan before seeing the oncologist and and you're anxious about that CT scan.
所有这些事情。
All these things.
现在有一个虚拟护理团队成员代表你统筹所有这些事务，这会是什么样子？
What does it look like for now there to be a virtual care team member to actually help quarterback all of those things on your behalf?
这才是这些能力真正潜力的体现。
That is what the true promise of these capabilities is.
假设我是某家医院系统的首席执行官。
So let's say that I'm the CEO of a hospital system.
你所描述的听起来令人无比兴奋。
And, like, what you're describing sounds incredibly exciting.
但如今，我有五个不同的供应商都在争夺这块蛋糕。
But today, I've got five different vendors who are all buying for, you know, that, like, that pie, essentially.
我有我的电子健康记录系统。
So you got your I have my EHR.
我还有一众基础模型公司，它们都在推出医疗健康产品，并声称要进入这一领域。
I've got, you know, the foundation model companies who are all doing you know, launching health care products and, you know, claiming that they're gonna get into this space.
还有Ambience，它们最初从AI记录员起步，但正在迅速扩展。
I've got Ambience, you know, kind of starting with AI Scribe, but expanding rapidly.
我还有我的AI收入周期供应商。
I've got my AI revenue cycle players.
我有我的AI临床决策支持供应商。
I have my AI clinical decision support players.
你知道，这些对我来说是五大主要供应商类别，它们都在朝着这个愿景汇聚。
You know, those are kind of, like, to me, the five major categories of players that are kind of all converging on this vision.
你对我的推销是什么？
What's your pitch to me?
告诉我，我该如何理解整个领域？我该如何考虑在这些类别中进行投资，才能确保长期有效，比如五年的时间跨度，而不是每隔六个月就得更换供应商？
Like, tell me how I grok this whole space, and how do I think about making investments against those categories in such a way that is durable for me, you know, for over multiple years, like a five year time horizon, as opposed to me being in a situation where I need to rip and replace vendors, like, every every six months?
是的。
Yeah.
从运营者的角度来看，我会从两个角度来思考。
The way that I would think about it from the shoe of the operator is probably twofold two lenses.
首先，关于AI，有很多兴奋和噪音，但营销和噪音与我购买后，我的临床医生是否会真正使用它，之间有着巨大区别。
The first is there's a lot of a lot of excitement and noise around AI, but there's a big difference between marketing and noise versus if I bought this, will my clinicians actually use it?
因为如果你没有达到足够的采用率，其他一切都无从谈起。
Because if if folk if there if you don't have the right level of adoption, nothing really matters.
对吧？
Right?
所以我认为，目前对于许多AI应用场景来说，评估结果都相当负面——这些公司非常擅长描绘愿景，但一到实际落地，这些技术的采用和使用情况却令人非常失望。
And so I think that's probably where, right now, the jury is looking really negative for a lot of AI use cases, which is these companies are very good at pitching a vision, but then when it comes to brass tacks, the adoption utilization of these technologies is extremely underwhelming.
这个时代的一个酷之处在于，你可以直接让医生去使用它，没错。
And that's one of the cool things about this era is that you can literally just let the doctors use it Yep.
然后用他们自己的脚投票。
And kind of put with their own feet.
对吧？
Right?
百分之百。
A 100%.
我认为这一点对任何直接与Ambience合作，或与曾与Ambience合作的组织交流过的机构来说，都已非常清晰。
And I think that's that already is it's it's so clear for any organization that either works directly with Ambience or just talks to an organization that has worked with that works with Ambience.
这些公司大多尝试过我们这个类别中的每一个玩家，而我们已经出色地占据了企业内每位临床医生诊疗流程中的关键窗口。
And most of these have tried every single one of the players in in the category that we've done an incredible job of actually owning the window of care in front of every single clinician inside of the enterprise.
如果你做不到这一点，那只是基本要求。
And if you can't do that, that's table stakes.
嗯。
Mhmm.
所以第一个衡量标准是：你能否真正实现足够的采用率和使用率？
And so that's the first lens is can you actually get the level of adoption utilization?
第二个是，医疗系统并不是传统的企业级SaaS公司。
And the second is, look, health systems are not traditional enterprise p d PSaaS enterprises.
你没有数亿美元的现金可以随意花在那些酷炫的‘玩具’上。
You don't have hundreds of millions of dollars of cash lying around to be able to invest in in toys, in cool toys.
因此，问题就变成了：我该如何为这些项目融资？
And so the question then becomes, how do I actually fund these things?
所以我们采取的第二个视角是：最终，AI令人兴奋的原因之一在于，这是我们行业首次认为，有一类技术能够从根本上改变运营利润率。
And so the second lens that we take is, at the end of the day, one of the reasons why AI is exciting is because this is the first time that we think as an industry that there is a class of technologies that can fundamentally change operating margin.
是的。
Mhmm.
我认为，如果能做到这一点，那些能够加速其AI产品迭代速度的组织就能够成功采用AI。
And I think if you get this right, the organizations that unlock their AI product clock speed, they'll be able to adopt AI.
他们将能够释放新的运营利润率形式。
They'll be able to unlock new forms of operating margin.
这种运营利润率使他们能够投入更多资源来吸引更优秀的人才。
That operating margin allows them to invest more in tools that attract better talent.
更优秀的人才意味着更高的业务量。
That better talent means more volume.
更高的业务量等于更多的收入。
More volume equals more revenue.
更多的收入意味着有能力投资于更多的AI技术。
More revenue equals the ability to invest in more AI.
你会因此解锁一个惊人的飞轮效应。
You get sort of unlock this crazy flywheel.
我认为从CEO和董事会的角度来看，那些能够有效实现这一点的组织会不断积累优势，成为患者首选的医疗目的地。
And I think from the CEO's perspective and the board perspective, the organizations that figure out how to do this effectively, they compound and become the destination of choice for patients.
而那些做不到的，则面临被整合的风险。
And the ones that don't, they're at risk of consolidation.
因此，真正的问题变成了：我究竟该和谁合作来改变运营利润率？
And so then the question really becomes, who can I actually work with to change operating margin?
电子健康记录系统会与我建立合作伙伴关系，承诺帮助我改变运营利润率吗？
Are the EHR's gonna work with me in a partnership model where they're committing to change operating margin?
我认为这件事对很多人来说太可怕了，因为它要求你的产品真正有效。
I think it's a thing that's so scary for people to do because it requires your products to actually work.
你必须真正擅长管理。
You have to actually be good at management.
你必须擅长衡量和归因。
You have to be good at measurement and attribution.
但我觉得，因为我们已经建立了所有这些能力，所以我们愿意去一家医疗系统说：我们会言行一致，我们的成功将取决于你我共同帮助你改变运营利润率的能力。
But I think because we've built out all those capabilities, we're willing to go to a health system and say, we we will actually put our money where our mouth is, and our success is gonna be dependent on your ability, our ability to help you change operating margin.
是的
Mhmm.
因此，我认为这两者的结合在关系初期创造了一系列确凿的证据，几乎就像一种宗教式的转变——当你走进一个使用这种环境的机构时，你根本无法从一个房间走到另一个房间而不被走廊里的人拦住。
And so I think the combination of those two things creates a set of proof points in the beginning of a relationship where it's almost like a religious conversion that happens, where now you walk into an organization that leverages the ambience, and you can't go from room to room without being stopped in the hallway.
这培养了信任感和对未来的期待，让你有资格随着时间推移开展更多工作。
And that builds the level of sort of trust and excitement for the future that earns you the right to do more over over time.
是的
Mhmm.
对吧？
Right?
所以我认为，在与马基学术医学中心的客户合作六个月内，我们就从‘我们在做文书助手’转变为‘我们希望尽快把你们路线图上的所有功能都实现’。
So I think within six months of working with the Markey Academic Medical Center customers we have, we go from, hey.
我们从做文书助手，到希望尽快把你们路线图上的所有功能都实现。
We're working on the scribe to we want everything on your road map as quickly as possible.
我认为这是因为人们亲眼看到它真正奏效后产生的兴奋感，以及CFO第一次看到AI实施效果时说：这显然能带来数倍于投入的回报，我们终于获得了之前没有的新利润空间。
And I think it's because of that sort of, like, excitement from seeing it actually work, and then the CFO for the first time looking at AI implementation and saying, this clearly pays for itself many times over, and we can take the new margin we didn't have.
你知道吗，我们合作的一个中心，预计在归属争议之后，将产生超过3000万美元的新增利润。
You know, one one one center we work with, they're projecting over $30,000,000 of net new margin post attribution debate.
这完全是Ambience带来的额外收益。
That's just free by Ambience.
因为他们不再需要雇佣人工记录员了？那这里的归属指的是什么？
Because they don't have to hire human scribes anymore, or what what's attribution there?
其中很大一部分是收入周期管理（RCM）。
A big part of it is RCM.
很大一部分是提升了诊疗效率和可及性。
A big part of it is improving throughput and access.
嗯。
Mhmm.
但收入周期管理（RCM）是其中很重要的一部分。
But RCM is a big part of it.
嗯。
Mhmm.
现在你可能会想，我有这么多利润可以投入更多人工智能。
And now you're like, well, I have all this margin to invest in more AI.
嗯。
Mhmm.
而这正是开启这一飞轮效应的开端。
And that's sort of the beginning of unlocking that flywheel.
所以一旦你看到了这一点，我认为这从根本上改变了你对这一类别的看法。
And so once you see that, I think it just fundamentally changes your perspective on the category.
这也很有趣，因为我知道，就在一年前，当我与医院系统的高管们交谈时，还处于AI速记工具采用的第一波浪潮。
It's interesting too because I know, like, even just a year ago, I remember the first wave of AI scribe adoption when I talked to c suite executives at hospital systems.
我当时问，回报率是多少？
And I said, you know, what's the ROI?
是什么促使你们购买这些产品？
Like, what made you purchase these products?
他们基本上说，听好了，朱莉。
They they basically said, listen, Julie.
实际上，财务回报率并没有那么高。
Actually, there's not that much financial ROI.
我们这么做只是为了留住员工。
We're just doing it for retention.
我们希望员工开心。
We want, employee happiness.
我们希望医生们感受到我们在为他们着想，给他们每天来上班的理由。
We want our physicians to feel like we are looking out for them and giving them a reason to come to work every day.
现在看来，重点已经明显转向了真正的硬性回报，也就是第二阶段。
And now it seems like it's very much shifted, you know, to to actually hard ROI kind of in, like, phase two.
因为当时人们根本不知道这甚至可能实现，是的。
It's because people didn't know it was even possible Yeah.
在那个时间点。
At that point in time.
对吧？
Right?
要实现这一点，你必须能够追踪用户在电子健康记录（EHR）内的行为，将这些行为映射到具体的编码，从而避免CDI查询、防止拒付，并将这些环节整合起来，实现新增现金收入和降低收款成本，而且这种方式必须能够真正通过CFO办公室的审核。
To to to do this, you have to be able to track where user behavior is happening inside of the EHR, map that user behavior down to a code, that code then being submitted, a CDI query prevented, a denial prevented, and then putting that altogether to new cash you're collecting in reduction in in cost to collect, and doing that in a way that's gonna actually meet the muster of the CFO's office.
而要实现这一点，我们需要下载数据仓库，并构建一套真正适用于CFO办公室的完整分析系统。
And for us to do that requires us to basically download the data warehouse and build an entire analytics stack that would actually work for the CFO's office.
因此，如果不做这些，CFO当然永远不会相信。
And so, like, in the absence of doing that, of course, the CFO is never gonna believe it.
但现在，我们已经看到客户中出现了真实的证据：CFO们会向其他CFO们讲述这一领域能创造多大的价值，而这需要大量工作和深思熟虑，包括选择正确的应用场景、切实提升编码绩效、真正推动变革管理，以提升运营利润率。
But now we're seeing the real proof points across customers where CFOs will actually tell other CFOs how much value this this category can create, and it takes a lot of work and thoughtfulness on getting the right use cases, actually moving coding performance, actually doing change management to move the move the mark marker on operating margin.
是的。
Yeah.
我们之前只讨论了医疗提供方这一侧的市场，但显然，你所提到的这些内容对付款方至少也有影响。
We've talked solely about the provider side of the market, but, obviously, some of the things that you're you're speaking about have implications for certainly for payers at a minimum.
是的。
Yeah.
而且，你知道，医疗提供方部署AI，付款方部署对抗性AI，最终演变成一场围绕应收管理的机器人之间的生死对决。
And, you know, the the sort of steel cage death match of, you know, providers implement AI, payers implement counter AI, and then we have a late bot on bot crime around RCM.
这实际上正在发生，正如我们所知，人们甚至在他们的收益电话会议上也在讨论这个问题
That's Like, actually playing out as we know, and, you know, people are actually talking about on their earnings calls even
是的
Yeah.
一些大型全国性保险公司就是如此。
With some of the large national payers.
你认为这种情况会在哪里体现出来？
Where do you see that playing out?
你如何看待这个问题？你们是否在相关领域开展工作？如果能听到一些真实案例，了解我们如何解决这一挑战，那就太好了。
Like, how do you see and are if you guys are doing any work along those lines, it would be great to hear just, like, real life case studies of how do we solve that that challenge.
你可能知道，我们与许多一体化的组织合作。
You you you probably know we work with a lot of organizations that are integrated.
他们拥有保险计划，并与保险公司保持紧密关系。
They have a plan, close relationships with plans.
对的
Yep.
我认为我们发现的世界，从我们的角度来看，更加乐观，嗯。
I think what we're finding is that the world may our view is more optimistic Mhmm.
这意味着，如果你有一个系统，能够真正理解信息的真相，并且深刻理解它——这不仅仅是被动监听，还包括对所有过往上下文的深入把握。
Which is if you've got a system that actually can understand source of truth and understand it really, really well, which is not just ambient listening, it's also deeply understanding all the past context as well.
所以，你需要在系统记录之上叠加这样一层能力。
So it's like you you need that sort of, like, layer on top of the systems record.
但如果你能深刻理解信息的真相，确切知道就诊过程中发生了什么，并能以高保真度和清晰的审计轨迹回答任何问题，这不仅对医疗机构有利。
But if you deeply understand source of truth and you know exactly what happened in the visit and you can answer any question with high levels of fidelity with clear audit trails, it's not just a win for the organization on the health system side.
但从长远来看，我认为对支付方也有利，因为反过来，我们正在谈论一场AI与AI的军备竞赛。
But I I think long term, it ends up being a win for the payer as well because on the flip side, you know, we're talking about an AI versus AI arms race.
我们已经看到了人力与人力的军备竞赛——支付完整性团队正在建立，以与医疗机构的RCM团队协作。
We've already seen the labor versus labor arms race because you've got payment integrity teams being built out to sort of work with the RCM teams on the health system side.
但我认为，一旦有了共享的信息真相，未来五年内RCM的投资回报率可能会变为负值，支付完整性的投资回报率也很可能在五年内变为负值。
But I think what ends up happening is once you have a shared source of truth, then the the ROI of RCM becomes potentially negative over the next five years, and there's a very real likelihood that the the ROI of payment integrity also becomes negative over the next five years.
因此，合作才是理所当然的选择。
And it just makes sense to to collaborate.
所以从长远来看，我对这一点持乐观态度，即使短期内似乎有点难以应对。
So I think I'm long term optimistic on this one even if the short term seems a little bit tricky to navigate.
就像AI能够做到的事情的标准已经提高了。
As, like, the bar has gone up on, like, the things that AI can do.
我记得我们刚投资的时候，有很多事情。
Like, there's a lot of things that I remember, like, when we first invested.
那时候我们觉得，这简直是白日梦。
You know, we were kind of like, oh, a pipe dream.
我们需要额外开发，或者等AI模型发展到足够成熟才能实现XYZ。
We'll have to do additional development and or wait until AI models, you know, come along far enough for us to do x y z.
而如今，我们已经做到了。
And then here we are.
那些事情实际上已经变得非常可行。
You know, those things are actually very, very feasible.
我们的路演材料差不多还是原来的样子。
Our pitch deck's about the same.
宣传材料也差不多。
The pitch deck's about same.
真的改变了很多。
Really changed.
是的。
Yeah.
对。
Yeah.
但能力确实已经进化了。
But the capabilities have definitely evolved.
比如，现在还有什么难办的事情吗？
Like, is there any remaining like, what's hard today?
还有什么目前难以实现的？
Like, what's still hard to do?
你从客户那里听到他们想做什么，但以目前的技术还做不到？
What are you hearing from your customers that they want to do that, you know, we're not actually able to do yet with, you know, the capabilities that are out there?
那么，瓶颈是AI本身，还是你之前提到的关于最后一公里工作流、集成、RCM等所有其他因素呢？
And is it is it AI that's the bottleneck, or is it all the other stuff that you talked about with respect to, like, last mile workflow, integration, RCM, all that kind of stuff?
是的。
Yeah.
我认为我们现在处于一个很难分清界限的阶段。
I I think we're at a place where it's hard to disentangle.
比如，你会把基础模型层的瓶颈、训练后阶段的瓶颈，和产品层面的瓶颈分别怎么界定呢？
Like, what do you consider, like, a bottleneck at the foundation model layer versus a bottleneck in post training versus a bottleneck in product?
嗯。
Mhmm.
我认为对我们来说，一些仍然棘手和困难的使用场景是跨医疗场景的上下文串联。
I think some of the use cases that are still tricky and hard for us are as we're thinking about cascading context across care settings.
嗯。
Mhmm.
所以，当你面对一位被送进急诊室、随后转为住院治疗的患者时，你如何预测接下来最紧迫的行动是什么？
So how do you anticipate what's gonna happen when you've got a patient who's admitted to the ED, and now they're upgraded to the inpatient setting, and now you're trying to make predictions on what next fast action really is.
这部分工作仍然有些棘手。
Some of that some of that work is is still a little tricky.
我认为，一般来说，这类模型在预测建模方面尚未很好地解决。
I think, generally, like, predictive modeling is not particularly well solved yet by this class of models.
但总的来说，我们发现，如果你组建了一个拥有正确应用研发能力、正确内部临床专业知识和RCM专业知识的团队，并让它们与应用研发团队紧密配合，那么现在有太多东西可以构建，我们几乎感觉不到瓶颈。
But I think in in in general, what we're finding is that if you've built the right team with the right applied r and d expertise and the right sort of internal clinical subject matter expertise and RCM subject matter expertise to pair really, really well with applied r and d teams, we're in a world where, like, so much is there's just so much to build that we don't really feel that bottleneck.
我们面临的瓶颈更多在于理解问题的能力，以及能否组建起能够解决问题的团队，而不是其他任何因素。
We're almost just like our ability to understand the problem and and go go to have teams that can go tackle the problem is the bottleneck more than anything else.
嗯。
Mhmm.
你有没有想过让Ambience成为一个真正的平台，即向第三方开放你的能力，让他们基于你的系统进行开发，而你则成为类似EHR今天的后端层？
Do you ever envision Ambience becoming, like, a true platform in the sense that you open up your capabilities to third parties to then develop on top of your system, and you become kind of that that back end layer that the EHR plays today?
这与我们大多数客户都在积极讨论的问题。
It's an active conversation with most of our our our our customers.
许多学术机构内部都有产品和工程团队，希望开发各种各样的应用。
A lot of the academics have internal product and engineering teams that wanna build all sorts of stuff.
有时候这在我们的路线图上，我们会讨论如何在未来做出正确的共同投资。
Some sometimes that's on road map for us, and we talk about how do we wanna think about making the right shared investments over time.
有很多很好的想法，但可能并不在我们的路线图上。
There's a lot of great ideas that are likely just not on our road map.
因此，让我们更容易让他人在我们之上构建，这简直是超自然的，这不仅适用于我们的客户，也可能扩展到更广泛的生态系统。
And so this ability for us to make it easy for others to build on top is is supernatural, and that could extend to our customers, but it could also extend to the broader ecosystem.
嗯。
Mhmm.
很棒。
Cool.
我最后一个问题是关于你作为一家原生AI公司建设Ambience的内部经验。
And one last question for me that's more internal facing of, like, your experience just building Ambience as an AI native company.
当然，过去几年里，作为一家AI公司的员工，这种体验也发生了很大变化。
And, obviously, again, like, being an employee at an AI company has obviously also evolved quite a bit over the last few years.
比如，如今由于AI工具对员工的可用性，你们现在在哪些方面做了根本性的不同？这些是你两三年前没有做的，但你认为是巨大转折点的事情？
Like, what are some of the things that you're doing, like, fundamentally different today based on the availability of AI tools for your employees that you weren't doing, like, you know, two or three years ago that you think has been a a huge game changer?
我的意思是，如果倒退两年，很多体验对人们来说简直不可思议。
I mean, there's probably a number of of experiences that are wild to folks if you almost, like, rewind back time two years ago.
但在工程方面，单个工程师现在借助OPUS 4.5能完成的工作量真的非常惊人。
But on the engineering side, like, the amount of work that an individual engineer can get done now with OPUS 4.5 is is is quite insane.
我们发现，你只需要真正聪明的思考者，而不再像以前那样需要那么多人来完成大量工作。
I think it what we're finding is that you just need really smart thinkers, and you don't necessarily need as many people anymore to get lots and lots of work done.
是的。
Yeah.
那么，这改变了你们招聘工程师的标准吗？
So has it changed, like, the profile of engineer that you hire?
确实如此。
It does.
嗯。
Mhmm.
确实如此。
It does.
我认为，一般来说，在平台层面，你需要的是那些能够深入思考长期架构决策、预见未来趋势的人。
I I think, generally, on the platform side, you want people who can think really deeply about long term architectural choices and see around corners.
而在产品工程层面，你真正需要的是那种能够深入临床环境、与客户共处、密切配合领域专家，并且极其擅长需求收集的人。
And on the product on the product engineering side, what you're really looking for is the kind of person who can embed deeply in clinical environments, spend time with customers, work closely with subject matter experts, and do really like, a really, really good at requirements gathering.
嗯。
Mhmm.
而这实际上是打造优秀产品和功能的瓶颈。
And that's actually the bottleneck to building great great products and features.
在内部，我们大量使用AI进行研究和上下文共享。
Internally, we use AI a ton for research, from for sharing context.
我们经常思考的一件事是，当新员工入职时，如何将我们内部的决策过程记录下来。
One of the things we oftentimes think about is when a new employee onboards, how do we take almost our internal decision traces Mhmm.
把我们组织做决策的方式变得极其容易让新员工站在巨人的肩膀上。
On how we make decisions as an organization and make it really, really easy for a new employee to be able to stand under the shoulders of giants.
因为如果你加入一家新公司，通常你会完全不知道这家公司是如何运作的。
Because if you join a new company, oftentimes, you're like, you have no idea how this place works.
你根本不知道过去做出这些决策的历史背景。
You have no idea the historical context of decisions that were made.
当你即将做出一个新决策时，你完全不了解情况。
You have no idea if you're about to make a new decision.
你该怎么开始思考什么才是正确的框架呢？
How do you even begin thinking about what the right framework is?
以前是这样的，对。
Before and yeah.
完全正确。
A 100%.
所以我认为，这些用例目前仍处于早期阶段，但我们已经开始组建内部团队，思考如果我们把公司打造成一家AI优先的企业，会是什么样子。
And so I think still early on some of these use cases, but we're we're starting to build internal teams to think about if we were to internally make ourselves an AI first company, what that would look like.
嗯。
Mhmm.
因为我们的观点是，就像今天如果你从零开始构建一个医疗系统，你绝不会采用十年前的方式一样。
Because our thesis is in the same way that you wouldn't if you were building a health system from the ground up today, you would not do it the way you were building ten years ago.
如果你今天在创建一家公司，你不会用两年前的方式去做。
If you're building a company today, you would not do it the way you would be doing it two years ago.
嗯。
Mhmm.
我们还没提到什么，是你希望世界了解关于Ambience的内容？
What what did we not cover that you want the world to know about Ambience?
这是个好问题。
That's a good question.
能够参与解决这个问题，本身就有一种特定的谦逊与尊严。
There there is a specific level of of just humility and and and honor in being able to work on this problem.
我知道你对医疗系统有自己的个人经历。
I know you've got your own personal experiences with the health system.
我也有自己的经历。
I sure have my own.
Ambience的每一位成员加入，都是因为他们自己、亲人经历过医疗系统，或者在医疗系统内工作。
And every person at Ambience comes in because they or a loved one or had an experience with the health system or works inside of the health system.
这是一个长期以来前景黯淡的行业。
It's just one of those industries where the outlook looked bleak for quite some time.
人们一直在谈论这个系统已濒临崩溃。
Folks have been talking about the system being at a breaking point.
我认为，这是第一次出现了希望，那就是。
I think this is the first time where there's hope that, hey.
存在一条以更少资源做更多事情的路径。
There is a pathway to doing more with less.
存在一条让临床医生、护士的工作变得富有成就感的路径。
There is a pathway for the job of being a clinician, being a nurse, to be a fulfilling one.
存在一条让患者体验不再像有时那样混乱和充满绝望的路径。
There is a pathway for the experience of a patient not being as confusing and full of despair as sometimes it is.
我认为，在某种程度上，这使得现在从事医疗行业成为一个特殊的时刻，而过去，由于前景不明朗，我们未必能吸引到最优秀的人才。
And and I think in some ways that makes this a special time to be working in health care, whereas in the past, I think we didn't necessarily attract the best people and the best talent because it was unclear.
如果你希望对医疗行业产生影响，这真的可能吗？
Could you even have an impact if you wanted to on health care?
所以这一刻非常特别，我觉得大家心中都充满感激。
So this moment is very special, and I think there's just a a lot of gratitude.
我们都必须有能力去参与解决这个问题。
Think we all have to be in a position to even be able to contribute to the problem.
完全同意你关于进入医疗领域的卓越人才的观点，而且我认为医疗科技现在有资格与最顶尖的主流科技相提并论。
100% agree with that point about just the exceptional talent that's coming into health care and the fact that I think health tech is now earning its right to be compared to best in class broader tech
是的。
Yeah.
而不是过去那种总显得比其他领域低人一等的奇怪小众领域。
As opposed to this weird niche that just, like, always looks crappier than everything else.
我还会把这个观点延伸到医生这一侧。
And then I would also translate that to the physician side.
我最后想讲一个故事，我记得大概是三年前，你们的一位客户问我，是否可以把我的邮箱地址分享给一位医生。
And, you know, my last story here is just I remember it was probably three years ago when one of your customers asked me if they could share my email address with a doctor.
我当时说，当然可以。
And I was like, sure.
你知道吗，现在怎么样？
What you know, what's going on?
两天后，我收到了那位医生的邮件，他明确询问了Ambience公司某位投资者的联系方式，只为了告诉我：谢谢你投资这家公司，因为我从未在日常工作中体验过这种工具带来的喜悦，它让我决定继续当一名医生。
Two days later, I got an email from that doctor saying specifically asked for a contact information of one of the investors of Ambience because I just wanted you to know, thank you for investing in this company because I've never experienced the type of joy that I've experienced using this tool in my day to day job that has made me now wanna remain a doctor.
你看，这个人原本已经考虑辞职了，毕竟经历了COVID的种种恐怖，再加上你提到的那些层层叠加的压力。
Like, you this is a person who had, you been considering quitting their job, basically, after all the, you know, terror of COVID and, you know, and everything else, you know, compounding on top of that that you're talking about.
所以我完全同意你的观点，过去技术一直是这个行业的负担，就连我早年部署自己的软件时，医生们都会叹气，说：‘又来一个工具？’
So I 100% agree with you that after, nothing but, technology being a burden to this whole industry and even when I was deploying my own software back in the day, doctors would groan and they'd be like, oh, yet another tool?
为什么非要硬塞给我？
Like, why are you stuffing this down my throat?
突然之间，他们开始看到了阳光。
All of a sudden to an era where they can actually see sunshine.
我认为，患者在日常生活中体验到的科技魔法与工作中所接触的技术之间的差距，第一次真正缩小了，哪怕只是一点点，这从根本上改变了他们对技术的看法。
And I think the the delta between the magic of the tools that they're experiencing in their consumer lives and what they do in their work has, for the first time, narrowed, you know, just even a little bit to a point where it's just fundamentally changed the nature of how they view technology.
当我们稍微透露即将发布新产品时，临床医生们的反应简直就像在排队等待苹果的新品发布一样。
It's kind of wild When we even hint that we're about to release a new product, the energy we feel from the clinicians is almost like lining up for the next new Apple product.
如果你以前从未见过这种能量，对我们来说，每次创造出新东西时，那种近乎魔力的感觉真的非常好。
And, like, if you've just never seen that kind of energy before, and it feel I I think for us, it feels great that every time we create something, there's almost like this level of magic Yep.
这种魔力为临床医生营造出一种期待感。
That's created for the clinician that sort of builds up this anticipation.
我们也明白，这伴随着巨大的责任——一旦我们交付产品，就必须真正改变我们所服务的临床医生的生活。
And I think we're we're also we we understand that comes with a bunch of responsibility too, right, which is one then once we do deliver, that the products actually meaningfully change the lives of the of the clinicians we serve.
因此，我们必须持续肩负这份责任，一个季度接一个季度地坚持下去。
And so that's a responsibility on us too to keep doing that quarter after quarter after quarter.
是的。
Yeah.
太棒了。
Amazing.
和你聊这些总是很愉快，Nikhil，恭喜Ambience取得的所有进展。
Well, always a pleasure to talk to you about these things, Nikhil, and congrats on all the progress at Ambience.
谢谢你，Julie，感谢你从一开始就相信我们。
Thank you, Julie, and thank you for believing us since the very beginning.
是的
Yeah.
当然
Absolutely.
感谢您收听本期《Raising Health》。
Thanks for listening to this episode of Raising Health.
如果您喜欢本期节目，请务必点赞、评论、订阅、给我们打分或留下评价，并与您的朋友和家人分享。
If you like this episode, be sure to like, comment, subscribe, leave us a rating or review, and share it with your friends and family.
在X上关注我们@a16z，并在a16z.substack.com订阅我们的Substack频道。
Follow us on x at a sixteen z, and subscribe to our Substack at a16z.substack.com.
再次感谢您的收听，我们下一期节目再见。
Thanks again for listening, and I'll see you in the next episode.
提醒一下，本内容仅作参考用途，不应被视为法律、商业、税务或投资建议，也不应用于评估任何投资或证券，且并非面向任何a16z基金的投资者或潜在投资者。
As a reminder, the content here is for purposes only, should not be taken as legal business, tax, or investment advice, or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any a sixteen z fund.
请注意，a16z及其关联方可能仍持有本播客中讨论的公司的投资。
Please note that a sixteen z and its affiliates may also maintain investments in the companies discussed in this podcast.
如需更多详情，包括我们投资的链接，请访问 a16z.com/discoveries。
For more details, including a link to our investments, please see a 16z.com forward slash disclosures.
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。