第23集 - 忘掉2024吧。2025年将会发生什么

当人工智能真正从人脑汲取精髓时，它会变得更加可持续。我们的思维方式——包括你和我的思考——本就是自然的。量子是宇宙的一部分，我们可以利用它。纳米机器人，虽然看似一种逆向思维，但从机器人技术运作的角度看，它同样属于纳米层面的组成部分。我认为我们必须顺应自然，而非对抗自然。

AI becomes more sustainable when it actually derives its essence from human brain. It's natural how we think or how I and you think. Quantum is a part of the universe so we can utilize that. Nanorobots, yes it's kind of an inversion but it's also again a nano part in terms of how robotics can work also. I think we've got to move with nature rather than against nature.

所以我预见这种转变会在2025年，或许2030年发生——届时人类将真正进化为更智慧的物种，与自然协作而非对抗。这就是整个故事脉络。

So I'd see that change in 'twenty five, maybe in 2030 where humankind actually evolves into a kind of much more intelligent species to work with nature rather than against nature. So all of that story.

另外尼基尔，我认为我们别无选择。大自然已经告诉我们，迄今为止的行为方式再也无法持续。你们可以探索更好的路径。

Also because Nikhil I think there is no alternative here I think. The natural has already told us that the way we have been behaving at this far is longer sustainable. You guys can file a better way.

想要生存，我们就必须顺应自然。

So we want to survive, we need to move with nature.

我们当前对抗自然的阶段，显然已超出人类能力范围。

The phase where we are going against nature, think is definitely beyond us.

深度访谈科技领域最具影响力的创新领袖，独家解读全球主要及新兴生态系统的数据与洞察，追踪全球最前沿的科技与开放式创新动态——以上精彩内容尽在每期新节目中。欢迎收听《开放式创新对话》。

In-depth interviews with the most influential innovation executives of the tech arena. Exclusive data and insights on major and emerging ecosystems all around the world. The hottest news on technology and open innovation at a global level. This and much more in every new episode. Welcome to Open Innovation Talks.

关注桥梁动态，与行业领袖对话。

Mind the bridge, chat with industry leaders.

2024年将成为人工智能之年，尤其是生成式AI之年。去年我们多次讨论OpenAI，它很可能是2023年的头条新闻。当前有一个与此高度相关的重要进展——印度对Janti City的回应项目，名为Project Hindus（印度河项目）。该项目由印度大型企业Tech Mahindra发起，今天我很荣幸邀请到Tech Mahindra全球创客实验室负责人Nihil Malhotra。嗨，Nihil。

2024 will be the year of AI and particularly will be the year of generative AI. Again, last year we have a lot of discussion about OpenAI, again, OpenAI probably has been the headlines of 2023. There is something that is happening related to this specific topic that is super relevant and this is what is called the Indian answer to Janti City, that is what is called Project Hindus. Project Hindus is something that's been launched by a big Indian company that is Tech Mahindra, and today I have the pleasure to have with me Nihil Malhotra, who is a global head, makers lab of Tech Mahindra. Hi, Nihil.

你好，Alberto。最近怎么样？

Hi, Alberto. How are you?

非常好，非常好。很高兴再次邀请到你。时隔一年再次交流，这是个很好的契机来了解发展演变。我们非常期待深入了解这个至关重要的Indus项目，评估其背后的动因，更广泛地理解Tech在开放创新方面的定位，以及Mahindra作为大型企业的角色。或许我们可以从Mahindra集团谈起，接着讨论Tech Mahindra，最后深入探讨这个令我无比期待的项目。

Very good, very good. It's a pleasure to have you back with us. Again, since probably it's been past a year, so I think it's a good touch base point to understand what has been evolution and we are super excited to hear more about this project Indus that I think is very, very important to properly assess and understand what are the reason behind and more broadly to understand what Tech is about too in terms of open innovation and moreover of Mahindra that is a big large civil. So probably maybe we can start from Mahindra then go into Tech Mahindra and then start digging into this project industry that I'm super excited to hear from.

当然，很荣幸回来。我们可以从Mahindra集团说起。作为全球性企业，我们业务多元——按产量计算是全球最大拖拉机制造商，路上常见的XUV400等车型来自我们的汽车部门，还有农业装备板块等众多业务。

Sure, absolutely. And good to be back. We can probably start from Mahindra. I think Mahindra is a global organization, has a lot of businesses. Today we rank some of if you think we are the largest manufacturers of tractors by volume, if you've seen an XUV 400 or if you've seen some of our cars on your roads, know that we have an automotive division, we have farms division, we do loads of business over here.

Tech Mahindra是集团中相对较小的部分，实质上是其IT和技术部门。我担任该公司首席创新官兼全球创客实验室负责人。2024年初我们有几个重点方向：首当其冲是人工智能，

Tech Mahindra is a small part of Mahindra's which is basically and that's the IT arm or the technology arm for Tech Mahindra. From where I belong. And I'm the chief innovation officer and the global head of Maker's Lab in that particular company which is Tech Mahindra. Now I think when we started this year in 2024 there were two or three big bets that were going on. I think one was obviously AI, artificial intelligence.

其次是可持续发展——如何让技术更具可持续性；第三是贯穿全年的主题：开放创新。即如何加强协作？如何让更多初创企业和年轻人参与我们的项目？

The second was sustainability and how do we make whatever technology that we make more sustainable. And third obviously which has been the theme around the year, what do you mean by open innovation? Can you collaborate more? Can you get more startups involved in terms of what you're doing? Can you get more kids involved in terms of what you're doing?

以及如何构建人才与领导力梯队？这不仅是集团的战略愿景，也渗透到我们部门的支柱理念。熟悉Mahindra的人都知道，我们的愿景支柱是'RISE'（崛起）。

And can you create that talent and leadership pipeline? So that's generally been the vision and strategy for not just the group, it's also been for my group. It just percolates down. And these are just pillars of our vision. The pillars of vision are if you know Mahindra, RISE.

崛起是愿景的重要支柱之一。它意味着推动社会积极变革，即如果我们崛起，也必须带动你们一同崛起，因为无论作为供应商还是客户，你们都以某种方式与我打交道。这关乎另类思维，对吧？每天都要以不同方式思考，对解决方案要有不同见解，甚至对遇到的问题也要有不同视角。关键在于推动这种积极性，不仅为了商业，更为了整个社会。

Rise is one of the biggest pillars of vision. It just means driving positive change to the society, which means if we rise, I have to make you rise as well because you deal with me in some way or the other as a vendor or as a customer. It's about alternative thinking, right? Think differently every single day and think differently about solutions, think differently even about a problem that you have. And it's about driving that positivity in terms of what we do not just for businesses but also society at large.

这对我们所有人都至关重要。这些是根植于我们文化体系中的核心理念。2024年伊始，我们的董事总经理安尼什（来自马恒达集团）就确立了必须推动技术发展的愿景。我认为这项业务现在将由技术引领。因此我们的工作也必须以技术为导向。

And that's very, very important for all of us. Those are the ethos that are built within our culture set. The vision for us when 2024 started and our MD Anish of course from Mahindra set up was that we have to drive tech. I think this business is now going to be tech led. So we actually have to be tech led as well in terms of what we do.

从集团角度来看，我们需要以协同方式向前推进。我们不应满足于像科技公司或汽车公司那样在不同领域各自为政，而应思考如何形成推动前进的协同效应。人才与领导力梯队建设是我们最重视的领域之一。我们如何连接初创企业？如何实现开放式创新？

From a group's perspective, there needs to be a kind of synergistic way in terms of how we move forward. We should not be very happy in terms of what we do in different places like a tech company or an auto company, but can we have that synergies in terms of what we drive forward. Talent and leadership pipeline is one of our utmost importance in terms of wish to do. And how do we connect to startups? How do we do open innovation?

协作对我们而言极其重要。'印度河计划'落在我肩上时，正值山姆·沃尔夫曼访印期间——当时有人向他提出挑战：印度能否打造出类似OpenAI的ChatGPT？考虑到OpenAI投入的巨额资金及其领域先驱地位，印度能否实现同等成就？于是我们的前任CEO CP·古尔纳尼在深夜11:30发推回应了这个挑战，他只说了句'接受挑战'。

How do we collaboration is very, very important for us. Project Indus happened to be something on my lap when Sam Wolfman was in India and somebody raised a challenge to him saying, can India produce chat GPT like OpenAI? And the amount of money OpenAI of course has spent and they are the pioneers in the field. Can India do the same and can India replicate the same? So our ex CEO of course, CP Gurnani wrote a tweet at about 11:30PM to his challenge and he said, and he just said challenge accepted.

这就是我们接受挑战的始末。作为首席创新官，这个挑战落到我肩上：我们想要实现什么以及如何实现。本可以采取无数种方案——我可以选用现有模型，可以采用Lama模型，也可以选择ChatGPT模型。全球有数百种模型在运行。

And that's how we accepted that challenge in terms of what we do. And that challenge came to me being the chief innovation officer, what we wish to do and how we want to do it. So there were n number of things I could have done. So I could have used the model, I could have taken the model of Lama, could I have taken the model of ChatGPT. There are hundreds of models which are running around in the world.

但我们决定另辟蹊径，阿尔贝托。以印度为例，这是个极其多元的国家，官方认定的母语就有23种，还有1645种方言或语言变体。

We thought we'd do something very different, Alberto, right? So if you look at India for example, right? So India is a very diverse country. Officially it has got 23 mother tongues that people speak. And also sixteen forty five dialects or ways in which the language is spoken.

深入印度腹地，虽然这23种语言保持不变，但方言或用词差异形成了约19,200种不同方言。其中有些方言仅在该国偏远角落使用，使用者可能只有10万、20万人，有些方言的使用者甚至达到数百万之众。

If you go just deeper down in India, these 23 languages remain the same. But the way the dialect or people speak different words, that is about 19,200 different dialects. Some of these 19,200 different dialects are spoken in small corners of the country. Some of them are spoken by 100,000, 200,000 people. Some of them even spoken by millions and 2,000,000 of people.

我们将此视为一项挑战，决定像Chachi Preeti那样从零开始构建模型，但会选择使用最多人使用的语言——印地语。因此命名为Indus（印度河），这源于印度河文明的名字。这个名字也与印度产生良好共鸣。我们选择印地语作为起点，是因为全球约有6.15亿人使用印地语，这比以英语为母语的人口还要多。

So what we took as a challenge was we said we'll build the model grounds up like Chachi Preeti, but we'll take in the language which is spoken by maximum number of people, which is Hindi. Hence the name Indus because Indus comes from the name of Indus civilization. It's also about a name that resonates well with India. And we started with Hindi because Hindi is about six fifteen million speakers in the world. And that is more than people who consider English as their first language.

因此选择印地语对我们非常合适。但当我们深入研究印地语时，发现它现在有37种不同的变体。印地语包含37到40种方言，所以我们决定首个模型将基于印地语及其方言构建，这将是一个完全以印地语为基础的模型。

So it was very apt for us to take Hindi. But when we went into the deeper roots of Hindi, that Hindi is now spoken by in 37 different ways. So 37 to 40 different dialects come in Hindi. So we thought that our first model would be built using Hindi and its dialects. It would be a completely Hindi based model.

该模型将完全不包含英语词元，所有分词和词汇都纯粹来自印地语。这个模型将开始实现Chargebeepty的所有功能（其功能已广为人知）。令我非常高兴的是，模型发布日期定在12月19日——这天恰巧是我们前CEO的生日，也是他在公司的最后工作日。

It will have no tokens of English. It will be completely tokenized or words would be only of Hindi. And the model will start doing whatever Chargebeepty does and the world knows what Chargebeepty does. I am very happy that and the date set for that release of model was kept on December 19. Happened to be our ex CEO's birthday and also his last day in the organization.

这个时机非常巧妙，无论从哪个流程来看都很完美，对吧？所以我们选择在12月19日这天发布它。

So this is very well timed and whatever process, right? So we kept it on December 19 and we released it.

所以你们是特意选在这天作为给他的礼物。

So you came on sign for giving him a gift.

没错，完全正确。这就是我们送给他的礼物。我们以测试版形式内部发布了这个名为Alberto的模型。如今我们的模型拥有约5.39亿参数和100亿个词元——这里说的100亿词元是指印地语及其方言的100亿个子词。

Yes, absolutely. That was his gift that were given. So we released that model, Alberto, and we released that internally as a beta. Today our model is about five thirty nine million parameters and about 10,000,000,000 tokens. When I say 10,000,000,000 tokens, 10,000,000,000 sub words of Hindi and its dialects.

就我们所完成的规模而言，我们可能是印地语词元的最大来源。而且这些都是纯粹的词元。很多人会拿英语模型套上印地语表层，而我们的模型是完全基于纯印地语构建的。这就是Indus的故事，关于我们如何实现的过程。

We may probably be the largest source of these tokens in Hindi in terms of what we've done. And these are pure tokens, right? So a lot of people what they do is they take an English model, put Hindi on top of it. This is purely built on only Hindi and stuff. That's the story of Indus, right, so in terms of what we did.

那么挑战已被接受，我要说挑战已经完成，现在

So the challenge has been accepted, the challenge has been I will say accomplished and now it's

算是完成了。我想我们还有时间，我认为我们还有两个月来完成它。

sort of accomplished. Think we still have time, I think we still have two months in which we will accomplish it.

好的，我们一定会看看你们一直在做什么，然后最终验证承诺。但我认为这非常有意义，进一步扩展这个问题：印度本身就是一个大陆，一个独立的世界，因此它可能有自己的需求，但我相信也蕴藏着巨大的机遇和潜力。我希望我们和观众都能理解印度的潜力，因为有时我们会忘记这可能是世界最大的支柱之一。

Okay we will definitely take a look at what you've been doing and then see at the end is the promise. But I think it makes a lot of sense and then to broaden up the question is, again, India is a continent, is a world by itself and so probably it has its own needs but also I believe has a gigantic opportunities and gigantic potential. So I'd like us and that our audience to understand the potential of India because I think sometimes we forget this is probably one of the largest pillars of the world.

不，我同意。完全同意你的观点，阿尔。绝对同意。记得我们上次也聊过，印度就像是世界的边缘地带——任何你未曾见过的应用场景，都能在这里找到，对吧？比如语言，世界上没有哪个国家能比印度拥有更多语言。

No, I agree. Completely agree with you, Al. Absolutely. I think that it's kind of a you remember we talked last time as well, it's that fringe of the world where wherever you have not seen a use case you will find it in India, right? So now you talk about languages, there wouldn't be any more languages in the world than it is in one country.

几乎所有的数据都需要被收集、被细化处理等等。因此从技术角度来看，这使得印度在我们工作中显得非常特殊。

So you put almost all of the data, it has to be collected, it has to be grained and fine grained and so on and so forth. So I think that makes India very special in terms of what we do from a technology standpoint.

我们将再次从'印度教徒计划'讨论到Tech Mahindra和马恒达更广泛的创新方法。你能否举例说明通过利用与初创企业的合作推向市场的创新案例？哪些最具潜力的印度初创企业可能成为印度以外公司的解决方案提供商？请用具体案例帮助我们理解应该关注印度的哪些方面。

And we're going have the discussion again from Project Hindus to the larger innovation approach of Tech Mahindra and Mahindra again. You refer to some example of innovation that you are able to put on the market, leveraging the collaboration with startups, what are the most interesting Indian startups that might have a potential to actually be solution providers also for company outside of India, again help us to understand with some examples of what we should look at looking at India.

是的，阿尔贝托，我认为印度当前正涌现大量AI初创企业。我可以举几个例子说明我们的做法——记得上次我们说过世界经济论坛将我们认定为开放创新领域之一对吧？我们倾向于开放创新，不会将其束之高阁。我们希望吸引更多人协作，汇聚更多智慧。关于Indus项目，我可以给你一个非常有趣的案例。

Yes, I think India is blooming with AI startups at this time Alberto, So but I'll give you a few examples of how we if you remember last time we talked we said that the World Economic Forum recognizes us as one of those open innovation areas, right? So we tend to open up the innovation. We don't keep it scuttled and hidden inside. We want more and more people to collaborate, more and more minds to come in in terms of what they want to do. Only for I can give you one very interesting example for Indus.

当Indus项目启动时，这就是其中一个典型案例。Indus的发布对印度而言堪称里程碑事件，人们开始互相庆贺，认为我们具备了挑战实力等等。有个来自二级小镇的年轻人直接给我们CEO写信，说只想参与Indus项目。他当时才十二年级，甚至还没大学毕业。

When Indus was launched, this is one of the examples. When Indus was launched it was a kind of an event for India where people started patting their backs and saying look we can challenge and so on and so forth. A young kid from a very small tier two village writes into our CEO and says I want to work only on Indus. And he is a twelfth standard student. He has not finished his graduation.

我们的前CEO CP把联系方式转交给我。他问什么时候能开始？我说你可以实习形式加入，先跟着团队工作等等。但毕竟你才十二年级，通常我们只接收大学毕业的实习生。

And our CEO, ex CEO, CP passes that contact to me. And he said when do I start? And I said you can start as an intern, you can start and work with us and so on and so forth. But you are a twelfth standard student. So normally we give internship after you've done graduation and so on and so forth.

但这孩子第二天就带着父亲来到我的实验室，斩钉截铁地说这就是他唯一想做的事。他只想证明我们能建成这个模型，愿意提供一切可能的帮助，但要求我们保持工作方式的开放性。这恰好体现了我们在创新层面的开放态度。

But this guy lands up in my place where I have the lab next day takes a place with his father, comes into the lab and says that's the only thing that I want to do. I just want to prove that we can build the model. I will help you in any case possible. But I want this openness in terms of what we do. So that actually gives you an idea of the openness in terms of what we do from an innovation standpoint.

基于此，我们需要多方支持——对我们而言开放式创新意味着国家三大支柱：学术界、企业（包括初创公司）和政府。若这三方能形成铁三角，国家就能持续发展。我们与各方都建立了联系：与高校合作研究模型伦理、消除深度伪造、开展量子机器学习研究。

Now based on this we also needed a lot of help from so for us open innovation means three facets of a country. So it means academia, it means corporates and startups, and it means the government. If these three pillars continue to work as trifecta, the country continues to work and progress. So we tend to have connections with all of them. So some of the work that we are doing with academia is of course in terms of how we make the models more ethical, in terms of how we release, remove some of the deep fakes from the models, in terms of research on quantum machine learning that we're trying to do with academia.

合作对象包括印度理工学院、ICER等院校及地方小型学院，这些正是创新萌芽的温床。与政府方面，无论是2022年疫情期间还是现在，我们都在协助解决AI应用、系统整合等政府面临的难题。同样重要的是初创企业，必须给它们成长空间。

These are the IITs and the ICERs and some of the small colleges of the country where we work in. That's where the innovation is just being incubated, right? So this is just the incubation stage. With the government, a lot of work that we have done in 2022 during COVID times or which we are doing now, right, is basically to help solve some of the problems that the government faces, right, in terms of AI, in terms of integration of solutions and so on and so forth. The other thing which is very important is the startups and we need to give them that space, right?

我们同时担任Atal孵化中心的顾问，这是印度政府发起的扶持计划——所有新创公司都先在AIC孵化器成长。AIC会负责初创企业前三年的发展指导、融资机制和市场策略等。我们与这些机器人初创公司深度协作，它们并非工业机器人方向，而是致力于自动化替代人工——比如为井下作业工人研发自动化设备。

So we are also advisors or part of the Attle Incubation Center, which is a kind of a movement that the Indian government or basically the think tank of India started saying every new startup that comes in actually gets incubated in AIC, which is our incubation centers. The AIC takes care of the first two or three years of how the startup progresses forward, what's the kind of funding mechanism, how should they take the market, so on and so forth. So all of that AIC we actually coordinate and collaborate with, right? So these startups include startups on robotics and these robotics are very different. Are not industrial robotics but they are robotics like for example India also has a lot of workers who work on these main holes and and they have to pick that can we automate their work itself.

这里有自动驾驶初创企业，有研究超级高铁的团队，还有扎根印度次大陆的核心AI服务公司。有个叫'人工大脑'的量子计算初创公司正与我们合作，还有专注量子模拟的Boson Fusai团队。我们在澳大利亚等地也开展了非常前沿的合作项目。

There are autonomous cars based startups which are working in there. Are hyperloop based startups that are working in India. And then there are very core AI based services startups that we connect with in terms of the Indian Subcontinent. Some interesting startups that are coming in India are also a combination of quantum and I can give you one startup called Artificial Brain which is basically working with us. There's a very interesting startup called Boson Fusai which is actually a startup which is working with us in terms of quantum simulation and how do we do this and very interesting work that we're doing in some parts like Australia.

我们正在努力开展工作，可以说我们正尝试在澳大利亚与一些初创企业和学术机构合作进行基于药物的建模研究。同时也在探索如何改进澳大利亚的天气预测技术。这让我们面临一系列关于工作方向的难题。目前来看，绝大多数初创企业都集中在人工智能领域，毕竟这是当下的热门话题。而量子技术几乎在转瞬间也成为了另一个风口。

So we are trying to work and I would say that we are trying to do certain drug based modeling in Australia with some of their startups and academia. We are trying to also look at how can we do weather prediction in Australia better. So there's a whole conundrum in terms of what we do. Largely if you consider it, maximum startups will be found in AI at this point because it's the buzzword. The second you would find quantum has become another buzzword.

量子计算领域有初创企业，量子传感、量子教育领域也有新兴企业正在崛起，它们正与我们展开合作。第三大领域可能集中在机器人技术、增强现实和虚拟现实等方面。这些不同类型的初创企业构成了非常活跃的创业生态。同时我们实验室内部也形成了良好的文化氛围，许多高校学生以实习生身份参与这些项目，这对我们而言非常重要。

You have startups in quantum computing, you have startups in quantum sensing, quantum education that are coming up and they are collaborating with us. And the third is possibly in areas of robotics, AR, VR and so on and so forth. So these are different startups. So we have a very healthy startup culture and we also have a very healthy culture inside the lab where a lot of these college students come in and they participate in these projects as interns. And that's a very important thing for us.

这正是人才梯队和领导力培养的关键渠道。我的实验室随时都提供支持——实际上我们在全球分布着五六个实验室。每年我们为70到100名实习生提供前沿领域的实践机会，包括印度本土和海外项目。这就是我们在开放式创新方面的实践。

And that's where the talent and leadership pipeline really comes through. So at any point in time my lab supports and we don't have one lab, we have about five and six across the world. We support almost about 70 to 100 interns every year in terms of what they're doing on the cutting edge, so India and abroad. So that's what we do in terms of open innovation.

自上次交流以来，有个明显变化就是元宇宙概念。我记得两年前它还高居热门榜首，当时万物皆可元宇宙，而现在人们大多在回避或低调处理。您对元宇宙现状怎么看？您如何看待这个趋势？

One thing that's probably has changed since the time we spoke is metaverse. I remember that it was one on the top list, again, two years ago was everything was metaverse and now people are mostly striking down or hiding. What is your take on metaverse? How do you see that?

作为实践者，我们团队在过去一年半里已为16到20家客户提供了接近生产级的元宇宙应用案例。您说得没错，现在确实在降温，但我认为衰退主要集中在消费者元宇宙领域。从元宇宙视角看，最明显的衰退迹象是人们和公司不再热衷于宣称'我们是元宇宙先行者'这种炫耀性话语。

So look I can say it from the side of a practitioner for somebody and for someone whose team has almost produced production level use cases of metaverse to about the extent of 16 to 20 customers. So 16 to 20 customers in the last one and a half years have been served by us. Now it's on the decline, you're absolutely right, but I think the decline has been largely on the consumer metaverse side. And what we have seen as a decline from metaverse point of view is the bragging rights that people and companies used to say, look, am the first in the metaverse. You can come in and view it.

用户参与度和互动量都大幅下降，这直接导致元宇宙热度减退。但工业元宇宙仍在持续发展——这里我指的是数字孪生技术。

You can come in and do some stuff. That has significantly gone down. And that's why you see a dip in terms of how the metaverse does. But whereas the industrial metaverse still continues to happen. And when I say industrial metaverse, I am talking about digital twins.

人们仍然需要数字孪生作为物理系统的虚拟副本来进行精准监控和管理。无论是未来工厂规划中的地块评估还是设备预测性维护，数字孪生都不可或缺。现在的趋势是，元宇宙概念已转型为'未来工厂'的组成部分，这被业界称为工业元宇宙，正被视为制造业的新革命。

People still need digital twins as a virtual replica of a physical system so that they can monitor and manage it correctly. People still need digital twins to actually find out what's going to happen to a piece of land or a piece of machinery if I am, let's say, talking about factory of the future. So what has happened is that metaverse portion has now come in into something called factory of the future, right? So people are now start talking about that's the new revolution for factories. And that's what we call it as industrial metaverse.

我认为消费者元宇宙可能不会发展到那种程度。因为我觉得那更多是从炫耀的角度出发，更像是打开了一个让人们去感受事物、观察三维世界的门户。然后你会发现消费者元宇宙中出现了一些挑战，对吧？已经有案例表明人们觉得在那个元宇宙中不安全，与陌生人交谈会感到不安，对吧？所以这部分正在衰退。

Consumer metaverse may not grow to that extent is my idea. Because I think that was more in bragging perspective, that was more in terms of the opening of the kind of a portal where people would go and feel things, look at a three d world. And then you see certain challenges that have come up in consumer metaverse, right? So there have been cases where people have said that they don't feel safe in that metaverse, You don't feel safe by talking to strangers in that metaverse, right? So that portion is down.

但我认为工业元宇宙仍在持续运作，在我们所从事的领域继续稳步推进。作为术语的'元宇宙'可能已经降温，但作为工业流程，它依然保持着这种形态。

But I think industrial metaverse continues to work and continues to chug along in terms of what we do, right? So as a terminology metaverse may have gone down, but as a process, as an industrial process it still continues to be this.

是的，同意。当前的热点都集中在AI和生成式AI上，这是现在人们追捧的方向。确实存在一些泡沫，我认为有些热潮是基于趋势的，所以会有起有落。但同样，背后存在一些实质性需求，我认为这些需求会持续存在。

Yeah, agree. Again, currently the hype is everything is on AI and generative AI is currently what people like. Yeah, there are some hypes. I believe that some hypes are based on trends and so some be upwards and downwards. But again, there are some substantial needs behind that I think they're going to stay.

正如你提到的，我相信元宇宙背后的一些需求，比如工业应用、数字孪生等，都是非常坚实的。实际上这方面有很多进化，当我们超越AI的热潮时，这些实质性需求就会显现。让我再次请教你这位未来技术专家的见解：展望2024和2025年，有哪些即将爆发或持续火热而我们目前尚未充分关注的重点领域？

And I believe some of the needs behind the metaverse, as you mentioned, the industrial application, digital twins are super solid needs. Actually, there is a lot of evolution on that side, as well as we go beyond the hype of AI, some very solid needs are gonna say. Again, let me pick your future technologist brain and use it again. If we need to look at 2024 and 2025, what are the main things that are about to happen or they will be hot that we are currently a bit off rather?

我认为首要的是——虽然2024年我们正在经历就不提了——2025年在我看来会出现以下变化：第一，生成式AI的热度会降温到我们今天看到的程度。原因是很快会有很多人和组织意识到，他们花费大量资金完成的工作本可以通过完全不同于Jainai的方式实现。当然有些工作无法替代，但部分工作确实可以。所以这个热潮会消退。

I think first thing and I would not call 2024 because we are living in it, but 2025 we'll see as per me these following changes. Think number one, the hype of generative AI would go down in the sense that we have seen today. And the reason for that is that a lot of people would soon realize and organizations would soon realize that there's a large amount of money that they're spending in terms of doing some work that could have been done in a very different way from Jainai. Of course some work cannot be done but some of the work can be done outside of Jainai. So that hype will go down.

第二，人们很快会意识到当前形态的AI不可持续。还记得两三年前的Alberto吗？当时我们的建议是'减少飞行，降低碳足迹'。而现在AI的碳计算量——或者说资源消耗——已经达到人类乘飞机出行的57倍，对吧？这完全改变了技术影响的视角。

Second, people would soon realize that AI in its current avatar is not sustainable. If you remember Alberto about two or three years back, your and my chart was please don't travel, reduce your carbon footprint by traveling less on planes. Well AI seems to have trumped all of that, right? They are now creating carbon and compute power or let's say resources 57 times more than a human would actually travel on a plane, right? So that's a very different perspective on what tech has changed.

部分AI工程师或研究人员必须转向探索可持续的AI实现方式。什么是可持续的AI？我目前还不确定。但现阶段我认为有两个方向：一是借鉴人脑机制，让它像人脑一样高效运作。

And focus of some of the AI engineers or AI researchers has to go in in terms of finding out a sustainable means of AI. Now what does that sustainable means of AI mean? I am not sure. But for me at this point there are two ways of doing this. One is of course get inspired from the human brain, make it much more easier just like a human brain does.

人脑仅需25瓦就能完成所有工作，而训练一个模型却需要约十亿瓦的电力，对吧？所以这毫无差别。实际上在我们尝试的领域，两者存在天壤之别。第二个可能的变化是，量子技术将在2025年某些生产应用场景中崭露头角。

A human brain does everything in 25 watts. Whereas if I have to train a model it actually takes about one gigawatt of power, right? So there is no difference. There is a huge difference in terms of there is no comparison in terms of what we are trying to do. Number two change that may happen is quantum will start rearing its head in 2025 in terms of some production use cases.

我知道在某些领域可能言之过早，但已有部分基础应用场景正在形成。量子安全和量子优化将在特定应用场景中达到一定水平，更多企业将更积极地公开他们在量子计算领域的尝试。这是第二个变化。生物医学即将迎来重大变革，我认为2025年我们看待疾病、药物乃至人体机制的方式都将发生快速转变。目前医学大致存在两种或三种体系。

I know I may be early for certain cases, but there are certain enabling based use cases already there. Quantum security and quantum optimization to a certain level may come in certain use cases and will become more companies will become more vocal in terms of what they're trying to do in quantum computing. That's the second change. Biomedicine is due for a big change, and I think 2025 would also see a very rapid change in terms of how we look at diseases, how we look at drugs, how we actually look at even some of the mechanisms of how we deal with body, right. So at this point in time there are two formats or maybe three formats of medicine.

对抗疗法主张防患于未然；顺势疗法讲究随波逐流，让身体自然痊愈；而阿育吠陀医学则倡导借助自然之力疗愈。未来世界或将关注这些医学体系的融合，形成新的生物医学范式。我认为医疗机器人或纳米机器人将日益重要。

The allopathic way, stop everything before it occurs. Homeopathic way in which terms of let it go with I'll travel with you as a stream and let me see if the body gets cured. And the Ayurvedic way which says let nature cure you in some format or the other. So some other format of medicine or maybe a combination of these I think the world would start looking at in terms of biomedicine. I think robotics in medicine or nanorobots will actually start gaining prominence.

到2025年，人类基因组测序已完成，我们将探索如何利用这些数据实现个性化医疗。比如定制药物方案，以及如何保障数据安全等。网络安全领域在2025年必将迎来巨变。

I think 2025 would also see that we've already done human genome sequencing. We will also look at how some of that human genome sequencing that we've done can be used to customize it for personal gains, right? So what do I do with my medicine? How do I take care of let's say things and security? Cybersecurity is due for a huge change in 2025.

网络安全将因量子技术发生根本性变革——包括量子防护策略等方方面面。这将是颠覆性的转变。当前城市建设模式非线即圆，要么向外扩张要么垂直发展，这种模式在2025年必将改变，因为大都市的基础设施已不堪重负。

Cybersecurity will see a major change in terms of what we do for quantum, how we actually deal with quantum and so on and so forth. So that is going to be a massive, massive shift in terms of what we do. Construction may seem to the way we build colonies and societies today, it's either a linear way in terms of it's either an expansive way where we go out like a circle or it's like a linear way in terms of what we build in flats and structures like that. That is bound for a very big change in 2025, right? Because we are buckling our infrastructure in major cities.

我们需要全新的解决方案。空中交通、机器人出租车等领域或将出现创业公司的新芽。这些都是2025年可能出现的景象。我对太空探索充满期待，更对技术未来几年的发展潜力抱持极大希望。

I think we need to find a very different mechanism in terms of what we're doing. Aerial modes of communication, robotaxis or aerial taxis, you may see a few green shoots there in terms of startups that we come. These are largely things that you may see in 2025. Very hopeful. I think space travel, I'm hopeful about space, I'm hopeful about what we can do, but I'm very hopeful in terms of what tech can play in the next couple of years.

不，我完全赞同。刚才送女儿回牛津时——她攻读物理专业——我们还讨论到量子密码学的应用。这确实是极具商业价值的创新领域。人工智能虽然常被过度炒作，但确实蕴含实质内容。不过我特别认同您关于可持续性的观点：技术必须更环保且更易用。量子技术方面，我与您意见完全一致。

No, I agree, I agree again. I was driving my daughter back to Oxford, she studied physics and she was telling me about quantum applied to cryptography. And I think that's a big potential revenue of innovation. So I agree with you that obviously AI is always on hype, there is a lot of substance but I think I really agree on the sustainability angle that the need to be becoming more sustainable and also to be perceived in a way that is more usable. Quantum, totally agree with you.

重点领域应该是医药、网络安全、纳米技术，还有交通运输和建筑业。这类行业应用进展往往非常缓慢，但我们至少能看到一些指明方向的可能性。

It should be medicine, cybersecurity, nanotechnologies and then again transportation, construction. Think probably we won't see again because this kind of sector is very very slow to apply but we see something that at least can show the way that we'll be doing.

阿尔贝托，我不确定这种变革是否会在2025年发生。但作为人类，我们长期以来的行为实际上是在对抗自然——我们建房圈地、制造武器弹药等等。或许未来两三年该转变思路，用科技顺应自然，这才是对人类有益的。

I I don't know whether that change will happen in 2025 or not Alberto, but I think for a long period of time as a human race we have actually tried to suppress or move, I call it the anti nature move. So we try and do everything against the nature. So you want to build houses, we want to capture land, we want to create ammunitions and whatever and guns and so on and so forth. I think maybe there is time in the next two, three years to take the essence of technology and move with nature. And that's beneficial for the humankind.

可以明确地说，如今全球所有首都都饱受污染困扰。无论你到哪儿，不是流鼻涕就是喉咙不适，各种病症频发。究其根源，是我们用各种方式压制自然的结果。我们从未真正掌控自然，或许该尝试用科技与自然和谐共处。

Can tell you safely none of the capitals of the world today are devoid of pollution. You land up anywhere, you either get a runny nose or you get your throat choked or you get some other ailments and the world is actually suffering. The reason for that is we have tried to suppress nature in some way or the other. But I don't think we have a handle on nature. I think if we can move with nature using tech.

因此当人工智能真正汲取人脑精髓时，它才更具可持续性。量子是宇宙的组成部分可供利用，纳米机器人虽是逆向思维产物，但也属于微观运作方式。我们必须学会顺应自然而非对抗自然。

So that's why AI becomes more sustainable when it actually derives its essence from human brain. It's natural how we think or how I and you think. Quantum is a part of the universe so we can utilize that. Nanorobots, yes it's kind of an inversion but it's also again a nano part in terms of how robotics can work. So I think we've got to move with nature rather than against nature.

所以我预见2025年或2030年会出现转折，届时人类将进化为更高智慧的物种，学会与自然协作而非对抗。大致就是这样的发展轨迹。

So I'd see that change in '25, maybe in 2030 where humankind actually evolves into a kind of much more intelligent species to work with nature rather than against nature. So sort of that story.

尼基尔，我认为我们别无选择。自然已经警示我们，当前的行为模式不可持续。我们必须找到更好的方式。所以我们

Also because Nikhil I think there is no alternative here I think and the nature also told us that the way we have been behaving at this far is longer sustainable. We need to a better way. So we

要想生存就必须

want to survive we need

顺应自然而行。我们当前这种违背自然的阶段，我认为确实超出了我们的能力范围。所以Niki，非常感谢这次对话，真的非常鼓舞人心，期待2025年再与你交流，看看你的哪些预测正在实现，就像你今天看到我一样。非常感谢你分享的宝贵信息。

to move with nature. This phase where we are going against nature, think is definitely beyond us. So Niki, thank you so much for this conversation, mean, usual was super inspiring and looking forward to to speak with you in 2025 to to check which of your forecasts are in the way as you as you saw you see me today. You so much for the good information. You.

感谢您今天的参与，我们下期《开放创新对话·Mind the Bridge·行业领袖访谈》再见。

Thank you for being with us today and see you in the next episode of Open Innovation Talks, Mind the Bridge, Chat with Industry Leaders.