00:00:02
Speaker 1: Bloomberg Audio Studios, Podcasts, Radio News.

00:00:18
Speaker 2: Hello and welcome to another episode of The Odd Laws podcast.

00:00:21
Speaker 3: I'm Joe Wisenthal and I'm Tracy Alloway.

00:00:23
Speaker 2: So we were in a Hong Kong recently. That was a lot of fun.

00:00:25
Speaker 4: I'd love to be back.

00:00:26
Speaker 3: Yeah, I haven't been back for four years. Not much has changed. Actually, I was kind of surprised and you expected LUs than I expected. But I am really glad we went back because obviously one of the big talking points in markets right now is competition between US versus Chinese AI. And we finally got a chance to talk to a couple of high level executives of Chinese tech companies who are actually making all the big capital allocation decisions when it comes to the AI race.

00:00:55
Speaker 2: Right, it felt like we're in this moment where there's been the I mean, the way I think about it, there's been Chinese internet giants, but they concentrated on China, right, there's been American internet giants that basically had the rest of the world. And whether we're talking about AI or self driving cars, we're going to see the first sort of like real head to head battle on internet companies specifically, and where they're like competing for playing the same game on some of the same markets. And of course we know American companies can use AI models built by China, et cetera, and so there are all kinds of options for people. So it's like really interesting to say, like, Okay, this clash is actually like it's happening.

00:01:31
Speaker 3: It's a good time to talk to a Chinese tech executive for.

00:01:34
Speaker 2: Sure, that's right. So the reason we were back in Hong Kong is because we were at the Bloomberg invest Conference. Though we also through an Odd Lots trivia night while we were in our.

00:01:42
Speaker 3: First non US overseas all Thoughts quiz night.

00:01:45
Speaker 2: Yeah, that was a lot of fun and I'm sure we'll come back and do that again. But we were at the Bloomberg invest conference and so we had the chance to speak with the CFO of by Do, Henry Hoo. So check it out. We truly have the perfect guest we're going to be speaking with by Do CFO Henry. So, Henry, thank you so much for coming on.

00:02:03
Speaker 5: Outlot, thanks for having me.

00:02:05
Speaker 4: And it's a great season, I think Hong Kong and it definitely is great to see in both Joan Tracy thank you.

00:02:11
Speaker 2: Very nice of you to say. So why do we start with this? You know, obviously I feel like half the conversations are probably about AI these days. But within AI, BYD is a full stack player, right You have cloud, you have the application layer, you have your own chips and of course your own model. As the CFO, you might have to think about prioritization, et cetera. Is there one layer of the stack that you feel is a must win for by do when you think about resource allocation? Is there a layer where it's like, okay, this is an area where we have to win.

00:02:45
Speaker 5: Thank you so much.

00:02:46
Speaker 4: And I think you'll probably put the tough question in end, but I think it's probably the most difficult question to start with. So I think the very unique thing today is I think the entire AI has been shifting from infrastructure to applications and from model to agents. I think that's actually the bad job. I think within that is, frankly speaking, right now, is very difficult to say at this moment which parties must have because in my view, the chap is infrastructures. You need to have a great model to bridge the capability. The cloud is a deployment of that capability, and obviously the monitization and all the ROI questions, especially for the people like me, I say, oh, we focus on that is on application layers. So without any of that, this ROI cannot work. So to answer a question, I think the key thing if I have to pick one, is Cloud, because Cloud at this moment is a platform. You cannot only hosting Arnie, which is our own model, but also I can work in a very open to hosting other models. And the my chip which connecting to my cloud platform can also help on inference because right now the play training is important, but eighty percent of the incremental demand today on a token are inference related. I think this part of the full picture is what I want to emphasize. But you know, given a tough question, if I want a big one as a student ABCD, I want big number C, which is my cloud.

00:04:07
Speaker 3: I'm going to ask you a question which I think is going to become standard for financial journalists in the same way we ask about headcount and expansion plans. What's your token budget? Is it bigger than Joe's? I mean the token the token budget for buy does or how do you measure? I'll ask it in a slightly different way. How do you measure what you're gaining from your token spend? How do you measure productivity?

00:04:31
Speaker 5: Yeah?

00:04:31
Speaker 4: Sure, I want to categorize probably two buckets. One, if we consume computing power to reach a higher level of technology standard you know, the ADI, the how would model form and also how harnys can be designed to deliver better results, So the are R and D efforts. However, as also a tech house, we also deliver those know how to our external clients with different verticals. I think if I'm measuring our internal consumption of the token, I really want to like how better and how efficient our technology can be developed. So that's on one side. However, on the other end, what I think the ROI is more relevant is how many real tasks that the open claw and for example our own application called do mate also is a real agency in human digits and other things can do the task. So I think these two different measurements are important in the way that right now there are two things are better than last year when it's a foundational model getting much better. And number two is the framework i e. Open Claw and other things can link up the foundation model capability to the rail world task from chatting on something to doing something and completing on something. I think completion part of the tokens is more important today, and I think consumption internally will actually encourage the people do that. But think about that, even last year or year before, everyone is actually beefing up budgets. I think that budget is all there, but the completion utask is more important.

00:06:03
Speaker 2: But let me just press you on this question a little bit further. So let's say Tracy and I are Let's say we worked for Baydo in the same department, I don't know, some department of yours. Would we have identical token budgets or would you have one way of saying, you know what, Joe, Tracy is actually finding ways to get more value out of AI than you are, So I'm going to increase her budget. Like do you make decisions like that? And do you have measurement techniques to see like this person really should get ten times the token budget of another person because they have figured out how to get a lot of juice from the squeeze, so to speak.

00:06:41
Speaker 4: You know, Joe, given the question asking I seeing pro next time, I give you multiple con Yeah, I think right now, the technology evolved very fast. I think that's the beauty part of AI. So we don't want to constrain by us out by before thinking through something, we just install certain policy by saying, you know, these are employees with defined the token number by the titles or similarities. I don't think that's the way it works. So I think we want to more open and more nimbo in a way that given enough token to the individuals to empower their internal RD efforts. But on the other end, we do have a lot of efforts to make sure the token costs become dramatic coming down. I think the cost is coming out very fast before you even think about getting a policy. Maybe the unit cost is coming down half in like a few weeks. So we need to think about the speed and the cost and output efficiency these three parameters as a package, not only on the number of tokens. That's when I say, on the other end, you feel very interesting facts. So right now you know we are recruiting a lot of younger talents to then even for Baydu, which is you know, twenty plus years listed a public company. So my seeing is the kids actually getting smarter than people expected, so they were not wasted the tokens you'll give to them, so they have a sensible judgment about what are the task they need to prioritize because they're working out agents and the models. The model actually helping them also prioritize autopaths they have. So I think the power of the technology today is it is not only at the tools. So that's actually the key concept on to mention it is not only a tool, it's a mindset, and the mindset become automatic and more intelligent. That if people can work on that well and have a new relationship with agents and with a model, some of the old questions we kind of struggle ourselves will be kind of diminished and less important.

00:08:30
Speaker 3: Okay, so no token maxing at BYDEO. But since you brought up talent, one thing I'm very curious about is we know the competition for the top engineers is so intense right now, and in the US we see these headlines where engineers are treated like sports stars. You know, they're being treated for millions of dollars or whatever. What is Bydeo's pitch to top talent, Like, if you're trying to attract someone to the company, what is it you say, to them that makes them want to work for buy do versus another tech.

00:09:00
Speaker 4: Firm's a great question, So let's bring a different perspective. I think, you know, we are a technology company. Previously, I think the priorities we empower our clients to be more intelligent. We give them more technology tools to help them to remove a move from the you know, the traditional it to the cloud environment, you know, such as that. But right now I think AI, especially for the big corporation like us, also change us as well. We need to think about the cultural change organization change not only as an organization and a company, but also how AI empower ourselves. So it's actually equally important to do something for client versus think about new tools affecting ourselves.

00:09:42
Speaker 5: So Trace is right.

00:09:44
Speaker 4: I think there are a few things we actually make a lot of different thinkings and some of the new initiatives. First of all, we're probably among a few companies in China still very open and even increasing the campus recruiting and the focus on the younger talents and number two recents. We also task the senior people not only look at you know, the current reporting structure, but also in the real mentor relationship with the younger, growing piece of the human capital in the company. But more importantly, I think it is really about giving people more autonomy to work in a company, so more trust and more autonomy and give them more real work. And you know, there's a one concept called one person company, right, so we're very happy to working with one person company because they actually use our AI to very nicely and willing to pay a lot of revenue to our products given the quality. However, within the company, we'll also encourage people to be the one person team so they can actually use the agents to work on a lot of internal tasks. So internally you have listen a little to call the doo Doo in Chinese is a very kind of nicky name, which is actually our internal kind of open clock, similar tools and which actually enhancing people's efficiency. And the true point I think give me more people, more autonomy, more trust and the more room to grow and attracting new talents I think equally all kind of very important to change yourselves. But also you know, reporting lines and organized instructure need to come with it to make sure that people can deliver the results. And the last note, I want to mention the key things that people see the application is important. They can work on a full stack in by DO, which is very unique value in the China tax space.

00:11:20
Speaker 2: You know, it just occurred to me. American companies are kind of becoming more Chinese in the sense that they're doing more vertical integration. Like that's sort of a long history here of sort of the whole thing, and now we see one of this phenomenon is that every American company like they want to even start designing and selling their own chips, their own silicon, which is something that you're doing your own business and you've had it for a while, and I'm trying to wrap my head, like what is the rationale? How much is it about just wanting to be able to control your own fate more and so wanting to control more of the supply chain versus having it that optimally aligns with the model that you're working on, because those are distinct priorities. So what is the real rationale for having custom Seligan?

00:12:10
Speaker 4: Yeah, I think thanks for the tech trend in the past kind of year and two.

00:12:15
Speaker 5: So if you look at the entire.

00:12:18
Speaker 4: Computing power consumed, for example, last year, most of the consumptions actually relating to the training of large foundation model, but right now you know many of them going to the influence and the complete tasks, and if you look at the different stacks, I think right now we are actually fitting to an incremental growing piece of the market which is well defined with a clear boundary, which is not focused on pre training for a very super scale foundation model. However, the influence application is important. So to a point, I think our cheir products appointing of cloud focusing on the inference and application is a unique way of we see the positive network effects. I think that's where the area will wants and also given the issue you mentioned, I think within that defined areas, we I feel pretty confident regarding all the issues you mentioned, and because on the supply and demand side, we can find the good match within the emergent market category, within the inference and application markets.

00:13:31
Speaker 3: So hypothetically you could try to do everything right the full stack, and I guess the capital investment you would need to do that is also hypothetically unlimited at this moment in time. And we hear these crazy numbers in the US about the hyperscalers spending hundreds of billions of dollars this year alone. But when you're looking at the different parts of the business. So you have a very mature internet search business, and then you have everything that you're doing with AI, including in structure, how are you actually allocating capital and then how are you actually I guess, balancing that with returning capital at the same time to shareholders.

00:14:10
Speaker 4: So I called it impossible triangle. So I kind of scratched my head every few months, every few quick weeks, depends on I also see the headline numbers brooke news with other peers as well, So sometimes I make a little bit kind of hesitating to make a statement, but I just want to tell the facts and tell the views. I want to separate them out. So in the recent quarter earnings, we you know, we mentioned we actually solve partially on these impossible triangles. One is our operating profit increase almost doubled on a Q on Q base and number two, our cloud revenue grew about seventy nine percent on a WILDWI basis, which is almost double off the wildwide growth rate for the cloud market in China. And number three is since Q three last year, our operating cash flow has turned positive. So positive operating cash flow, incremental operating profits and the higher growth and the market. However, my capbacks is not seeing kind of double even third multiple times. So I think the way of resolving that is as a CFO or as a management team of a heavy CABACS investor AI tech company right now need to find a way on one hand really drive the growth but also keep the density of the investment into air in a reasonable pacing. However, when you do that, you need to keep a conscious regarding OURI and our in mind to look at the entire cash cycle. For example, every dollar we spend today we probably need to wait for another probably twenty thirty forty months depends on the category to get full cash back. And during that frame, obviously there's a you know, price heights, memories, there's difficulty on IDC centers and a huge spending our service. So my point is as a SFO on every project, you need to look at the entire life cycle, not only at one time, but also the pacing important because the foundation model R and D always taking a few months. Right, So these are the things you need to keep our mind. But my statement today is as by do we want to invest probably in a more responsible manner to the shareholders, but do not diminish our ambitious to investment into AI. Keep the right density is important. But given the results for this quarter, I think we kind of resolve that at least for this quarter. So hopefully you can keep on working on that and maybe, you know, a half year later when we check on this point, we can still keep on the same pattern, you know, high growths, less dollars spend, but better II. I think that's probably the angle we want to achieve.

00:16:27
Speaker 2: Can I just mention something I'm very curious about? You know, I'd say the heads of the American AI labs maybe have varying degrees of AI psychosis. They have a lot of worries that the what they call alignment research, et cetera. Do you work on similar things or do you have the same concerns and do.

00:16:48
Speaker 6: You also have AI side, Like do you like you know, for speaking of like trying to make money, Like do you invest in or how much do you invest in what they would call AI safety or alignment and essentially making sure that the models that you're building don't go rogue and always work on behalf of human flourishing?

00:17:09
Speaker 2: Is that a thing that you allocate capital to?

00:17:11
Speaker 4: Yeah, that's great question. So there's an emerging area of example in this data sanity and all the kind of post training efforts need to work on that. You know, alignment obviously is one of that. But my point is if you look at this new concept of harness, right, it's not only about training and getting model on the leaderboard, but also more importantly to measure the robustness and all the things you mentioned. I think in the context in China tech sector, the engineering has to be and has been a good competitive advantage. So the harness from the data flightwell to the alignment, to the data quality and the labeling. I think the entire evil system has been robust. For if you think about even in the mobile internet work right, so as simple as data labeling to the alignment tracks and post training and SFT, I think this kind of the full chain of the capability in terms of the talents and the pool of resources and the cost of data sanity and all the tracks has been in my view a little bit kind of more efficient in a way that the ecosystem has been in place there, so the cost efficiency has been there. So my view is this is engineering, not theoretical quantum y right, So on that the engineering capability form. The China tech world and industry has been there with the key elements I mentioned right, talents, lower costs, more efficient I think these are the few things I just want to point out actually can help solve the issue. But as I mentioned it, the things are evolved very quickly, right, so you know, we don't worry too much about the issue you mentioned in local market.

00:18:48
Speaker 2: Yeah, so this is interesting. I'm curious. I want to press further on this because the American air I'm very anxious about this, and they publish these model reports and it says things that in the chain of thought, we were able to see that four percent of the time the model was able to identify that it was being tested, and therefore it changed its behavior in response to recognizing that it was tested. And this is a sign of potential misalignment. Are you doing the same sort of research and spending to establish that again, the models work for people and don't have a rogue goal, So.

00:19:26
Speaker 5: Yeah, sure, I think right now if you look at this, right.

00:19:28
Speaker 4: So, we are also part of the open source community, so many of the good model especially publishing recently, also will publish their salts. Makes sense, so we kind of follow the new sauts, but also doing our own pasts as well. So overall, I think people in the open source community today, in my view, is very collegial. So people still want to do a better model from kil model for everyone globally, not really on one country to different places yet.

00:19:52
Speaker 3: So actually, related to this, I'm going to ask something. Maybe it's slightly sensitive, but I think it's very important. So in the US, the AI companies, even as they talk about safety, they're basically self regulating, right Like they choose to put out these reports and judge their own models and things like that. In China, tell me if I'm wrong, but it feels very different. It feels like the government is more hands on when it comes to AI. China has been very explicit about this as an area national security, national strategy. So you're operating in an environment where you're firmly embedded in China's technological and industrial policy. How does that influence the development of your AI models and your broader tech.

00:20:35
Speaker 4: So obviously we're not inflation to commut of public policy, but I definitely happy to share some of our thoughts I have. I think in the world, in the China AI today, we believe we have a great group of very superior talents, not only the engineers, but also people actually design the framework. Right, So that's actually very important because it's not only about algorithm selves, it's about whole system, regarding infrastructure, regarding the data regulation, regarding the model, and the cloud. I think given the past kind of ten twenty years in China, giving this entire infrastructure has been upgraded to a level that is kind of worth leading. I think the policy is supporting to getting the moment we have today.

00:21:15
Speaker 5: It's already proven.

00:21:16
Speaker 4: We have a proven path to leading not only the technology renovation and innovation, but also the way you monitor that into the.

00:21:24
Speaker 5: Stage you already have today. So that's my first point.

00:21:26
Speaker 4: My second point is I think today the technology is growing very fast, and the tracks on the performance and on the data transparency and the rules regarding the data regulation, even without the AI model, even on the cloud age in the past kind of twenty years, for five years, has been getting more robust because if you think about that, in a cloud environment, you have almost similar issues, right who own the data, who use the data, who can access that? But today it's a new tool to actually cover all the things we are doing, so I think it is not a new concept for the policy makers think about. It's a new model, it is a new thing. It is really a new and better tools to utilize and access resource or already building and existing resources, will be building on existing platforms, which has been one hundred percent compliant. But also we have a lot of support, you know, from the policymakers, industry practitioners, academias, they actually all contributing to that. So overall, my feeling is it's a very transparent and open environment, not only China but also globally, and academias and industry practitioners actually contributing quite a lot of the good conversations to this environment. And my feeling is the policy makers through different channels has been very open also listening to the new frontier issues and questions.

00:22:46
Speaker 2: I know this is a business conference and we want to keep things very professional here and not engaging gossip, et cetera. But I have like one sort of I'm just curious about something, which is if the American ai CEOs the most hawkish on the sort of like chip exports on the China stuff is Dario, who used to be a buyer.

00:23:09
Speaker 6: Employee do you ever hear things in the office. Do people ever say like, oh, I remember that guy he was? You know, is there any Dario gossip that people talk about in the office from his stinted bid.

00:23:22
Speaker 4: So that's why I want to put the ball back to a court. I want to show another gossip which probably want to hear. So probably in the past kind of hundred days, right, it's open club become very popular, right, and educated market about how AI is really getting to the real task and the real word. Obviously in China, you know a different way. You have a new way calling you know, not only the claw but other nicky names.

00:23:45
Speaker 5: Right.

00:23:45
Speaker 4: So one day I saw Peter, who is the founder of the community which drives the open clock to be prevailed put in Instagram, yeah, saying you know he actually wanted to work with spy do big because you know open cloys are too. But the two is it's kind of eating up all the capacity. I eat the skills, right, so everyone is contributing to the skills, and the cloud is actually grab all the skills and do the work.

00:24:14
Speaker 1: Right.

00:24:14
Speaker 4: So I think one day we are pretty happy, you know, see Peter drop us a note very in a positive way. Because he kind of on one side noticed that the search is important capability of the skills I eat the skills in the open cloud environment. So actually ask him by do to work with him to beef up the search skills in order for open cloud to do a better work to accessing the real time information. Because today the fundational model is one kind of carve out. Is every few months you train a new model and the model itself in his mindset doesn't have the real time information. For example, the fundational model is that it doesn't capture you know, Joe and Tracy we're talking about today. You need to have a new skills accessing the od slots what is happening in real time. So I think you know Google globally and a Byeo of China, they are the powerhouse for searching real time information. So it has to be linking to the open clock. So I think that's actually one of the things we're pretty much happy to about. So you know, next day we ask our engineers to link up with Speter and we're actually part of this skill marketplace doing pretty okay. And right now I just want to share our foundation model called earning five point one right now is ranked as the globally number one in the text format of the Global ram Arena and the globally number five in the search skill capabilities globally in the RAM arena as well. So I think that's I want to give you another gossip. So but for the previous one, I probably can talk with you after this.

00:25:38
Speaker 5: Open Yeah, I.

00:25:39
Speaker 2: Know you didn't give us any Dario gossip, but implicitly because I know that the open clock guy, you know, it's originally called open claud and then anthropics suit him. And also he's got kind of annoyed because they didn't let the API users get full access. So I think that fellow who created open claw is not the biggest fan of Dario's approach. So I by by giving us answer, at least give us a little drama there. Thank you.

00:26:08
Speaker 3: You mentioned search a number of times already, and data is obviously very important to AI. We spoke with Gray Shao earlier in the week. She writes about AI on her sub stack, and I asked her if China has an edge when it comes to data collection, and she said she thought not really because a lot of the data that's been collected was unstructured, and so it was hard to harness for AI model training and inference purposes. He talked a little bit more about how you did that at BYDO because you've got a lot of data you're using it for Ernie. How is that transition process like actually carried out.

00:26:46
Speaker 4: Yeah, sure, I probably would talk about something that market has not noticed enough, and I will talk about what's a real challenge, right, So it's always two sides of a story. I seem I'm very happy to talk about Google versus what we think about by DO in some certain formats.

00:27:02
Speaker 5: A few things.

00:27:03
Speaker 4: I think the markets, not only the capital markets, but also the industry has kind of on the value a little bit regarding the same components we actually imagine with the same structure. Google is monetizing and have this integrated capacity. So first of all, Google has its own TPU, right, which empowered that cloud. So the GCP growth, the Google cloud is growing faster, which is part of the reason is the TPU. And so by Do we have our own trip department, and you know, based on the public information, we recently did the public filing on the spring of these assets, right, So the trip we have the theme with Google for the foundation model.

00:27:41
Speaker 5: We have our earnings sus mentioned.

00:27:43
Speaker 4: However in the physical AI cord applications or called the work model. So we have our robotaxi called Apollo Goal. Just want to share one number. I think the market sometimes I tell even my friends, it's kind of surprised that each week, including San Francisco and including all the cities asting taxes in US, we MO from Google delivered about five hundred thousand trips per week, and in the last quorder by do appollable in globally twenty seven cities delivered about three hundred and fifty thousand trips, which is only about twenty five percent field than Google. The part of that is not only about robotaxi. It's about how we're using the data empower our own foundations, models, trainings and also do influence but also have a lot of know how regarding the multi modile contents and all the different things and the more important if you look at the traditional search right now on this border also on the market, didn't notice that, you know, still even my friends telling me, oh, Henry, congratulations for that for fiel earnings. But you search it probably still eighty ninety percent of the revenue. But the chooses for this border is declined about forty eight percent, so it's already blow fifty percent.

00:28:49
Speaker 5: So the new growing area.

00:28:50
Speaker 4: For example, the digital humans and also our application software is becoming a powerhouse and growing very fast. So my point on that is if you look had key components or the blocks from the trip cloud, ROBOTAXI for AI, physical air applications and the software. And also one more thing I want to mention is Google linking with the YouTube for the multi model contents, and we actually have our control the subsidy called it iq in China, which is actually over fifty per cent of market share in China for certain long form contents in China as well, So we also have our closed loop of the data flight well as well, probably at a different scale, but I think it's still in the same format and the same model. So my view is yes, I think on website it is right that certain data and elements are in their own kind of pockets. But however, for by do we still have access to those pockets, probably better than other peers. But I want also very honest admit, right, given Joey's my girlfriend, I still own him a little kind of gossip after the session. I want to share my own challenge or it's in China. You have different camps, right, different camps, they kind of don't open up enough how to share the data which is reality known to the market for everyone. But my point is right now this you know, the agents and the foundation model become super smart and it has a great push to move everything to a public cloud. It's actually helped resolving that issue to be accessing more information. Last note I want to share is before AI can the public car penetration in China is about twenty thirty percent versus in US is kind of ninety percent. So that's why your comments I can totally understand because without AI, the gap is like this, but right now it's actually getting closer, but still there's a gap. But my confident coming from this gap will further narrowed because everything will be on cloud environment, everyone's access real time information. But also for by do we have a four stack and each components. Given the Google Pass has been proven to be right and more efficient, we just want to follow the same pattern and access and benefits from the different layer of the data itself.

00:31:14
Speaker 2: So I am very glad that you brought up the robotaxis, because first of all, I just love robo taxis period. They're very fun to ride in.

00:31:22
Speaker 3: You took me on my first I remember, and I was a big.

00:31:25
Speaker 2: I was like, Tracy, you gotta ride in Weimo, you gotta ride in a Weimo. And I think you're convinced.

00:31:28
Speaker 3: They're now we have to write it.

00:31:31
Speaker 2: But here's another besides. I'm also excited about them as a business story for a very specific reason, which is that when I think of like by Do, people call it the Google of China's but you know, separate markets, right, Google is something China. People in the US don't buy in large use by Do as far as I know. But there's gonna be cities now where there's gonna be direct competition between Weimo and Apollo, and including London, I think is going to be the first city where there's going to be head to head beut. And so this is exciting because I feel like, Okay, the tech Internet, the consumer facing Internet giants of the US, the consumer facing Internet giants of China are for the first time going to really be competing in certain identical consumer markets. And so what I'm curious about is like, not who you think is gonna win. I presume you think you're gonna win, But like, what is the dimension upon which the winner will emerge? Will it be the quality of the application, will it be who can produce and secure automobiles in volume? What is the most important dimension that will determine the winner, either globally or in.

00:32:42
Speaker 4: A specific city, So be fugetting that, just you know. Also, the car is one of my hobby too, just you know, Joe probably know I'm actually a risk car driver, so I got my risk my license as well, so every time I actually drive a little bit, so it's quite nice, enjoyable because I think that probably the remaining KAI can use gasoling and drive yourself probably twenty years from now, so before getting there, I think the key world is change.

00:33:07
Speaker 5: The car ownership. Okay, my view is in my own calculation.

00:33:12
Speaker 4: In US as an example, right now, each mile including insurance, gas price, parkings average is about sixty to eighty cents per mile. Is the tipping points between rent a car.

00:33:25
Speaker 5: Versus owner car.

00:33:26
Speaker 4: Okay, so if cheaper than that, people will buy a car, but more expensive people will rent a car. Right so, right now, for global player, I don't want the name name but the average ROBOTAXI costs today because the scale server still very small. It's about you know, one two or two point five dollars per mile, so that's a range about one to two dollars. So my point is this curve, just like agents getting more prevailing, is coming down very fast. So assuming when some point, you know, five years, six years, whatever, ten years, if globally the robotax deliver average price per mile coming down to let's say sixty eighty cents US per mile, then many people were thinking buying a car because today's very thankful, you know, Joe and Tricity. You're probably in Hong Kong. You know, the parking is so expensive, even more expensive than the gas, and the gas in Hong Kong.

00:34:13
Speaker 5: Also very expensive.

00:34:14
Speaker 4: So first of all, the car right now is all Evy drive car Number two, you don't have a buyer parking because you and we can drive in a car can go out and number three, while we're having this forty minutes, my car actually can go up pick up passengers and I can make some money for me.

00:34:28
Speaker 5: Right it's a new agent.

00:34:29
Speaker 4: So that's what I'm saying. It's a physical agents on the road to making money for myself. So my view is ROBOTAXI will change the human behavior getting out in terms of behavior pattern of transportations. That's one thing, right, So we and way More and all other players globally are going to that direction. So that's my vision for the market going forward. However, as you mentioned about the market as a player in the near chime competition, my view is right now the markets do.

00:34:58
Speaker 5: Very early and the ten is very high.

00:35:01
Speaker 4: So in the last quarter we shap our car in London and as you know, you know both way More and Us are starting open the market London. Hopefully next year you will see a car. And you know we have go partner with both Uber and the Lift and also with a Grab in South East countries, so next time you probably call a car from Uber or lift apps, you'll get a buy those car. So I think it's actually helping increasing the services because one interesting notice the human driver probably don't work in the midnight right in certain cities, but right now the car actually can work twenty four hours. So expanding a new market and right now is still a very low percentage of penetration, so still we have a lot of room to go on the other end to your question about the success factor, I seek the two things. When is a technology need to cutting edge and improving, The number.

00:35:47
Speaker 5: Two is operation efficiency.

00:35:48
Speaker 4: Right, so it's actually have a lot of work need to be operational driven. For example, how many locations you pick up passengers to meet more efficient?

00:35:57
Speaker 5: Right?

00:35:57
Speaker 4: The charging stations and all the different networks is actually very important. But given we are working on this business for kind of thirteen years so far, and I can tell you one interesting fact. Globally, there are only two cities right now have over you know, thousand cars in that scale, which is San Francisco and a win city in China, which is Google operating in San Francisco and Apollo Go from Baidu operating one city in China. But I think our kind of partnership with both you know, a Lift and the Uber globally with different cities, I think has been very collegial because the demand is much higher than the supply.

00:36:36
Speaker 3: Joe, I'm going to admit something slightly embarrassing. Actually you already know this, but I never learned to drive, partly because I grew up in Tokyo and then I moved to a bunch of other big cities and so I never needed to, and now I always joke that I'm basically I'm never.

00:36:50
Speaker 5: Going to learn.

00:36:51
Speaker 3: I'm just going to hold out for the self driving cars. So you know, fingers crossed, I hope. So I wanted to ask something about you know, you've mentioned agents a number of times and this seems to be becoming the hot new thing in AI, and I know your CEO has talked about how one of the key metrics for bydo is daily active agents. And my question is how does that actually turn into revenue or return from a cfo's perspective, because I understand with search, you know, you type something in you see the ads advertisers are paying you for that, but I'm very unclear how it works if the agent is actually going out and doing something.

00:37:30
Speaker 5: Yeah.

00:37:30
Speaker 4: So in the mobile internet, where you know everyone look have, for example, the DAO the daily active users because that either fulfilled information, quority demand and individual are primary users for many of the mobile applications in app stores, so that DAO was the primary matrix to measure that. However, in the recent conference, our chairman and founder of by Do, Robin mentioned, as in based on his leadership, that DAA, which is a daily active agents are the new kind of matrix to defining the success of agents. So I kind of very much agree on that. The reason is if you look at the tasks, it's actually spread out into different verticals right so right now it's very difficult to find a new way to identify how much people using especially how much value coming out from using AI. So the agents today is basically can deliver a final task, not only using as a tool for human beings. The agent is smart enoughs can think about that, planning the task and completing the task, and obviously in the way of interacting with human beings, it actually become more smarter and in the way that working with more efficient planning of that. So the truer question overall my thinking is the DAA will measure not only how many people using that, but also how difficult it is. True a question the result driven payment is coming up in the near trail. So I just want to share a few things. For example, right now we have three or four different key products, one of them in China qualify or it's really solving complicated issues for enterprises. It's very similar to our fur goal, which actually in the previous years to do the planning. But right now it's actually coming to the real world. So we install this agent to one of the biggest pots in China and help them deploy and planning for the shipments, the logistics. It's saving the cost of the idle time and improving the revenue of their parts. So the pots actually willing to share a certain profit generation with us. So the key things I'm observing is in the previous meetings. Even I'm a sapphold, but actually I'm tending a lot of you know, the meetings to meet with clients. In the previous meeting without AI, most of the meetings we are talking with is the CTO and the CIO of that company. Because it was a tool, it was a cost center, so they need to find a budget internally, Joe, Unit's not easy, right So they have their CFO and their CEO. But right now, most of the meeting we are having today is with the CEO himself. Because AI right now is not only about by Doo, It's really about helping our clients. So the client has to be a top down level of the initiatives to really drive AI internally. So I think our sales process become relatively more efficient in a way that we get into the number one decision makers. He has a budget and he knows that driving the part efficiency is important for his task, so he's willing to share certain economics with us. So I think the customization is also diminished because right now the agents can be used different pasts. You can repeating that success lower the unit basis of the cost, and the agents become more real and the clients see the value and the profits, so they have a higher willingness to pay and high ability to achieve that payment. So I think these four cycles actually in the AI world is very different with the traditional it.

00:40:55
Speaker 2: This is actually an interesting question because I've seen debate on this within AI about what is revenue look like or what is you know, a sales price look like? Because another thing people talk about is, for example, using an AI agent to say, resolve an insurance claim or something like that, and then the AI provider gets paid on say like you know, the number of successful claims resolved, et cetera. Are you bullish on that basic model where the payment is as you said, Okay, maybe they'll share revenue with you because they can measure that savings. Is that the model that you see across a range of AI applications where it's like a sort of per task or sort of very clearly linked to the efficiency game.

00:41:43
Speaker 4: Yeah, we have another kind of line up business we call the digital employees, which you know, Joe, probably you have the similar experience that if you have one season of the podcast, you're probably very energetic.

00:41:55
Speaker 3: Right.

00:41:56
Speaker 4: If you do that like ten twenty times in two weeks is very exhaustive, right, because you think about you're.

00:42:01
Speaker 2: Seeing you share my views, right, so way is that what you're insinuated we're going to get place because we get exhausted. But the AI didn't want.

00:42:08
Speaker 4: Get it, so so you know. So, So my point is the humans there, the motivation and the knowledge base have their own kind of territory to be frank, but if you look at the conversation, look at the quality of them, know how, if you really tap in a good manner, of course, the digital human actually can deliver efficiency and better execution quality. So one example I just want to share is e commerce is a big industry in China. Yeah, and a lot of live performance is really selling the products. It looks fun, but you know, the come out is the kol cannot work like twenty four hours, right, so and people cannot buying stuff like twenty four hours.

00:42:50
Speaker 5: But if you think.

00:42:51
Speaker 4: About you have a great quality of the human employee can help the merchant owners to sell into different time zones and also can speak Chinese English and for the different parts of audience to have the little jokes from their own countries. They actually can help you know, the ecommas revenue. So that's why we have their own kind of product called digital Employees. We actually help our merchant and the e commace store owners to really pushing on that and selling all the goods. And it actually can perform pretty well because on the you know, the Q and A sessions on the questions is actually reacting to the random users asking a wide range of questions. That knowledge actually is very fluent in a way that for the foundation model, it is the way it works, right, So I think that actually has different user case and we actually monitized by charging for example, the result improvement and all the different things.

00:43:42
Speaker 3: I've tried to buy things for twenty four hours straight before. I think when I first used Cowbow, I think I had Tawboo psychosis or something, and that's how my husband and I ended up with three couches in our apartment. There was about five hundred square feet large.

00:43:56
Speaker 4: So my little suggestion you need to have another agent help you to sell it right product. That's another way are.

00:44:03
Speaker 3: You going to spin out the chips business. We're at Bloomberg invest it's our first live recording of all lots.

00:44:08
Speaker 4: Let's break some news, all right, So it's coming to the to the money part. So I was not to be Frank. I was not even trained by by Finance. I become self by accident. So actually my BOSTH Bachelor and the master training was a trip designer myself. So I think after joining by DO, I definitely realized the Bayous chip product has been really really high quality and really good for the influence, for the for the all the things we talked about helping our DAA to grow as well. So based on the power of information, we already filed the confidential filing for spin off of our trip assets in Hong Kong and we are doing that and processing that process on track, and that is one part of the assets we try to unlock at this moment. But however, as I mentioned the cloud of foundation model, they are all very important. So after this being off, we hopefully can enhancing that ego system. And as you'll know, Chip is not only the hardware. I deeply understand, it is about ecosystems. We need to work pretty well with our customers and the suppliers and the software developers all at one goal, and I think to be a separalistic public company. It will help to achieve that goal not only as a hardware, but also the entire ecosystem as well. And our customer will view our trip products more neutral and independent products that can actually do more testing and more usage on their own cases.

00:45:33
Speaker 2: Yeah, it seems to call reading that you figured out a way that developers can easily port over their Kuda stack over to your stack without much trouble. Henry, thank you so much for coming on Our Loves our first live recording anywhere in Asia. Really appreciate it. Again, it's the perfect guest.

00:45:52
Speaker 5: Yeah again, thanks for having me.

00:45:53
Speaker 4: Thank you, thank you, Tracy, thank you.

00:46:08
Speaker 3: That was our conversation with Henry Hoot, the CFO of Baidu, recorded live at Bloomberg Asia invest I'm Tracy Alloway. You can follow me at Tracy Alloway.

00:46:18
Speaker 2: And I'm Joe Wisenthal. You can follow me at the Stalwart, follow our producers Carmen Rodriguez at Carmen Erman, dash Ol Bennett at Dashbot, Kilbrooks at Kilbrooks and Kevin Lozano at Kevin Lloyd Lozano. And from our Odd Laws content. Go to Bloomberg dot com slash odd Lots for the daily newsletter and all of our episodes, and you can shut about all of these topics twenty four to seven in our discord discord dot gg slash outlines.

00:46:40
Speaker 3: And if you enjoy odd Lots, if you like it when we talk to Chinese tech executives, and please leave us a positive review on your favorite podcast platform. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes absolutely ad free. All you need to do is find the Bloomberg channel on Apple Podcasts and follow the instructions there. Thanks for listening, Yeah,