A Conversation with Robin Li, Baidu Co-founder & CEO

it has the makings of one of the tech world's greatest comebacks chinese internet old-timer baidu has pulled itself back from the brink of near obscurity and it's now on the cusp of re-emerging as one of the country's most influential companies after losing its lead to a string of hard driving competitors bold bets on autonomous vehicles and ai chips are starting to come good it's smart speakers and cloud businesses are posting double-digit growth but a government crackdown on china's tech champions is casting a long shadow over the sector and that competition well it's only getting more heated now after more than three years the u.s educated founder and ceo of baidu robin lee has agreed to speak exclusively to bloomberg tv about his company's transformation and about his future this is a conversation with baidu ceo robin lee [Music] you're looking at almost record highs now for baidu's share price do you see that as a validation from investors for this strategy change this focus on ai on autonomous vehicles on semiconductors i think our strategy has not changed much over the past two or three years it's that the market takes time to realize that you know autonomous driving or ai has big potential and by doing uh is having a lead on that so two years ago or even one year ago when we spent a lot of money in ai in in autonomous driving it was a negative you know impact to to our stock price because you lose money but by doing that it hurts your eps nowadays it seems that investors sentiment has changed they they all of sudden realize that oh this thing has a big future and by those investment in ai and in intelligent driving will someday pay off and i'm happy that the market has started to pick up this kind of message but for us we've been investing not for one year not for three years not about five years we've been investing in ai for more than 10 years we we probably lost a lot of money by doing this but uh i i think eventually we will be rewarded by by doing this what is the rationale then for this secondary listing in hong kong i think it's quite straightforward by doing space in beijing and we want investors in asia those who are in the same time loan can easily invest in baidu they they can have a better opportunity to grow with us and uh you know there were u.s china tensions for the past few years i think investors also have this kind of concerns although we're not that you know concerned about uh d-listing but i i think uh having another exchange uh for for buy two stock is it's overall good for for the investors and also good for people around the china area to really share our growth and you're really talking about going out on a limb as the business eight years ago ten years ago to invest in these areas where many others weren't how much of an advantage then do you think that gives baidu going forward that gives us a lot of advantage because uh you know this kind of technical problems are very hard it takes time and it takes a lot of money to really get that done or i would even see today we're not done yet we're far from down we will continue to invest for the next 10 20 years and our pace of investment will not slow down we will continue to to add more to our r d budget so that we will maintain this kind of lead ii is a very broad market and it can apply can be applied to many many industries and there are lots of innovation that can be done when you get into those specific industries i'm uh very uh confident and also i'm very determined what are the growth prospects for the broader business looking like heading into the the rest of the year this year when we have an easy comp last year because of kelvin uh our business was affected so this year the the corbins will have a pretty good growth rate but more importantly the new bins the new ai enabled business be at ai cloud or intelligent driving uh it's been growing very fast well you you're right you so double digit growth is what we've seen in terms of your ai powered smart devices and the cloud unit of the business does that growth continue to accelerate yes especially for ai cloud we separately disclosed the the cloud revenue and each quarter you can see that uh it's growing very quickly and uh the growth rate for q4 was higher than q3 and the the growth rate for q3 was higher than q2 so it's quite obvious i believe advertising revenue accounts for about 70 of overall revenues non-advertising revenue is clearly increasing as we just touched on yeah at what point do you see non-advertising revenue overtaking advertising revenue and if so what time frame yeah it will happen i i cannot see you know accurately which year but i think the trend is quite clear uh this is not uh you know i do is not alone if you look at other uh you know large internet companies but most of them uh for for most of them advertising is a you know uh it's a minority of their revenue what kind of acquisitions might you be looking at at this stage we think technology is at the core competence of baidu we would look for technologies that could complement with our own strategy and the technology portfolio but in the meantime if we are short of let's say operating experiences or talents those are all possibilities and would you be looking domestically for those acquisitions primarily or overseas as well i would say primarily uh domestic we we were more uh optimistic about international before but because of the trade tension it's increasingly difficult for us for us to invest in the us you studied in the u.s of course you've built an incredibly successful business here in china where do you think the u.s china relationship is going ultimately my preference is certainly that the those countries can work together and i think by working together a lot of things can be done faster and the world will be better off i like you mentioned i spent quite a few years in the u.s i i know the u.s well and i love my country i i love china i think for baidu's success we obviously are rooted here in china we will benefit from a large very large market and the government policy to encourage innovation but we also benefit from our u.s partners they they supply parts they help us to to grow faster i very much like to stay coupled instead of decouple coming up why robin lee says his autonomous cars could soon be taking on tesla i'm quite confident that we we will be able to roll out the most advanced car in terms of [Music] autonomy [Music] you have a partnership with julie one of china's biggest car makers for autonomous electric vehicles what is the aspiration with that partnership we realized that we have never built cars before and we need help and julie is a natural uh fit because you know they are one of the leading car makers here in china they have a lot of experience they ship millions of cars every year and they have a good capability of managing the supply chain to ensure the quality and cost all that kind of things that we're not good at or when i talked to lee shufu you know the founder of gd we are very much on the same page he wanted baidu to control this car maker company and we want to get you know the expertise from their car making experi experience what's the production time frame how many units are you targeting on car it really uh typically would take about three years to get a car onto the market we'll try to you know get it on the market sooner rather than later it's probably too early to announce a number of units at this time because it's still early but i'm quite confident that we we will be able to roll out the most advanced car in terms of autonomy but before 2024 it sounds like that's right okay and you talked about the competition tesla of course but also xiaopong niyo we're also seeing the likes of foxconn uh huawei evergrand all kind of moving into this space very competitive yeah is there a risk of this being over saturated this market there are uh but uh again i think this is a very large market uh there will be many players in instead of just baidu but what i think most people don't realize that it requires a lot of investment in r d in technology and we started back in 2013. in the future i think everybody believes uh that you know autonomous feature is the most important feature for an electric vehicle right and uh just the autonomy itself requires a lot of investment in technology last year we spent about 20 billion rmb in r d this is more than any other ev company has spent and probably more than all of them combined the companies you you mentioned we are determined to spend a lot of money for the next 10 20 years to make sure we have the best and strongest autonomous platform sounds like robin lee versus elon musk do you think you'll be able to take on tesla uh i think we we are quite different you know we we are an open platform we want everybody to be successful we don't just want to build our own cars so in addition to the venture g do we also supply technology to other oems we want them to use our our autonomous driving capability too we are probably also running a feed around robot taxi we want the consumers to enjoy true self-driving cars sooner rather than later we want to save their right hailing cost there are lots of things we can do and we will partner with people to do that so this is very different from tesla okay i want to get more on the ai chips and focus because that's really interesting you've raised about 230 million us dollars to develop the chips unit of the business what are you hoping to achieve ultimately well it actually started did uh quite a long time ago probably 10 years ago when we were running a very large server farm and we we need to customize uh some of the servers so that it would run faster or cost us less and we found that by designing our own ai accelerator would do that job so we we started this kind of uh chip design quite a long time ago it's a very simple reason we want to save money for our servers uh so uh starting from last year we we had our first quinlan uh chip uh on the market it's not really on the market because uh over 50 percent of the chips was used by our our own server farm but uh it served the purpose right we we didn't have to pay a very high deep price for for third-party you know uh high-end ai chips we could use our own it saves money and runs faster it fits our needs better but then all of a sudden i realized this is not just for baidu it's actually a lot of industry a lot of other companies also want this kind of high-end ai chips and we're sick we're seeing that chips shortage play out globally right now so it seems like a good move it's that the kunlun chip is a seven nano millimeter chip you've said you're going to start mass production this year has that started and where do you see most of the demand coming from uh yeah the the seven nanometer lines quinlan ii is coming up uh this year and it's being uh produced uh by by the the fabs so uh we do expect that the performance will be three times better and uh the internal demand is already there the extra no demand we found that is quite quite strong because uh as you mentioned there's a shortage on high-end chips can you specify which sectors you expect to see most demand from it crosses many industries because you know ai can help in a lot of scenarios in a lot of cases uh and wherever you need high-end chips be it a data center or edge computing for smart transportation for energy sector for education or for finance they all need this kind of high-end chips to do complicated ai have there been any discussions internally about spinning that unit off at some point depending on how you define spinning off we already raised the third uh you know outside money so that operation will be run more independently than before but by doing still the the controlling shareholder we will still uh support them how hard it is and this focus on ai and on chips dovetails with government policy the five-year plan how much of that is by accident by coincidence by design how helpful is it well on the surface i think it's more like by accident as i mentioned we started this like 10 years ago are purely out of our own needs we just want to save money but along the way we gradually realized oh this is much more important than just internal use the the whole country a lot of companies also need this kind of high-end advanced achievements as ai is being adopted by many industries as people continue to realize that a higher computing power can generate better ai results the the you know market demand skyrocketed so we feel lucky uh from this perspective but it's also kind of um more like designed because we are a technology company our our mission is to make the complicated world simpler through technology this is one of the most important technology in the ai era if not us who else can do a better job coming up china's regulators crack down alibaba's been hit tencent is in the spotlight could baidu be next [Music] regulators here are now squarely focused on an anti-monopoly regulatory change reigning in financial technology how do you think the technological regulatory environment is going to change in china how is it going to impact a business like baidu the country has never emphasized innovation more than they do today right innovation is at the center of uh china's uh modernization according to the official uh documents and and uh i i think companies like us we invest you know uh more than fifteen percent of our total revenue every year uh to to to do r d uh i think the government like that and in the meanwhile i think anti-trust is also good for innovation you just cannot imagine you know the number one and number two guy all of a sudden merged and gaining more than 90 percent of market share in the us but that happened quite a few times in china before that's not good for innovation so i think that the antitrust push is kind of justified baidu has been fined recently for what the regulators say are monopolistic practices for not flagging deals to the regulators is there a risk that you're going to have to unwind some of your deals uh that's not very accurate to say because the law probably requires a certain kind of do seek for approval right for that particular deal we did not seek for approval in time and uh it doesn't mean that we had we had monopoly practices the fine was not for that and i also think it's not right for the government to block every deal right certain kind of liquidity is also good for for innovation you just cannot do monopoly things are there steps that you're taking now to adjust to the new regulatory environment here yeah if we do a deal of large scale we will seek for approval in advance the government is also encouraging companies to open up their data how do you think that is going to play out has that started yet that drive it's very much in line with our belief we run a lot of open platforms for autonomous driving it's apollo right for ai in general we have baidu bring which is also open we have pedophile you know the most popular uh chinese deep learning frame work uh those are also open source we want a lot of people to use our platform we're open to everyone in a lot of cases it's free for them it's not only good for the society it's also good for us we get a lot of feedbacks a lot of the focus has been on alibaba and ant and 10 cent do you feel like baidu is better off in some ways that you're slightly sheltered from this well i can't speak for them but like i mentioned we like openness we want to share data with other partners and we think that's good for our technology too another big topic here that involves technology a lot of debate online about this is the 996 the almost more than 70 hour work week that many technology companies and their employees go through and and and is the reality for many people do you think that pushed that drive the amount of hours that people have to put into their work in tech companies here do you think that has to change do you think that needs to be rethought 996. uh it depends if uh the the employees are voluntary i think that's okay especially for those young people if they want to become more successful they need to work harder than than their peers right this is just natural but if from a company perspective if you demand them to do 996 i think that's wrong i think employees have the right to to to work and to rest and to enjoy life it's their choice and at baidu we have this kind of philosophy right it's really up to you if you want to work harder we welcome that if you want to enjoy life that's okay too i want to switch focus really to to you and and your future we've seen some changes at the top of some of the big tech companies here in china in terms of leadership some big names stepping down making way for others is there a succession plan at baidu i always wanted to find someone who can replace me as ceo but in the meantime i do enjoy my current work i like technology i like other to see all the changes happen i i like to see you know our technology can contribute to the society so uh whether i'll be the ceo or not probably it's not that important it's just as long as i'm doing things that i like uh i'll be happy but having said that i increasingly spend a lot of time these days to work with the uh the the the baidu people so that their capability keeps improving especially in in the innovation front uh i think i need to train them i need to inspire them to to become more innovative that that that's uh over the past year or so i spent a lot of time on that okay so you've been looking for a ceo who is your number two though robin i wouldn't rank our you know executives from my number two to to number ten that they're all very talented uh hard-working and has has big potential uh i i i want them all to grow and become better okay and more but you're you're in no rush to retire anytime soon it sounds like uh well uh i i if i have to stay out of the ceo position i will robin lee's ambitious bets on chips and driverless cars has positioned his company to reap the rewards of the next technological revolution but competition is intense u.s china rivalry could upend the sector and china's regulations well they can change on a dime robin lee will need all the skill and luck he can muster to cement baidu's position at the top of china's ai table that was a conversation with baidu ceo robin lee on bloomberg television i'm tom mckenzie

Channel: Bloomberg Markets and Finance

Views: 29,518

Rating: 4.8381114 out of 5

Keywords: Bloomberg

Id: g332_cRrr_M

Length: 24min 5sec (1445 seconds)

Published: Thu Apr 01 2021