Why Startups in China Are Going Global with Jing Yang

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
I think this is a coming of age really like naturally, because
when I say coming of age, I meancollectively Chinese
entrepreneurs, yes, a lot of these founders are either my age
or even, you know, a few years younger than me.
I've, I've talked to some of them.
So I, I really can relate to whythey want to build a business
that targets the global market instead of just China, right.

(00:23):
So in the good old days, you canbuild a business and start from
China and then then think about expanding but beyond.
However, that is not possible anymore as long as it's a
consumer facing software productimmediately from day one.
You, you, you have to think about do I beautiful China or do
I beautiful Global X China? Because the examples of TikTok

(00:45):
and Machine and many other smaller companies have already
shown them that you cannot do both.
It's not possible to serve both.Welcome to Analyse Asia, the
premier podcast dedicated to dissecting the powers of
business, technology and media in Asia.
I'm Bernard El and China is definitely the talk of the town.

(01:08):
And what better way with me today is Jin Yang, Asia Bureau
Chief for the information to come back to tell me what is
really going on the pulse of allmost important tech developments
coming out of China and also Asia broadly.
And in this episode, of course, we want to talk about Byte
Dance, pivotal moment, the new wave of AI startups funded by

(01:28):
the Chinese venture capital and also Thailand abroad.
Jiang, it's great to have you back.
So can you tell us what you havebeen up to lately?
And of course, I know lip sync must have made you very busy
since early this year, but whereare you?
Yeah, I mean, obviously DeepSeekwas has been this really
surprising story that in my opinion just made the coverage

(01:50):
of China AI much, much more exciting.
And in addition to that, what wehave been busy, busy covering is
more like the DeepSeek effect, right.
How DeepSeek is changing everything from how you think
about the cost of developing AI to, you know, the, the, how it
affects the US, China chip war, etcetera.

(02:12):
And on top of that, and there's a lot of, you know, I and my
colleagues at The Information have been doing a lot of
interesting stories which we canget into in a little bit, such
as economic robotics, the latestof the China venture capital
landscape by dance, and also newwave of agentic startups,

(02:36):
agentic AI startups funded by Chinese entrepreneurs and
programmers, but targeting a global X China consumer base.
So I have been reading the information since I'm one of
your subscribers. I think I was started using your
deep research tool as well. So to, to really start, because
I, I'm focusing on a couple of stories.

(02:57):
I think between us, we have always been talking about Byte
Dance at the crossroads now. So maybe just to start with the
story of Byte Dance. So in your recent reporting, I
think Byte Dance 2024 results has been showing international
revenue growth, but a slowing profit margin.
I think now it exits matter, right, If I didn't get that
wrongly. So what are the major takeaways

(03:17):
there? Yeah.
So by Jones's revenue on a, you know quarterly basis has been
closing in on Meta and for quitesome time now and in the past
that milestone as of the first half of this year.
However, it's worth noting that on the revenue and excuse me on

(03:41):
profit and profit margin basis by don't still lags far behind
Meta. Meta actually has post has been
posting very strong results in the last few quarters.
And then it is true that you know, I think it's just the the
sort of the natural course of direction of travel.

(04:01):
When a company becomes big enough, the growth rate will
slow down, right. Actually if you look at buy
downs, a company that has hundreds of billions of dollars
in revenues still can grow like you know 2030% of quarterly
revenue basis is, is already quite impressive.
Many companies of that size of revenue either would have slowed

(04:25):
down or grow much less. And then yes, basically the last
two, 2 1/2 years, as I and my colleague Drew Osawa have been
reporting by Don's Financials, the overall theme is that the
international business, majorityof which comes from TikTok has

(04:49):
been contributing a bigger and bigger share of the overall in
overall revenue. However, we understand from
talking to our sources that TikTok overall is still not
profitable. Buy downs as a whole still
generates most of its money from.

(05:11):
By money I mean profit from the domestic business.
Which is doeing basically. Doeing and total right and then
from advertising and e-commerce,live commerce related to
revenue. So these are the biggest
drivers. So the is actually mainly the
revenues and profits mainly driven by the domestic economy

(05:33):
of China. Yes, because that is still the
lion's share of the company's revenue and profit.
It's just that, you know, you see for example, you know, three
years ago maybe the China business was off the top of my
head like more than 80% make up for more than 80% of the

(05:53):
revenue. Now it's like, you know, just
below 80, right? It's a, it's a very slow
transition, but you see sort of that's where the direction of
travel is. And then the other thing that is
worth noting is that as we know and we have reported several
times, the Baidon's founder Johnny Ming really is really

(06:18):
focused on and very hands on with about buy downs, AI
development. And then to that end, buy downs
actually has increased the CapExspending, which also has been
reflected in its earnings since this year.
And, and, and as a matter of fact that the company told
investors that they're, you know, they expect to continue to

(06:41):
be on sort of like a high CapEx trend as well because the
company feels is, is sort of, you know, now is the time to
invest. Seems like Jiang Yi mean is like
mark in existential mode, founder mode and just go all in
on AI. Yeah, I mean, actually there has
been. I don't know about in the US,

(07:02):
but definitely in China, I've seen talks of people comparing
Jiang and Zuckerberg as as sort of like, you know, in China
there's John and then in the US is there Zuckerberg and then
they they compare it for severalreasons.
First is that buy downs is, is quite a similar in terms of
business model, right? It's predominantly a social

(07:26):
media company now, very sort of all in on AI, but then also
because both entrepreneurs are very aggressive in hiring AI
talent. We all know what Meta has been
doing in the last couple of months.
But has Jangming went out and splashed, let's say, like the
same kind of salaries to deep sea research?

(07:48):
And not that we have seen obviously because to be honest,
I mean just on average this, youknow, the pay in China cannot be
comparing to begin with, right? It's about 1/5 right?
So, but if I were to give you 1 recent example, we it's also a
piece of news that my colleaguesinformation broke is that by

(08:10):
Dance successfully poached a long term.
And the veteran, you know, researcher at Google DeepMind,
someone who was in Silicon Valley to go work for by Dance.
And that person we own way now is one of the leaders of by
dances AI development. I see.
So that's where sort of you see the the parallel.

(08:33):
Actually, the revenues and profit parallel also makes it
really interesting because it makes me feel like it's also
like a bit like the Apple iOS, where Apple iOS is holding most
of the profits, but the revenue is smaller than Androids, which
have the most revenues, but actually their profit share is
actually extremely low. It feels that way for me when I

(08:55):
look at bike dance. What you mentioned about bike
dance just now as compared to Matter.
Yeah. But though I would say though
you know buy downs is a is, is aprivately how companies are not
public listed. So he does not have to succumb
to sort of quarterly earnings and the Wall Street pressure.
So, so you probably doesn't really need to like sort of

(09:16):
squeeze so much out of his revenue dollar.
But how long can they stay private, right?
I mean strike while OK strike, they stay private.
Literally in the trillion dollarquestion.
Correct. OK.
But we always talk about the much more bigger question.
I think TikTok is slowly sort ofhelp to reduce Byte Dance major
revenue share, which actually concentrated in China.

(09:38):
So they're actually getting moreinternational revenues.
How significant now is TikTok toByte Dance future, right?
Especially considering all the conversations that were that
were supposed to be some TikTok US view going on, There's a lot
of regulatory pushback and the role of also how the Chinese
government is looking at the deal itself.

(10:00):
Yeah, let me just try to unpack that a little bit on, you know,
it's a. Big subject.
Yeah, when we talk about Tiktok,I, I understand a lot of the
attention tends to be paid to the US business.
And obviously it's not just because it's a very big user
base for Tiktok, but it's also probably the most I generates

(10:23):
the most revenue for Tiktok. As I'm sure we know, you know,
Tiktok has very big user base inSoutheast Asia, but in terms of
the ARPU you can get clearly in the US is much higher.
So it's we're actually recordingthis episode at a very

(10:46):
interesting time. As we know, the US and China
just started the third round of trade talks in Spain yesterday,
September 14th. And what happened just just
before the talk started was thatthe Ministry of Commerce in
China on Friday issued some a statement for the in my memory,

(11:12):
I think for the first time confirmed that it will be
included on the agenda of the talks.
Because if you remember, I thinka while ago, U.S.
Treasury Secretary Scott Benson,I was interviewed on, I believe,
CNBC when he was asked whether TikTok was part of the trade
talks. He gave a really ambivalent

(11:33):
answer and said yes and no. So now that is I find it very
interesting that almost unprompted that the Chinese
Ministry of Commerce would send that a statement.
And obviously the statement alsosays many other things, right?
Like, you know, China wants to ensure the TikTok can continue

(11:55):
to operate in the, in the in theUS and then safeguard the rights
of Chinese companies. The USI mean, it doesn't really
go beyond what the Chinese government has said before on
this subject. But the the fact that they
issued the statement just beforethe talks started to me feels
like China is probably signalling that it is willing to

(12:20):
make a deal. I think if I was to take a step
back, right, We all know what happened in China between 2020
and 22 when there was this brutal crackdown on all the
Internet platforms. I think that tells you that
Chinese leader Xi Jinping, he personally doesn't, for the lack

(12:43):
of a better word, doesn't reallycare about the Internet
platforms. Like he doesn't seem to believe
that Internet platforms can contribute to China's economic
and the technological advancement the way that, you
know, sort of so-called heart attack or deep tech can.
So for a long time, I think people see like TikTok as a sort

(13:08):
of a just a card that China has on on his hands.
It's not something that China doesn't want to give up.
As a matter of fact, I think China is, especially after Trump
made it so clear how important TikTok is for him and his

(13:29):
administration. I mean, we, we, we saw recently
the White House also joined TikTok, right?
I think for China sort of feels like it is probably in the
slight upper hand in, in terms of the TikTok situation.
China knows that it has something that US really wants,
so that it is just waiting to see what can TikTok, what it can

(13:53):
exchange from TikTok. So I think anything, it's just
basically, in my opinion, impossible to predict Tik Tok's
future in the absence of any kind of resolution of this
issue. Yeah, I think we will eventually
get to some resolution of the issue because I think the bigger

(14:17):
conversation now in terms of theChinese government is in the
semiconductors. You can already see some of the
tech giants who used to be in the e-commerce, you know, like
Ali Ali Cloud is now building their own chip, right?
You could already see Huawei is now talking about the five
nanometers and seven nanometers.I think it seems that even the
Chinese tech companies seem to have also pivoted to heart tech

(14:40):
on there. And they are all, I would say to
be quite honest, while we talk about how great Deep Sea is, I
think Quan is a pretty interesting model because Quan
is the model that actually Deep Sea did distill from.
So and I think I think the latest one was actually top of
the generative AI leaderboard. So, but coming back to the

(15:00):
question. In terms of the open source.
That's right, Yeah. Yeah.
So from your perspective now, right, how does now Bike Dance
compare, say with the global peers?
If you think about a Matter and Google in terms of AI
innovation, I think Matter, I think more or less they have
made a big reset. They're going, they hired a
super team. Now they're supposed to help

(15:21):
them turn around. I think Google has has come back
quite a lot. I think they're getting stronger
and stronger and putting probably the likes of Anthropic
and Open AI on the heels. But where?
Where is by Dance from, from your point of view?
Yeah. I mean, by Dance has sort of
gone quiet in the sense that, you know, I remember, I think

(15:45):
like all over 2024, I think by Dance was like the indisputable
leader in China. That was before Quinn, you know,
gaining traction before deep seabecame a phenomenon.
So even till this day, you know,Dobao Chava still ranks as the
number one consumer app in the generative air space.

(16:07):
However, Bajas hasn't said anything at all or done anything
in the open source, you know, camp like it hasn't said or done
or released any of his models oropen sourced any of his models.
He also stopped participating inall the leaderboard ranking.

(16:31):
I don't know if you noticed. Yes, I do.
Yeah, so. Which surprises me because they
show a lot of research papers I think in 2024, but it never was
ever close to like launching something that is as profound as
what Deep Sea has managed to do.Or even some of the other major
large language models like Kimi or Triple for example.

(16:53):
Is it just because maybe they didn't have the team to do it?
Or maybe they back PED? They didn't.
They were caught on the back peddling back foot.
So now they are like trying to chase chase up.
I don't know why they don't participate in the leaderboard
anymore. It definitely is.

(17:13):
It's just sort of adds to the the layer of opaqueness.
Yeah. However, I definitely would not
write off their, you know, wherethey are.
I think. I mean, my understanding is they
have a lot of people working on this and then they are doing

(17:35):
great work. They are just, for whatever
reason, haven't told the world. Or maybe they are working on a
close model that goes. It could be the possibility.
But yeah, I mean, there's no Chinese close models yet.
Yeah, they have always been onlydone close models, right?
Yes, correct. So and then even if you have a

(17:59):
close model, but if you don't participate in the leaderboard,
then still there's very little the outside world can know where
you are. Or maybe they may be the 2026
surprise. I mean, I'm just waiting for the
shooter job, I think. I'm sure a lot of people are.
I'm not going to. I agree with you.
I'm not writing them off. They have produced one of the

(18:19):
best AI applications before ChatGPT, so I'm not going to
write them off. They probably have one of the
most advanced teams. But to your question about the
comparison with Meta and Google,I mean the one thing I would say
is like we reported this as wellearlier this year.
So we know that buy downs in addition to, you know the

(18:40):
Chinese language models that they have, right, which already
have been used to power its various applications.
They have also been working on English language LLM and that
they actually have teams in the US and Singapore that have been
working on this. However, we just haven't.

(19:02):
Again, we just haven't seen anything released from that
team. Neither have we seen evidence of
TikTok using any of Bidens's English language LMS.
So that's adds to another layer of curiosity in my opinion.
So one thing I was surprised by was Byte Dance didn't really

(19:26):
compete in anything that is textbased.
I think it went straight for themultimodal because their
strength is in image and video. They just haven't put something
relatively significant that would change the multimodal
space. It seems that currently Google
is leading with of course recently the Nano Banana image

(19:47):
creation piece or even bit of the VO3, But I haven't really
seen a lot that is coming out from by dance because I actually
naturally expected them to have a very good multimodal model as
well. Yeah, I mean, I mean back when,
you know, on text or image to video generation was still an
excellent thing. People were already predicting

(20:08):
that by dance is going to be theclear leader in this, right.
But then you know to a point which is simply haven't seen.
And is mind blowing yet OK. So, so I think, I think there's
a lot to talk about by that, butI've got, I've got to shift
gears a bit because I've got so much interesting stuff that I
want to ask you. So the and I hardly get a chance
to get you here. So let's talk about venture

(20:29):
capital. So there is a story that you
broke on Hongshan was I was finding very fascinating.
I think Hongshan have 9 billion in capital.
I think it was one of the biggest raise that I think it
was 3-4 years ago when it was announced.
Yeah, yeah. In China.
Too, and they have only deployed1/4 of that 9 billion.
So what's the problem? Yeah.
So, yeah, that story was something that I have been

(20:53):
working on for a long time. So I think the talks in the
China VC land have always been that they haven't, they have
they have trouble deploying all that huge war chest raised in
the summer over 2022, which was also 12 months before the break
up within you know, Sequoia, US,China, Southeast Asia, right.

(21:19):
But obviously, it's very hard toget the actual deployment ratio.
So finally, you know, the star was aligned.
I was able to get all these numbers and I did that story, I
think it was out like 3 weeks ago, right?
And so basically while I was doing that story, I also looked

(21:43):
into sort of the underlying reasons and it was actually a
lot more complicated than I imagined.
So what what you have happened is that you know, in 23
everybody. So basically the, the post COVID
economic rebound in China just never happened.
And then and, and then that sortof a depressed to the stock

(22:05):
market. And then in the same time, you
know, I mean, we're talking about the onshore Asia stock
market, right? But in the same time, at the
same time in the, in the offshore or the OR the OR the OR
the global stock markets valuation has also been
depressed because of that interest rates and war and
whatever, right? If the USIPO market hasn't

(22:29):
really recovered until very recently, right?
So then that basically was the macro reason that soldier made
exits very difficult. When exit become difficult, that
means there will be very few bigticket growth to late stage

(22:51):
deals. Yes, deals that in which you can
write like three $400 million checks.
Those deals basically just stopped existing.
If you think about it, the sort of the golden era right when Ant
Financial raised its last round,the C round in 2018, that was

(23:16):
several billion dollars. That was, at the time, the
single biggest VC fundraising. Well, that includes a lot of
sovereign wealth funds as well. Yeah, yeah, JSA, domestic and
then the the top tier investors in that round, each of them
wrote 500 million. Nowadays in China you couldn't

(23:37):
even find that many, but I was $500 million round U.
Three Asia conference last week and I was told that there was a
200 million seat plus round. I I.
Forgot which which company was it was.
I think it was a robotics company.
It was an Angel round and the person has been rebuilt that
product yet? Oh yeah, if it's in robotics,

(23:58):
yeah, there's there's a lot of controversy.
Correct. Yeah.
So yeah, I mean, I wish I have so much money they can deploy
some of that, right? Well, it also it's not like you
deploy money for the sake of a deployment.
Right. No, no, no, I know that.
But but given that robotics is so hot in China, right and there
are very few, in fact, you can almost say that with almost all

(24:22):
the global VCs moving out of China, Hong China is probably
best suited to capture most of the distance wave of AI Internet
companies. I mean, China is going to most
slightly dominate in robotics. So why not deploy it now rather
than, you know, somebody care. I mean, 200 million Angel run

(24:45):
Angel funded business. I think that's something like if
you do that in the US oh, sorry,Miriam Murati was 2 billion for
Angel round. So, you know, yeah, that's about
right, right. And also Ilia, let's ask 1
billion, yeah. First round, Yeah, that's right.
No, I mean, here's the thing, right?
That's just one example. If you look at the entire

(25:05):
opportunity set, right? I, I think it's, it's, it's
undeniable that there are just not that many, you know, you
know, growth to late stage dealsjust become really scarce.
And then the reason for that is,is first, as I said, you know,
exit become really hard and as it become hard, there are two

(25:27):
reasons for that. I, I just talked about the first
reason being the macro public market valuation.
But the second reason is more China unique.
You know, the Chinese securitiesregulator, CSR, say in the
beginning of 2023 started implementing the new overseas
listing regulation, which made it much, much harder for any

(25:48):
Chinese companies to list overseas.
Yeah, doesn't matter as long as you do the listing outside of
the Asia market. And then there's not just IPO,
it captures follow on as well any, any types of equity
fundraising. So that also has made exits very

(26:11):
difficult. And then then in addition to
that, overall, I'm sure you, we all know that China's economy
just hasn't been doing really particularly well in the last
two years since COVID, right, Since COVID ended.
Then that basically has another a piece of impact on that.

(26:35):
Then then on top of all of this,the, the, the, the break up or
the divorce with secure US also has had a set of impact, right?
For example, now that US China relations have become so tense,
it's just impossible now for anyChina based venture capital

(27:00):
firms to do investments in the USI mean, just 5-6 years ago, a
lot of Chinese VC firms have like teams and even funds
dedicated to, you know, the Bay Area, right?
That just becomes a lot harder, right?
In the Hongshan story I did withmy colleague Drew we we

(27:27):
mentioned is 1 specific case, which is like a very rare deal
that Hongshan invested this year, which is a which is around
funding 8 sleep, which is like asort of smart.
Mattress, yes. Yeah.

(27:48):
The founder of The Information is a fan of that bad choice.
She's always talks about it, more or less.
I they, I think by the way, they, they say they have the
plan to expand to Singapore and China.
I'm dying for that mattress, butmy wife disagrees with me.
I simply haven't had the opportunity.
It's a type of thing, I'm sure. But anyway, so in that deal

(28:11):
which is really not sort of likea direct investment in AI.
Yeah, yeah, right. It's a mattress business.
Exactly. It's like a mattress Topper that
has some sort of smart sensory stuff, right?
And. But is this cooling thing that's
supposed to make the big cool, you can control the temperature
or that? Yeah, yeah.
But it also collects the the thethe the users data like your

(28:36):
temperatures, yeah blood pressure that kind of thing.
So anyway, so even a deal like that, Hongshan had to make sure
that its investments is is is islow enough to not warrant a
board seat so that so that it won't trigger a review by
Sevillas. So it just dawned on me the

(28:57):
Hongshan's 9 billion fund, is ita multi stage fund or is it
something? Like it's a series of funds,
Yeah. So it's a, it's a four funds
seed venture growth and late stage.
I see. So maybe the growth and the late
stage, they may not be deploying, but maybe the seed
and Series A they're still deploying.
Yeah, that's exactly what my reporting showed.

(29:18):
So like they actually for the seed fund is almost 50% deployed
now three years after it's been.Raised.
But can they go back to all the LP's and say, well, maybe it's
better to now just deploy in theseed and a round and then they
can always revise the investmentthesis, right?
Exactly. I mean, as my story also

(29:40):
suggests, you know, some of the Hong Shanxi LP think that they
if they if there's just not, nota good opportunity set to deploy
for you know, gross and late stage deals, maybe you can
consider returning some of the capital just for just out of
goodwill, right. Obviously no GP is obligated to
do that. Or maybe if they are looking at

(30:03):
raising the next seed slash venture fund, they can consider
maybe just using right, like repurposing the undeployed
capital in the gross or late stage fund into the new seed
venture. So I, I think moving forward,
right, just thinking about Hong Shan, right, then I want to so

(30:27):
flip to the startups, right? So now I think you've got things
like you where Jan Spar, Lofa, they're all targeting global
users. You also have banners who have
now moved to Singapore. And I don't know to the FT
recently published this article that they seems to be they incur
the wrath of the Chinese users and also incur the wrath of the

(30:47):
US venture capital. So I think there's still a shift
of most of these companies are are already going for global now
rather than just a local domestic market.
But I mean, if for any observer like myself looking into China,
when I look at the start outs, the competition within that
market actually makes them very,very resilient.

(31:08):
When once they come out of China, they become actually a
very big force. I mean, to be quite honest, even
in Southeast Asia, I see Chinesefounders moving around.
I usually just feel that, you know, it's not the Southeast
Asia founders I'm all worried about.
It's the 996 Chinese founder that I'm that is really on the
top of my mind. Yeah.
How, how, how do you think? How do you how do you look at

(31:29):
them? Yeah.
I mean, I think this is a sort of a coming of age really like
naturally, because when I say coming of age, I mean
collectively Chinese entrepreneurs, yes.
And if I think a lot of these founders are either my age or
even, you know, a few years younger than me, I've talked to

(31:50):
some of them. And so I really can relate to
why they want to build a business that, you know, that
targets the global market instead of just China, right?
So in the good old days, you canbuild a business and start from
China and then then think about expanding beyond.
However, that is not possible anymore.

(32:11):
It's as long as it is a softwarecompany that is consumer facing,
as long as it's a consumer facing software product
immediately from day one, you, you, you have to think about do
I beautiful China or do I beautiful Global X China?
Because the examples of TikTok and machine and many other

(32:31):
smaller companies have already shown them that you cannot do
both. It's not possible to serve both.
But by that's they both right, Imean, there's no.
But they they built DOI for China and then just try.
For the rest of the world. Yeah, exactly.
So they cannot have one homogeneous like product for

(32:54):
overall right because of a censorship and data regulation.
Whereas Xian and Timo, they wellTimo has Pintodo before and then
have a separate product dedicated to the US market.
OK, so Shen is probably The onlyexception that goes global.
Yeah, from day one, like becausethey by the time that, you know,

(33:14):
they their business, they figureout their business model that
the e-commerce marketing challenge is too competitive.
Do you think that like for example, the menace situation
was that really the Chinese market is so competitive because
the. No, I don't think that's the
case. So what I was going to say is
that if you are a consumer facing so let's you know, maybe
take e-commerce out of the picture because it's not really

(33:37):
it's more of like you are providing services, right?
Like it's, it's, it's not the same as the content, right?
But if you look at all the othercompanies just mentioned, they
do touch on some kind of content, right?
That because when you provide software to your users to help

(34:01):
you build whatever kind of agentic tasks that the user want
to do definitely involves generating content right?
Now, once he does that, then it you are faced with, you know,
the, the censorship problem in China, right?
That's just like at the very like operational level, right?

(34:23):
At a higher level, then you alsohave all these problems with,
you know, how do you want to be perceived?
So basically the mindset of a lot of these founders is that
they grow up and, and, and, and were also trained in an era when
China was like fastly integrating with the world.

(34:44):
So then they basically they're upbringing.
When I say upbringing, it's not just, you know, education, but
as well as how they were trainedin their career.
Their entire academic and professional operating is very
globalized. If you think about the the the

(35:04):
the 30 and a 40 year old Chinese, the average is 30 and
40 year old Chinese in their knowledge of America of the West
is definitely a lot better than,you know, their countries of
China. Reminds me of a very interesting
quote, but I forgot who said it.Is that the Chinese of the 30s

(35:26):
forties that you mentioned understands America very well.
But if you do the reverse, they don't know anything about China
or they see like some black box as well.
And I think that's where they the, the different versions of,
you know, both sides. Why there's so much differences?
Because the the US doesn't understand China the same way.

(35:50):
Yeah, yeah. I mean, in a way, I mean the way
I like to look, look at is, is, is, is, is a way, in a way it
reflects sort of America's superpower as a softer power.
It either reflects sort of, you know, the, the, the, the, the
global, the, the global order that was, you know, set since
the end of World War 2, right. But those are the fundamental

(36:14):
factors by going back to like these founders, I think they
harbor this vision like I want to build something like if I
because if you are a builder, you want your the thing you are
building to be really successfuland how to measure that success.
By the whole world. Exactly right, especially after

(36:35):
they saw TikTok, even though, you know, TikTok got a leg up
from the the acquisition of Musicali still right?
I mean, even Musicali was built by Chinese programmers, right?
So and then on top of that, as you know, you know, China has,
you know, a very deep bench of very small programmers and

(36:59):
engineers through, you know, decades of education emphasize
on STEM and also through, you know, so many Chinese students
studying overseas. So naturally, you know, you
know, you have, you know, the desire to compete at a global

(37:21):
level and the ability to do so then that hence we now see, you
know, this sort of wave of AI agent, agent AI startups that
are trying to to do this. So I think one thing I just want
to go before we move into another segment on, on robotics

(37:42):
is I know we covered UAI a lot. So it's been uniquely
positioning itself in the app creation space, right?
Specifically with the concept ofbyte coding.
So is this going to be how you think Chinese startups going
global, similar to what Cap Cut did in video editing?

(38:02):
That's definitely what the founder told me that he wanted
to do, right. And then he was also, he also
worked at the cap cut team when he was at buy downs.
But whether that can succeed, it's too soon to tell.
The company is still at a very early stage.
I don't know. Have you actually tried to?
Play with ourselves. I'm still navigating.

(38:26):
I find that the the Chinese generative AI applications tend
to be more holistic. That means they try to put 2-3
different point solutions and try to give you a slow solution.
Sometimes it works and sometimesit doesn't work because you
because one of the things I think most generative AI

(38:46):
solutions to me are or I call point solutions.
That means it does one thing anddoes one thing really well.
And then, but I didn't want thatjust one thing.
I want this to touch this thing that I do.
I want you to like like give me an example like gamma is a, is a
presentation generator. Right now I have to go to
ChatGPT, make a few prompts and then cut and paste the
information and then do the generation.

(39:08):
What I would love to do is finish finish the the the
content generation in tragic PT corrected everything.
Then I said, can you send this to gamma and generate me the
presentation right. And that hasn't happened.
I have to still do the cut and paste, which is the frustrating
on it. Whereas I see like Kimi, when I
tried it myself, I realized, OK,they're trying to combine these

(39:31):
two things, but maybe it's it's the way how we are so used to in
the rest of the world. We're so used to the US Internet
ecosystem. When we use the Chinese apps,
despite I have one single folderin my phone, it's called Chinese
China. And every day I would just take
a look at try every UI to just to make sure I understand how

(39:54):
the Chinese is thinking about it.
I still don't, I still find it. It's like, you know, the
Galapagos island. It's like this Galapagos island
with this sort of apps with the rest of the apps that I'm using.
That's the that's the feeling I'm getting from, from from
using you where I mean Kimi or all the now I recently I'm also

(40:16):
starting to try some of the Chinese LMS as well.
It gives me that sense that theyare very holistic.
But somehow, maybe it's just theusage, it still feels very
Chinese to me and not the way how the US apps are done.
Yeah, yeah, I see what you mean.And also I think the even the

(40:41):
user interface, the UI design and a lot of the things can also
be a bit different. I wouldn't say this, right.
I think all the agenda AI start-ups targeting consumers,
not just in China, right, not just those funded by Chinese
founders, but all over, right. I think they're all under or

(41:04):
going to be under a lot of threat by the bigger companies.
Yeah, because now they are, theyall realized they have to move
on to this part as well. I mean, even in vibe coding you
see, you know anis anisphere, which is the OR cursor, right

(41:26):
cursor built a lot of its basically built its AI agents on
top of cloud. And then now I thought big is
also trying to do like coming upwith web coding solutions based
on cloud. So it's just getting really
cutthroat. It is because it's like what Sam
Orman say, right? If you are building an AI

(41:48):
company, you what is good when, when you look forward to the
next 12 months, if a better large language model comes out
and whatever you're building benefits from there, you should
be happy. But in another 12 months,
whether it's open AI or entropicclone, what you are doing, you
will be feeling shitty, which isprecisely what's happening to

(42:11):
Cursor right now. I already have people dropping
out. Cursor even including my own
team has now completely fully converted to cloud code.
Yeah, so that is telling me quite a lot about.
Yeah, it's like once the the bigcompanies that decide identified

(42:31):
spot this area and that this is the area we now want to get
into, then you will be crushed very quickly, right.
I mean, this is exactly what happened then with the
Wingsource. Maybe, maybe it's the US
companies behaving, learning to be behaving like Chinese
companies. Maybe they don't admit it,
right? But I heard recently, apparently
according to some VCs is like based on I think you know,

(42:53):
R.E.M. Data they, they think that the
the US San Francisco startups are working 996 because a lot of
people were expansing on Saturdays for food in their
offices. And then there was this new wave
about, you know, US founders talking about going 9962.
So maybe, maybe it's the other way around.

(43:14):
The reverse culture shift had actually started, but we just
don't see it. But it's actually beginning.
Well, but I do think, I mean, I don't know Silicon Valley that
well to say. Yeah, but you have my.
Impression has been that they also work very hard when needed,
no? No, they work very hard.
But I think there is this new phenomenon.

(43:36):
I think it started with a coupleof years when Mike already said,
look, you guys in Bay Area askedfor what Wellness spa, you know,
perks for whatever, go to China,learn from the 99 states.
And then it just becomes like, and obviously.
And then when you start having Elon saying you've got to be
hardcore, yeah. Oh, yeah, yeah, yeah, yeah,

(43:57):
yeah, right. He was doing it like, yeah,
there's like so everybody and then suddenly, you know, when
Google is showing off all the perks now and even matters also
cutting off the perks. It's like it's almost it's kind
of a shift. Yeah.
But I think part of it is reallyinduced by the Chinese like
companies. Probably.
Yeah, it is like 8. Come on, the Chinese guys are

(44:18):
kicking out as we have to go 996, which I and I think even
the part where the big companiesstarting to challenge a small
company actually starting to feel it's a little bit like how
Chinese companies going after the small startups at the same
time. Yeah.
I mean, however though I, I, I, I will say I think in, in, in,

(44:39):
in previous like the pre, the recent AI wave, right.
I think that's has always been the way like the ecosystem has.
I mean the capitalism and venture capital, the nature of
what we have has always been structured that way, right.
If you can't out compete your competitors you better off being
acquired. True, but you know.

(44:59):
It's a bit like the. Incumbents.
It's it's a bit like an old friend of mine who is your
former predecessor always say Amazon is like a Chinese
company. It's just that they don't know
it yet. OK, I I want to get to the last
part, which actually is the it'sthe part I'm most excited about
because I really want to have a robot.
So one of the things is about the humanoid robots and also the

(45:21):
growing US China competition in physical AII think the question
is why does it? Because I think a lot of robots
are still single function. Why is this such a big area now?
It's becoming like, you know, both sides are trying to build
their own humanoid robots. When I was talking to people in
this space, what they told me isthat, you know, the the arrival

(45:42):
of on chargeability in late 2022made out people in the field in
researching physical area and robotics realize that this is
now the moment. This is now the sort of the
magical wand that can really, you know, take us, you know,

(46:05):
forward just as you know, when deep learning revived the
dominant research of artificial intelligence in the 60s, right.
So then what we're seeing now really is a result of of, you
know, people excitement and thenthe advancement to the

(46:25):
researchers have made in the last two years, since 2 1/2
years since, right. And I just, I think it just so
happens that US and China both have, you know, the researchers
in physical AI, both have, you know, identified this area.
And then, you know, both sort ofsaw the opportunity that that,
that that came with generative AI, right?

(46:50):
And then it's definitely, I think one of the most
controversial areas. When I say controversial, I mean
you know, is this thing even real or is just another giant
bubble that is waiting to burst?Yeah, because one thing that I
always never understand is they always keep talking about human

(47:12):
or robots. But maybe for some tasks you
don't need, humans are actually not very well formed of the
task. Maybe some other particular form
factor might be better. But it seems to be trending
towards having humanoid boxing, etcetera.
Like what they saw in the recentthe World of AI in Shanghai.
The AIC conference, which is nowlike everyone went there is

(47:34):
like, wow, you know, now there'sa boxing robot.
Yeah. Yeah, no, but in the US too,
right? You see some of the start-ups
that demonstrated, you know, backflip, right.
But how do you commercialize a, a robot that does backflip?
What is the the use case for that, right?
People also question that, right?

(47:56):
So here is in terms of the form factor, I'm just, you know,
going to tell you what people inthe industry tell me, right?
So there are two reasons they tell me why a humanoid has a
compelling case. First is that you know our world
like human being. We build our world based on our
form factor as human, so that a human or robotic, if it works,

(48:22):
would would maximize the environment that we already have
built. Because then you basically have
a robot roaming around in the form factor that is already
optimized for human. So that is sort of the practical
reason, right? And then there is also this
romantic or philosophical reasonI was told is that, you know,

(48:44):
it's just this complex that men want to be God, to play God,
what men want to recreate itself.
On, well, the same level. Form factor, form factor and
intelligence. Yes, OK, But then there is
something like in Nate Herbert'sJune and even in Asimov's

(49:06):
Foundation, the robots eventually got wiped out because
of that problem that they becamesentient so.
Yeah, I mean, obviously that's why I I I saw I say this is the
philosophical reason, right? Yeah, on.
However, going back to what he was saying, actually if you look
at on some of the companies I talked to in China, actually a

(49:28):
lot of them are pretty practical.
So for example, if you look at Tesla Optimus, right, it has the
dexterous hands, like, you know,55 fingers, all the motors.
And 400 actuators, which Elon Musk mentioned in the recent All
in Summit, happened in 70. Five.
Yeah. And then and then also two legs

(49:52):
and then the legs also have the,the the food also have all the
toes, right? So that's like sort of, you
know, following the human form factor to the very last minute
detail, right. However, you can see some other
companies in China there will belike, you know, the technology
right now is just not there for a fully motorized dexterous

(50:13):
hands or two legs. But then for the entire humanoid
sector to have a future, it needs to have some kind of a
commercial use case first. Then why don't we build, for
example, a gripper like it's essentially like this, right?

(50:33):
And also a wheelbase instead of having legs.
It's for fulfillment packing exactly exactly for ecology.
And then while we do not well, well, we will not this company
say you know well, we will not stop working on Dexter's hands
and then you know to food solution on we want to

(50:56):
commercialize, you know the technology we know we know now
that works right. So that's the sort of a more
practical approach. Which one is better we don't
know. I think it's too soon to tell.
But let's say like the top threecompanies, UB tech unit 3
robotics, right? I will also classify DJI as one

(51:19):
of them as well. Are there any like next wave of
robotic companies coming up or is still the this big tree?
There is a lot that are I mean, I don't want to give any names
because, you know, first I don'twant I don't, I haven't talked
to every single one of them. And even if I have it's, you

(51:42):
know, things move so fast, right?
All I can tell you is that yes, there is a new wave of companies
from China and there's also a new wave of companies that
supply specific parts. For example, Tesla Optimus
actually has to source parts forthe tech store scans from China,

(52:03):
right yes. And then this actually gets
here's what I really want to talk about in terms of the US
China competition in human oil robotics first, I think both in
terms of technology, I think both of Usus and China are sort
of on the same starting point, right.
No one now has like a a clear upper hand in terms of solving

(52:25):
the biggest problems and the biggest bottlenecks in this
field. Right.
But manufacturing, still China is leading.
So here is where it gets interesting.
A lot of people think that Chinamay already have a lead or is
going to pull a lead in the space because of the
manufacturing prowess that Chinahas amassed.
While that is not untrue, that actually is not the reason why

(52:49):
China may eventually have to pull the lead in the space.
Because let's say all the factories in China, right, the
supply chain, if they managed toproduce better hands and
whatever parts they will not sayno to say any America company

(53:11):
that come say I want to order 10,000 hands from you, right,
right. So that that is why you know,
the the best smartphone still is, you know.
Made in China. Yeah, it's made in China, but
it's an American company, right?So that is not where, you know,

(53:31):
China, you know, before, before iPhone, you know, there was no
smartphone brands in China, right.
So actually where China the the people in the humanoid industry
in China. Tell me where China's real
advantage actually is. Is is is from or the?
It's actually one to 100. No, it's the test beds.

(53:53):
Correct. It's a test beds and then what
happens is that once there's a ground breaking design from the
US, they can evolve it to the next level with the
manufacturing techniques and theand taking it to scale.
It's just a little bit like deepsea, right?
I mean, deep sea is built on thefact that they saw what was
happening with the US. It's kind of, I don't know, is

(54:13):
it innovation? Yes, it's a kind of different
kind of innovation, but it's more continuous innovation.
Or did I catch? Did I miss what you're saying?
No, I, I, yeah, let me just clarify a little bit.
Let's just say right there is this, you know, group of
startups in China and a group ofstartups in the US, right.
So training this physical AI device is extremely hard because

(54:36):
you first need to gather the data and you need to work on a
really complicated model. And then you need to gather the
data. And I don't let's not go to the
data. The data is yeah, we assume
that. But then after but then you have
to this like rounds and rounds of iteration.
You have to do so, for example, you have this robot and then you

(54:56):
need to actually deploy this role of robots into like, say a
factory floor and see how works and then gather, you know, the,
the, the, on a trial and error basis, the things that worked
and have not worked and then fine tune that.
But then if you look at it in the USI mean, what kind of
advanced manufacturing does the US still have these days beyond

(55:20):
Tesla, but in China, right? All these, you know, 'cause the,
the, the, the first batch of, ofusers of humanoid is actually,
you know, EV makers, car makers,right.
And with China's advanced manufacturing, it does not only
give China age in the manufacturing in the supply

(55:41):
chain perspective, it gives you a huge advantage in being this
giant like test bed this experiment, right?
Yeah, where your robots can be deployed.
And eventually someone would be the one that innovates the
iPhone in China, essentially therobot in China.
That's what you're saying. Yeah, yeah.
So because you know, even if youbuild like say other than with

(56:06):
the exception of Tesla Optimus, right, the other companies, the
other robotic companies in the US, they will, it will be very
hard for them to find like factory floors in the US where
they can actually deploy a roboton a trial basis and then gather
the data and then find to the results based on that and then
do it like 100 times. OK, right.
Yeah, OK. So because we are still running,

(56:27):
running a bit out of time. So I'm going to have two quick
questions and then you just gaveme what's on your mind.
OK, What's the one thing people ask you about the Chinese
startup ecosystem that you wish that they would ask you more?
What's the one question that youwish that they would ask you
more, but they don't? I want people to ask me or have
a conversation with me about redefining what is a Chinese

(56:51):
startup these days. OK.
What's the Chinese startup there?
I think the definition has to bemuch have, has to be broaden a
lot more now by for example Manus again, right?
Is it a Chinese startup or not? You can say yes or no startup
built by Chinese. But they started from China
right after they became famous, then moved to Singapore, right?

(57:14):
And then you also have like say,James Spock, which is started
they, they, they, they were founded in California, but found
they were incorporated in California, but founded by, you
know, Chinese entrepreneurs. Is that Chinese or not?
I will say yes and no, right. It's just these days, you know,

(57:34):
in sort of the Western contest, the name Chinese, the term
Chinese has become somewhat a toxic label.
So a lot of Chinese entrepreneureven want to share that image of
themselves. Yeah, yeah, yeah.
They they do. Their founders, even in this
region, were not there to take Chinese money because they want
to go into the US. Market, yeah, Yeah.
And then that is all driven by geopolitics.
But you know, I don't think I don't need to tell you if you

(57:56):
look at I think at least half ofthe researchers that meta
approaches from open air are Chinese.
Yes, right. We have a saying in China these
days now that the that the AI race is now between the Chinese
in China and. The Chinese in the US?
Yep. OK, last question.

(58:16):
OK, looking into 2026, what are you going to be watching?
Wow. I mean, everything we just
talked about, I think it's very interesting where I, I really
feel like, you know, we're at a,a, a, a point where so many
forces are so many fundamental forces are shifting and maybe
like working against each other.But obviously the biggest, you

(58:39):
know, question mark is US China,right?
That will just determine many, many, many things.
Okay so Jim, we must do this again.
Maybe when I'm in Hong Kong we will find a studio.
I'm pretty sure the process Singapore guys can find me a
studio there and we should do itface to face.
Okay yeah so in closing, 2 quickones.
Any recommendations recently? Yes, I actually do.

(59:01):
So I highly recommend the new book Empire of AI, written by my
friend and former Worcester Journal colleague Karen Howe.
Yes, I'm going to get her on theshow.
I haven't finalized the date yet, but I'm I, I will try to.
We we are more or less locked down with the date already.
Just a quick plug. So I'm actually hosting a book
talk with the Karen at the Foreign Correspondents Club in

(59:24):
Hong Kong. Tell.
Her that next week to interview.Thank you as well.
But if any of your listeners arein Hong Kong, please, you're
welcome to join the the the the event in person at the SEC.
In Hong Kong before that, but I'll.
Try and then the other book I want to recommend.
It's not a super new book but I only started reading it

(59:48):
recently. It's the autobiography of the
worlds I see. It's just such a great book.
OK, yeah. So how did my audience find you
and the information? So the information you can find
us www.theinformation.com. And since you mentioned the deep

(01:00:09):
research, yeah, I really like it.
By the way, OK, I don't know howyou as a subscriber finds it.
I tested because I haven't gone to your next the user I tried
it. I actually enjoy like if I want
to find very very quick quips where I need to verify some
information like you know, I do recall was this company invested

(01:00:32):
in it? I find it better than
Crunchbase. Yeah, because it's it's based on
our own reporting. Yes, because you all created
that information. Yeah.
So maybe that that is the difference.
Where else Crunchbase, sometimesthe information is not correct
because it's self reported. Yeah, you understand what I'm
saying. So in that sense, that's why.
Yeah, so deep research sort of resolves, completely eliminates

(01:00:55):
the AI black box issue because it's all trained on our
reporting, right? But anyway, yeah, so I'm on
Twitter at Jinyang HK. I still like to call it Twitter
old fashioned that way. And then, you know, you're

(01:01:15):
welcome to e-mail me jane@theinformation.com with any
thoughts, comments, feedback, criticism.
I'm sure you're going to be in Singapore more often because
manners are here now. Yeah, either I or my colleagues.
OK, yeah, definitely can. Find us on YouTube, Spotify and
everywhere now and subscribe to us.
And Jian, we'll talk again soon.Thank you for having me.

All Episodes

Episode Transcript

Popular Podcasts

CrimeLess: Hillbilly Heist

Crime Junkie

Stuff You Should Know

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}Why Startups in China Are Going Global with Jing Yang

Episode Transcript

Popular Podcasts

.css-r6mb8g{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:1;overflow:hidden;}CrimeLess: Hillbilly Heist

Crime Junkie

Stuff You Should Know

All Episodes

Why Startups in China Are Going Global with Jing Yang

CrimeLess: Hillbilly Heist