Innovation, DeepSeek, and China

Investment Analysis Clubs Macro Economic Trends and Risks
1

Here is an interesting interview with the head of DeepSeek, Lian Wenfeng, who seems to have Musk’s genius for accomplishing the improbable without the megalomania. Wenfeng’s objective is not to maximize profits or save the world. It is to achieve human-level AGI, which is a bit scary as I don’t think he ever considers the impact of creating a human-like intelligence that is smarter than humans.

Anyway, here are some interesting excerpts with short comments. Deepseek: The Quiet Giant Leading China’s AI Race

China has been an AI follower than a leader:

> “China should gradually become a contributor instead of freeriding . In the past 30+ years of the IT wave, we basically didn’t participate in real technological innovation. We’re used to Moore’s Law falling out of the sky, lying at home waiting 18 months for better hardware and software to emerge…But in fact, this is something that has been created through the tireless efforts of generations of Western-led tech communities. It’s just because we weren’t previously involved in this process that we’ve ignored its existence.”

China’s deficiency in innovation:

" What we lack in innovation is definitely not capital, but a lack of confidence and knowledge of how to organize high-density talent for effective innovation…In the past 30 years, we’ve emphasized only making money while neglecting innovation. Innovation isn’t entirely business-driven; it also requires curiosity and a desire to create. We’re just constrained by old habits, but this is tied to a particular economic phase."

How Deepseek fosters innovation is very nonCommunist. Hungry ambitious and smart young people are put in a room with little supervision, lots of resources, and a simple but ambitious objective—create AGI.

> “We are mostly fresh graduates from top universities, PhD candidates in their fourth or fifth year, and some young people who graduated just a few years ago…The team behind the V2 model doesn’t include anyone returning to China from overseas — they are all local…DeepSeek is still entirely bottom-up. We generally don’t predefine roles; instead, the division of labor occurs naturally. Everyone has their own unique journey, and they bring ideas with them, so there’s no need to push anyone. While we explore, if someone sees a problem, they will naturally discuss it with someone else…Anyone on the team can access GPUs or people at any time. If someone has an idea, they can access the training cluster cards anytime without approval. Similarly, since we don’t have hierarchies or separate departments, people can collaborate across teams, as long as there’s mutual interest…Our hiring standard has always been passion and curiosity. Many of our team members have unusual experiences, and that is very interesting. Their desire to do research often comes before making money.”

The ultimate meritocracy in a socialist setting. Sounds a bit like the old NASA, or Bell Labs. But very different from the top-down approach of a centralized authority micro-managing research as is common to many Asian governments and companies, and where the US seems to be headed.

Finally, Wenfeng gets to the core of how to create an innovative society: (had to insert “apple” because of TMF censor).

" In the future, hard"apple"core innovation will become increasingly common. It’s not easy to understand right now, because society as a whole needs to be educated on this point. Once society allows people dedicated to hard"apple"core innovation to achieve fame and fortune, then our collective mindset will adapt. We just need some examples and a process."

That in a nutshell is the Silicon Valley. Here is China’s problem as I see it. Xi wants stability and control but innovation requires and creates disruption and anarchy. Seems mutually exclusive. DeepSeek may be more the Chinese exception than the rule until a change in government.

5 Likes
2

Again you rely on xenophobic stereotypes (a few of which may be true, or partially correct.)

It is without exception that “following” economies play catch-up by copying the leaders, at least for a time. We would hardly expect African nations to develop a cell phone infrastructure without seeing what advanced economies are doing, and to try to emulate it as quickly as they can. (For the record, it also allowed them to bypass the expensive “copper wires to every home” phase - obviously at the cost of being a backwards society for an extra 100 years.) Yet now cell phones are common across the entire African population, and they have even invented payment methods which serve them better than US apps do.

Likewise, the Chinese are a “come from behind” society, having been a subsistence rural economy just 50 years ago. Drawing conclusions about their ability to innovate misses the point entirely since there was little need to do so during a time of “catch up”.

Now it may be true that the Chinese will fail to innovate, although the emergence of a vibrant EV manufacturing center, EV battery development research, high speed rail, renewable energy, quantum computing, lunar sampling missions, and now AI might cause you (and others) to reconsider that conclusion which I think premature.

The future of the Chinese economy lies in innovation, and everyone in China knows it. But that hasn’t always been true. Innovation didn’t drive the manufacturing miracle that has unfolded in China over the past half century, during which some 700 million people have been raised—or lifted themselves—out of desperate poverty. Instead the driver has in large part been what might be called brute-force imitation.
But can China innovate? Can it compete at a global level with developed nations that have built their economies on innovation for decades? Many observers are doubtful. In recent years, they note, the West has steadily produced an abundance of innovations and innovators, while China has produced relatively few. In March 2014 this magazine published “[Why China Can’t Innovate,” by Regina M. Abrami, William C. Kirby, and F. Warren McFarlan, an article that captured the conventional wisdom. The authors’ arguments were sound and well supported at the time. But just two years later eight of the 10 companies that had reached a $1 billion valuation in the shortest time ever were Chinese—and six of those eight were founded the year that article was published.
Those are startling numbers for a country that in 2020 ranked only 14th on the Global Innovation Index. Something clearly propelled those Chinese companies to the top, but the metrics we use to evaluate innovation have missed it…. in the past five years, as an “innovation cold war” has taken shape between world powers, China has achieved a kind of parity with the United States—and the driving force behind its success may not be its innovators at all.
To understand what’s powering the global rise of Chinese companies, we need to recognize that China now has at its disposal a resource that no other country has: a vast population that has lived through unprecedented amounts of change and, consequently, has developed an astonishing propensity for adopting and adapting to innovations, at a speed and scale that is unmatched elsewhere on earth.
It’s *that* aspect of China’s innovation ecosystem—its hundreds of millions of hyper-adoptive and hyper-adaptive consumers—that makes China so globally competitive today. In the end, innovations must be judged by people’s willingness to use them. And on that front China has no peer.
3 Likes
3

That made me laugh. It seems that TMF is heading the way of China and China is head the way of the old TMF.

1 Like
4

Nope, I am making sociological/economic stereotypes. My claim is innovation increases the more a society recognizes and encourages individual over collective accomplishments and the more accepting it is to social disruption and change. Because that is what innovation does. It disrupts and causes change. That is why theocracies tend not to be the most technologically innovative regardless of how much money they have.

Innovation also tends to occur from the bottom up than the top down. That is the antithesis of the communist Chinese central planning of the economy. Chinese scientific community is also very hierarchical, with a relatively few “superstar” scientists managing large groups of younger scientists. By management I mean directing the research and allocating funds. This is different from the US system of lots of independent labs competing for individual government grants. I think the US system is a bottom up approach that tends to be much more innovative.

3 Likes
5

I didn’t see any xenophobia in that post, but maybe I missed it.

Which statement was xenophobic?

1 Like