To follow this blog by email, give your address here...

Sunday, December 20, 2009

China Ascendant? (a comment on Robin Hanson's comment on...)

This blog post is an edited version of a comment I made on Robin Hanson's recent China Ascendant post on the Overcoming Bias blog. So, read Robin's post before reading this one!



Also, it is best read as a sort of post-script to my recent article on the Chinese Singularity in H+ Magazine. So maybe you should read that article first too ;-)

Ultimately, it may not be so important whether the US or China or India or Europe leads the advance of science and technology during the next decades.

Certainly, if you're a Singularitarian like me -- the Singularity is about the fate of mind and humanity, not the fate of nations ... and if/when it comes, it will quickly bring us beyond the state where national boundaries are a big deal.

But in current practical terms, the "where" question is an interesting one.... Especially, if a lot of the relevant developments are going to happen outside the Western world, this is worth knowing because it's going to affect a lot of decisions people have to make.

So, to get to the topic of Robin Hanson's blog post: China ascendant ???

My answer to that question is always: Maybe.

In his post, Robin makes the statement:

If China continues to outgrow the West, it will likely be because they do a few things very right, as did the West before.

The point I want to make here is a simple one: One of the things China is doing much better than the US, these days, is thinking medium-term and long-term rather than just short-term.

Perhaps long-range planning will be one of the "few things" China does "very right," to use Robin's language.

China is planning decades ahead, in their technology and science development, in their energy and financial policies, and many other areas as well.

Whereas in the US, we seem to be mired in a "next quarter" or "next election" mentality.

However, the matter isn't as simple as it seems...

It's interesting to observe that the American system sometimes does great mid-range planning accidentally (or, to use a more charitable word: implicitly)...

For instance, the dot-com boom seems kinda stupid in hindsight (trust me; I played my own small part in the stupidity!) ... but on closer inspection, a lot of the "wasted" venture $$ that went into the dot-com boom funded

  • the build-out of Internet infrastructure of various kinds
  • the prototyping of technologies that later became refined and successful.

Those VCs would not have funded infrastructure buildout or technologically prototyping explicitly, but they funded it accidentally, er, implicitly.

So in this case, the US system planned things 10 years in advance implicitly, without any one person explicitly trying to do so.

We can't explain the dot-com boom example by simplistic "market economics" arguments -- because on average, the investment of time and $$ in the dot-com boom wasn't worth it for the participants (and they probably weren't rational to expect that it would be worth it for them). Most of their work and $$ ultimately went to benefit people other than those who invested in the boom. But we can say that, in this case, the whole complex mess of the US economic system did implicitly perform some effective long-range planning.

Yet, this kind of implicit long-term planning has its limits, and seems to be failing in key areas like my own research area of AGI. The US is shortchanging AGI research badly compared to Europe as well as Asia, because our economic system is biased toward shortsightedness.

There are strong arguments that long-range state-driven planning and funding has benefited developing countries -- Singapore, South Korea and Brazil being some prime examples. In these cases, it supported the development of infrastructures that probably would not have developed in a less state-centric arrangement like we have in the US.

So, one interesting question is whether explicit or implicit long-range planning is going to be more effective in the next decades as technology and science continue to accelerate (or, to put the question more honestly but more confusingly: what COMBINATIONS of explicit and implicit long-range planning are going to work better)?

My gut feel is that the "mainly implicit" approach isn't going to do it. I think that if the US government doesn't take a strong hand in pushing for (and funding) adventurous, advanced technology and science development, then China will pull ahead of us within the next decades. I don't trust the US "market oligarchy" system to implicitly carry out the needed long-range planning.

The reason I have this feeling is that, in one advanced, accelerating technology area after another, I see a contradiction between the best path to short-term financial profit and the best path to medium-term scientific progress. For instance,

  • In AI, the quest for short-term profits biases toward narrow AI, yet the best medium-term research path is to focus on AGI
  • In nanotech, the best medium-term research path is Drexler's path which works toward molecular assemblers, but the best path to short-term profits is to focus on weak nanotechnology like most of the venture-funded "nano" firms are doing now
  • In life extension, the best short-term path is to focus on remedies for aging-related diseases, but the best medium-term path is either to understand the fundamental mechanisms of aging, or to work on radical cures to aging-related damage as Aubrey de Grey has suggested
  • In robotics, the path to short-term profit is industrial robots or Roombas, but the path to profound medium-term progress is more flexibly capable autonomous (humanoid or wheeled) mobile robots with capable hands, sensitive skin, etc. (and note how all the good robots are made in Japan, Korea or Europe these days, with government funding)
In area after area of critical technology and science, the short-term profit focus is almost certainly going to mislead us. What is needed is the ability to take the path NOW that is going to yield the best results 1-3 decades from now. I am very uncertain whether such an ability exists in the US, and it seems more clear to me that it exists in China.

The Chinese government is trying to figure out how to combine the explicit planning of their centralized agencies, with the implicit planning of the modern market ecosystem. They definitely don't have it figured out yet. But my feel is that, even if they make a lot of stupid mistakes as they feel their way into the future, their greater propensity for thinking in terms of DECADES rather than years or quarters, is going to be a huge advantage for them....

China has a lot of disadvantages compared to the US, including

  • a less rich recent science and engineering tradition
  • an immature ecosystem for academic/business collaboration
  • a culture that sometimes discourages effective brainstorming and teamwork
  • a less international scientific community
  • an unfortunate habit of blocking parts of the Internet (which doesn't prevent Chinese researchers from getting the world's scientific knowledge, but does prevent them from participating fully in the emerging Global Brain as represented by Web 2.0 technologies like Twitter, Facebook and so forth)

However, it may be that all these disadvantages are outweighed by the one big advantage of being better at long-range planning.

As Robin points out, dramatic success is often a matter of getting just a few things VERY RIGHT.