Re: Nanotech Scenarios

Anders Sandberg (
01 Jun 1999 11:26:15 +0200

"Chandra Patel" <> writes:

> Here's a bunch of questions I'm currently wrestling with:
> Will the widespread hype begin before or after the development of an
> Assembler? Will the public be informed about the capabilities of an
> Assembler prior to its development or only afterwards?

It largely depends on people like us and the Foresight Institute, as well as the public relations people and news media. Likely, there will be bursts of hype (there are already, and unfortunately it is sometimes we who originate them) but if the development of nanotechnology is slow and there are few obviously amazing effects then it will gradually settle in (just look at the spread of bioengineering in the chemical industry). If there are some obvious breakthroughs or telegenic effects, then the hype will roar.

> Who has the best chance of developing the first Assembler? the U.S.
> government? another government? coalitions? corporations? conglomerates? the
> nanotech Galileo tinkering in vis lab?

Everybody but the Galileo, really. While individual geniuses can do amazing work, it seems like getting broad technologies like nanotechnology or computers to work requires broad research efforts - not necessarily done by the same organisations. Many seems to think that there will be a race towards the first assembler, but it is more likely people will be moving in that direction looking for the other priofitable stuff that is between now and then (the crash programs to get AI have all failed, while the developments in other areas such as search engines, parsing, neural networks etc are advancing at their own pace in that direction).

It is important to remember that no single group can likely develop nanotechnology on its own - it requires very broad knowledge and technology which is distributed across many disciplines. So everybody relies on the products of everybody else (or it will become extremely costly and inefficient - there is a reason for defence contractors and companies sponsoring basic research).

> How will the nature of the first developer affect how events progress in the
> MNT+x era? Will it matter who the first developer is because governments
> will seize the technology?

Suppose a farsighted government wanted to sieze the technology of the Internet in the late 80's, having realized the potential. Would it have worked? Answer: no, because the information was already out there in journals, in the minds of thousands of researchers, engineers and experts. It might have delayed the application, but it would not have stopped the creation of other nets.

The same goes for nanotechnology: if the US government declares nanotech a matter of national security, it becomes a huge trade advantage for Japan (or China or Sweden or Brunei) to sell it. And the information will largely exist in the public.

Of course, this all depends on the existence of governments with sufficient strength or idiocy to try something like this. Given the timeframe of nanotech development and the current world trends, this is not entirely likely.

> Will the financial markets act similar to Hanson's idea futures market and
> begin factoring in the eminent development of the Assembler into its
> valuations? This assumes that the finance types identify what affects such
> an event would have on valuations.

Maybe. Once nanotech industries start to show progress, there will likely be investment interest. The "sudden jump" scenario some here favors actually means nobody will invest much in nanotech until after it has become mature, which seems rather unlikely (beside the main assumption of a sudden breakthrough). A gradual development will instead see a steadily increase in funding (and likely financial hype - watch out for nanobubbles).

> What will the economic impact be? Will pre-MNT capital lose all its value?
> What will be the commodity of the post-MNT era? The most obvious answer is
> 'the ability to make Assemblers'. But what about creativity, Assembler
> software, Assembler software debugging software, easily mutable materials,
> heat dissipation technology, etc.?

MNT changes a lot of stuff, but capital in airlines, medicine, services, information, entertainment etc will obviously remain and likely grow even more as MNT is applied. Just because you can turn butylyl into diamond (early assembler) doesn't mean I have to sell my shares in a construction company; when you can churn out diamond beams the company will simply use them. And when nanodevices can grow buildings from scratch, the construction company will still be there, but it will lug around the equipment to set up growth baths and do the surveying/growing/testing instead of the old construction work. Maybe in the long run the firm will just be a single person subcontracting a truck, a few loading/unloading robots, some chemical production and a network of designers and telepresent building inspectors, but the capital will remain.

And of course, Assembler programming will boom. Just two words: molecular debugging.

> Should Assembler software be open source?

This is really an ethical question. Being an academic I'm of course all in favor of open source nanotech.

> Does it matter if Assembler > software is efficient?

Inefficient software can work well if it is introduced faster than the competitors superior products (witness certain software companies). Hopefully quality will win in the end.

> What weight should be given to the 'crazy screw with
> a problem' factor where the screw decides to implement some "goo" scenario?

Goo might be a lesser problem than an assembler making nerve gas out of sewer water. The problem here isn't exactly nanotech, but the existence of these screws and powerful technologies. Screws (and screwups) will likely always exist, but simple safeguards can likely limit the problem a bit (don't sell ammonium nitrate to the drooling character with a unabomber pin) but in the end it is a matter of setting up a working protection/law enforcement system. Definitely not a trivial problem, especially since we want it to work in an open society. But at the time individual maniacs have access to nanotech advanced enough to produce significantly dangerous effects we can expect that people and organisations pooling resources can produce countermeasures significantly better.

> What would be the motivation of a 'crazy screw' since many ailments that
> drive 'fanatics' will be remedied or have the potential to be remedied with
> MNT?

Personality disorders, religious, political or other ideological convictions, lack of forethought ("I wonder what happens if I do *this*?") and simple mistakes ("Oh, I forgot the stop command... darn!").

> What is the future of governments in the MNT+x era?

What is the future of governments in the pre MNT era? We are clearly seeing that information age societies are chafing under industrial-age type governments, and that some reform is necessary. Whether this is towards some kind of flexible networked governments, recursive federalism, global fluidity, nightwatch states or something completely different is uncertain. But MNT will make this drive even stronger, especially since the pace of change will accelerate making many regulations, institutions and laws completely out of date.

> What platform is MNT/OS likely to be based upon?

Oh no, you are trying to get us into a religious war here! :-)

Seriously, why should it be based on any existing platform?

Anders Sandberg                                      Towards Ascension!                  
GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y