Immediate Consequences
From Wise Nano
The long term effects of MNT appear to converge on one of three possibilities, Friendly AI,utter disaster, and forcibly imposed partial-stasis. Because one of these resolutions appears to be extremely likely within a year of the development of MNT (sooner if MNT is developed later and software techniques are more advanced) it is most urgent that effort be concentrated on outlining the size, timing, and probability of scenarios within the first year after MNT is developed. I will try to describe consequences under the assumption that the pace of MNT development is gradual. For example, under the assumption that MNT is ultimately developed by a mid-sized team (budget<$10,000,000/year for <10 years) within a large corporation, and that this takes place approximately fifteen years from now (for instance, assemblers are invented on January 1st of 2020).
I have chosen this date because it appears to me to be about the median best-guess assuming that no massive resources are put into MNT by a government or multi-national, and because it puts the event in question far enough in the future that our efforts to prepare, if taken promptly, can be expected to have a substantial impact. It is also chosen because it seems to me to be near the outer limit of the probable life expectancy of a geo-political situation that can be extrapolated from our own. Governments of developed nations will probably not be quite bankrupt yet, significant labor will still be needed for manufacturing and agriculture, and no other technologies of transhumanist import are likely to have moved out of the laboratories, with the exceptions of "mainstream" nanotechnology and biotechnology and of greatly improved surveilance technologies, which can confidently be predicted to have come into their own by this time.
I will assume that belief in MNT has not substantially changed from its current level. Further safe assumptions include the assumption that a nanofactory and control software can be produced faster and more cheaply than an assembler, but that this does not guarantee that they will be pre-designed. Scenarios should consider consequences with and without significant pre-design. Realism must allow for the development of technologies that have not been thoroughly analyzed, such as electronic nanomanufactured computers far superior to rod logic.
The first consequence of MNT that I would like to point out is that as soon as a substantial fraction of the financial community recognized the possibility of MNT there would be massive anticipatory disruption. The prices of resources that were believed likely to lose value would lose value immediately. Uncertainty would almost certainly lead to greatly increased interest rates, making it unlikely that the unemployed would find new work even if they had skills that were still valued. The increase in interest rates might be so great as to rapidly force most (already massively indebted, assume debt/GDP ratios > 1/1) national governments to pay their bills by printing money rather than by borrowing. In the short term, real-estate prices would explode. After a few months they might crash, either due to autonomous people's lesser dependence on prime locations or due to concern about the maintenance of property rights for non-portable forms of property. The technical basis for cheap comfortable autonomous living would probably take several months to develop however. Autonomy would also create a labor shortage (probably concurrent with, not in place of the job shortage) because people who could afford to live autonomously might refuse to work. Conscripted labor might be used in many countries.
The primary determinant of the nature of long-term equilibrium is almost definitely the identity of the first group to develop some form of general artificial intelligence. If this GAI is created via directed evolution or other non-analytical methods other than uploading, this is almost certain to lead to human extinction. If uploading is pioneered by individuals who appreciate the threat of GAI or of uncontrolled digital human evolution, it is almost certain that they can deploy robots capable of eliminating these threats. Because nanotech driven neurological modifications are likely to be relatively safe and powerful compared to those available today, and because nanotech will provide a powerful set of tools for studying the brain compared to any currently available, it is very likely that neurologically augmented humans will participate in developing the first GAI. One of the most urgent tasks facing us may be the analysis of the unintended consequences of using nanotech tools to enhance the sorts of reasoning that would be needed.
