Center for Responsible Nanotechnolgy (CRN)

Results of Our Ongoing Research

These pages, marked with GREEN headings, are published for comment and criticism. These are not our final findings; some of these opinions will probably change. LOG OF UPDATES

CRN Research: Overview of Current Findings

Why Early Development May Be Safest

Overview: Early development of molecular nanotechnology (MNT) increases some risks, but reduces others; overall, we think it's safest to develop as soon as possible. This is a preliminary conclusion, and we may change our opinion, but there are solid reasons for taking this position. The development of MNT seems inevitable sooner or later. If development is delayed, it will rapidly become easier and cheaper, thus harder to control. Also, it's probably the case that early development will allow more time to develop MNT-based protective technologies—which may be necessary to cope with some dangerous MNT-based technologies. Finally, if it's done right, molecular manufacturing could save millions of lives per year and greatly decrease the environmental damage we're already doing. The costs of delay (opportunity costs) are significant, and may even outweigh the risks of development.

MNT is inevitable; the question is when.	Science and technology are rapidly gaining competence at the nanometer scale. According to Ray Kurzweil's recent testimony to the US Congress, "most of technology will be 'nanotechnology' by the 2020s." In other words, before 2030, most fields of technology will make routine use of nanometer-scale components. At some point after that, the seemingly miraculous MNT will be commonplace: regardless of whether Drexler-style nanosystems are ever built, automation and miniaturization will have duplicated the important aspects of the technology. But MNT almost certainly will be developed earlier. There will be strong economic pressure to develop it as soon as the cost of development falls within the range of corporate R&D. Given the national security implications, it's likely that governments will be working on it well before then. And, as we explain here, there may be reasons to develop it internationally, before national programs can get started.
Development will rapidly become cheaper.	Technically, the development of MNT depends on a design and on a molecular manipulation capability. Chemistry, scanning probe microscopy, optics, lithography, and a variety of other fields are rapidly advancing our ability to design, create, and manipulate molecular structures. Just in the last few years, several new families of large designer molecules have been discovered. Computer simulation is also developing rapidly as computers become more powerful and new algorithms are discovered. These trends, and several others, will continue. As more options become available, the design of a molecular manufacturing system will become easier. Development efforts will require far less investigation of fruitless possibilities. Today, an MNT development program would cost many billions of dollars. Sometime in the future, probably well within this century, it will be a science fair project. Between the two extremes is either a rapidly falling curve or a sudden, unpredictable decrease in cost.
Delay may lead to multiple MNT projects.	If MNT is not developed as soon as possible, the rapidly falling cost will allow several players—corporations and/or nations—to pursue independent development projects. A delay could happen for several reasons. Overly pessimistic opinions about the feasibility of MNT could reduce initial interest. Environmental or social concerns, or simple Luddism, could delay the research. Spending large amounts of money requires either political will or corporate boldness, which could be lacking at the crucial time.
	If MNT development is significantly delayed for any reason, then by the time a project is started, development will be considerably easier. Political and economic pressure for development will rapidly increase. The rapidly falling cost of development will allow more groups to enter the race, while also greatly improving the cost/benefit ratio. Similarly, there will be a rapid increase in the number of foreign powers who could make a credible attempt at developing what is (among other things) a massive military force multiplier; once one program starts, a perceived "nanotechnology gap" could lead to crash programs in a number of countries that do not fully trust each other. As the number of contenders in an arms race increases, the risk of preemptive strikes probably increases as well.
One early project is easier to control.	If MNT is developed in several projects almost simultaneously, each owner will be able to choose what to do with it. There will be less scrutiny of each project. Any controls that need to be imposed will require much more effort. Conversely, a single project provides a single point to monitor and control. An early project, started when the resources required are still quite large, reduces the uncertainty about who else could be working on MNT development. It may also reduce the incentive for other projects to start later; many intellectual property rights, and some national security benefits, of an MNT program will be lost if it can't keep up with the first project.
Early MNT gives us a head start at defensive technologies.	Some of the problems that MNT could create may only be dealt with effectively by MNT-based technologies. For example, as noted by Robert Freitas, widespread detection networks may be necessary to deal effectively with grey goo. A system that can sample large volumes of air or water for sub-micron particles, and respond with sufficient speed to clean up an infestation, could probably only be built by MNT. Nanotech-built weapons may pose a far greater threat to human well-being. It would be a good idea to start practical engineering on defensive MNT-built technologies well in advance of the development of aggressive or dangerous technologies. This might be helped by developing MNT early, on the theory that early development will allow more selection—at least at first—of who gets to do research with the technology.
Early MNT can solve tangible problems.	Technology, applied appropriately, can mitigate many current problems. Large areas of the world currently suffer from a lack of technological infrastructure. This is currently a self-perpetuating problem. Portable, rapid, flexible manufacturing could solve it quickly. Health requires sanitation; efficient trading and democratic government require communications. Sanitation and communication could be supplied almost trivially with MNT. This would save millions of lives in the poorest areas of the world, and greatly increase global prosperity (which would provide vast new markets for commercial enterprises).
	Advanced technology can reduce much of the current environmental burden. From a hut heated by a smoky dung fire to a mansion with kilowatts of incandescent lights (which are only 1% efficient), people worldwide throw away most of the energy they consume. The same is true of potable water—most of it is used for industry and agriculture. In countries fortunate enough to have modern medicine, present-day techniques require awesome quantities of material and labor to keep their populations somewhat healthy.
	MNT will not magically invent the solutions for most of these problems. But once a solution is developed, it can be applied quickly and globally at very low cost. If MNT is developed even a few years early, and used well, tens or hundreds of millions of lives will be saved. Any risk that is exacerbated by early development must be balanced against this very significant benefit.

DEVIL'S ADVOCATE —

If it's so dangerous, why allow development at all?

As we said above, sometime in the next few decades, MNT will become very easy to develop for any country or large corporation. We don't believe that development can be prevented forever.

Wouldn't it be better to wait until we know more about how to deal with the risks?

Some knowledge can only be gained in the lab. If we wait, people will be thinking up new kinds of weapons, and getting better at using MEMS and other small technologies. MNT will be disruptive, and hard to control, whenever it happens. We think that there'll be less disruption if it's developed early enough that it's only developed in one place. We're open to argument on this point. But remember that millions of people will die tragically and preventably each year that it's not developed. That's a certainty, not a risk, but any risk must be balanced against it.

[Comment submitted by Noah Ennis] Very interesting series of articles. I wonder, though, if "millions of people will die each year nanotechnology isn't developed" is not a false dilemma. After all, the choice is not strictly between developing nanotechnology and letting millions in the third world perish. The choice is between those two extremes and the middle ground of alleviating poverty through existing technologies -- for example, vaccines and mosquito nets -- which are already well within the West's technological and economic capabilities to produce in large quantities. The reason they don't is not economic at all, but political. The conclusions drawn by CRN about MNT point to far more powerful intellectual property protections, as well as centralized control of the manufacturing, which seem to make the situation worse, not better. So although MNT has a huge amount to offer for mitigating the effects of poverty and developing third world infrastructure, posing its development as a moral imperative is somewhat of a red herring when there are alternatives within our means that are subject to the same imperative without any of the relevant moral risks. There are many superlative reasons for advocating MNT, but the moral implication that failure to do so costs X many lives, inasmuch as MNT is a technological instead of political solution, strikes me as misleading.

Thanks, Noah. Your point is well taken. Millions of people around the world are suffering and dying needlessly today, not because we lack basic technologies that could help them, but because we lack political will to implement basic solutions. A large part of CRN's work is aimed at understanding these political mechanisms and the underlying social systems that drive them. We agree that creating a powerful new technology, like MNT, is only half the battle -- or maybe much less than half. The real challenge is in finding effective ways to guide the development and proliferation of the technology so that the most beneficial outcomes can be achieved, while the greatest risks can be averted. We've said all along that it won't be easy, and that's why we urge responsible government agencies, educational institutions, concerned businesses, and civil society groups to adopt some or all of CRN's Thirty Studies as an important first step toward clarifying the many issues involved.

Next Page: The Need for International Development

Previous Page: A Solution that Balances Many Interests

Title Page: Overview of Current Findings