Jun 9, 2004
Drexler dubs “grey goo” fears obsolete
Eric Drexler, the “father of nanotechnology”, has revealed his latest views on molecular manufacturing. Writing in Nanotechnology journal, Drexler explains how he now believes that self-replication - the initial source of “grey goo” fears in which nanomachines run amok and overwhelm the world - is not an essential part of the molecular manufacturing process. Nanotechweb.org’s Liz Kalaugher contacted Drexler to find out more.
nanotechweb.org (NTW): What are the most important themes of your paper in Nanotechnology?
Eric Drexler (ED): The paper explains updated molecular manufacturing concepts and why these no longer include tiny self-replicating machines. This makes fears of accidental runaway replication - loosely based on my 1986 grey-goo scenario - quite obsolete. Chris Phoenix [of the Center for Responsible Nanotechnology] and I wrote the paper to counter the main threat posed by grey goo, which is that all the hype diverts attention from more important issues - research directions, development paths, and the role of advanced nanotechnologies in medicine, the environment, the economy and in strategic competition.
NTW: How have your views on molecular manufacturing and nanotechnology changed as the field has advanced?
ED: Research and thinking in this area has come a long way since my first refereed journal article on the subject in 1981 and my 1986 book Engines of Creation. The PNAS journal article was based on a biological model of molecular machine systems - hence the early focus on self-replication - but the logic of the technology led to the very different, non-biological approach described by Nanosystems in 1992 and in the more recent literature. The common theme is mechanical control of molecular assembly, but a lot has changed. In some ways current ideas have moved closer to Richard Feynman’s original 1959 vision.
In particular, it turns out that developing manufacturing systems that use tiny, self-replicating machines would be needlessly inefficient and complicated. The simpler, more efficient, and more obviously safe approach is to make nanoscale tools and put them together in factories big enough to make what you want. Throughout history, people have used tools to make more and better tools. That’s how we got from blacksmith’s tools to automated industries. The natural path for nanotechnology is similar.
This new picture shows Ralph Merkle’s convergent assembly architecture for a factory that joins small parts to form larger ones, starting with nanoscale blocks. The machines in this would work like the conveyor belts and assembly robots in a factory, doing similar jobs. If you pulled one out, it would be as inert as a light bulb pulled from its socket.
A physical analysis indicates that with molecular manufacturing a desktop-scale factory will be able to convert simple raw materials into a billion-processor computer in less time than it takes to boot a modern Windows machine. I found this surprising, but the basic physical principles are straightforward - simpler in some ways than what we see in labs today, because all the steps would be better controlled. Today, we don’t yet have the tools necessary to build the tools to build a factory like that, but hard, creative work in the labs is chipping away at that barrier, and new developments are likely to accelerate the pace of change.
The analysis for this sort of system is written up in Nanosystems, and some of the more recent work is referenced on my website, e-drexler.com. The analysis is basically an exercise in molecular modelling, systems engineering and applied physics. I invite further critical review of the literature, but so far it’s stood up remarkably well.
A major area of progress has been in understanding how to get from, say, protein engineering and current nanofabrication as we see them today, to a first-generation programmable molecular assembly machine. From there, the scale-up to larger machines seems more straightforward. More recently, researchers have been examining different pathways that could be quicker than those outlined in Nanosystems.
It’s important to remember that history is full of unintended consequences. One unintended consequence of more conventional nanotechnology work is progress toward molecular manufacturing. We’ll be better off if we try to see where we’re going.
NTW: Is grey goo fact or fiction?
ED: The popular version of the grey-goo idea seems to be that nanotechnology is dangerous because it means building tiny self-replicating robots that could accidentally run away, multiply and eat the world. But there’s no need to build anything remotely resembling a runaway replicator, which would be a pointless and difficult engineering task. I worry instead about simpler, more dangerous things that powerful groups might build deliberately - products like cheap, abundant, high-performance weapons with a billion processors in the guidance systems.
NTW: How safe do you think molecular manufacturing will be?
ED: Molecular manufacturing itself should be safer and cleaner than ordinary manufacturing, yet also faster and more capable. Some products, like computers and cheap solar collectors, seem very benign and attractive. Other products, in the wrong hands, could pose a threat. The Center for Responsible Nanotechnology has investigated ways to regulate the products of molecular manufacturing. The Foresight Institute has issued guidelines for future product safety. We need more thinking and discussion along these lines.
NTW: What, if any, regulations do you think are appropriate for the field of nanotechnology?
ED: The appropriate regulation depends on the processes and products. I think there’s been an over-emphasis on the need for new regulations, in part because public perceptions have confused future possibilities with current products. Existing consumer and environmental safety regulations provide a good framework, but they need to be applied, and adjusted to recognize novel properties resulting from small size rather than different chemistry.
As the technologies advance and become more powerful, I think we’ll face some novel issues. These deserve thought today, but the need for regulatory action is still down the road.
NTW: Do you feel that the public perception of nanotechnology differs from the reality?
Public perceptions are still forming - a survey conducted in the UK showed that only 19% of respondents were able to give some kind of definition - but there’s already plenty of confusion. The history of the term has a lot to do with this.
When “nanotechnology” first came to public attention in the late 1980s through my book Engines of Creation, the term referred to early ideas about future molecular manufacturing and what it could do. This promised great things - inexpensive, enormously useful products - but from the start the technical ideas got popularized and twisted and mixed with hype about runaway replicators.
The public now confuses today’s nanotechnologies - a diverse set of products with nanoscale components - with obsolete ideas about future molecular manufacturing nanotechnologies. This has led to confusion and a potential regulatory backlash. Ordinary, manageable risks, such as issues of nanoparticle toxicology, become magnified in the eyes of those failing to distinguish current developments from powerful future transformations that accidentally ended up sharing the same name.
Efforts have been made to get rid of this confusion by denying that the powerful future technologies are even possible, but these denials have failed - I think for good reason. We need another, more reality-based way to move forward, to start to unwind this confusion. I see two basic steps:
The first step is to recognize that a physics-based analysis - which can be subjected to further peer review, revision and extension - has given a glimpse of what future nanotechnologies will be able to do. To put this glimpse in context, we should try to show the public a realistic timeline. This will place grand expectations and concerns for the future where they belong, out of the way of current lab research, which shouldn’t be burdened by them.
The second step is to show the public a more accurate picture of where advanced nanotechnology can lead - to replace the image of scary self-replicating nanobugs with a useful appliance, a bit like a Star Trek replicator. Of course this also gives cause for worry, but the chief concerns centre on the ongoing economic and strategic competition for powerful new abilities. This sort of concern motivates further research, not suppression, so the spillover for nanoscience and industry should be positive.
NTW: How do you see nanotechnology developing over the next 10 years?
ED: In the last 10 years we’ve seen many steps toward molecular manufacturing, including the advent of the scanning tunnelling microscope/atomic force microscope (STM/AFM), and rapid progress in computational nanotechnology and protein engineering. I expect the field to continue to grow at an exponential rate with further innovation in tools for sensing, manipulation and the design of atomically precise structures.
Given the substantial strategic and economic implications, I expect that competitiveness issues will play an increasing role in nanotechnology investment and development. As I’ve said before, developing molecular manufacturing will require systems engineering, and it will be interesting to see which parties harness the world’s growing scientific knowledge to a focused systems engineering effort.