Eric Drexler on Alternatives to Grey Goo

Printable Version (PDF)

Little did I expect to be writing two articles on "grey goo" in less than a week, but Eric Drexler has just published and interesting article (with Chris Phoenix) with his updated perspective on this subject. The articles is published by the Institute of Physics, and can be downloaded free, for one month (registration required).

As I noted in a previous post, Drexler had warned in his 1990 book Engines of Creation, of a worst-case nanotechnology catastrophe in which the biosphere is destroyed by self-replicating bacteria-like nanomachines. This scenario was, and continues to be referred to as the "grey goo" scenario.

Phoenix and Drexler now worry that too much emphasis is being placed on the "grey goo" scenario, and too little on other dangers posed by nanotechnology. Each part of this is interesting.

On the positive side, they think we need not worry unduly about grey goo because current views of nanomanufacturing focus not on self-replicating assemblers, but rather on minaturized versions of conventional factories. This is extremely good news, and I think does go a long way to justifying his claim that the "grey goo" worries are not massively relevant to the kind of nanomanufacturing in which we are likely to engage. We need to be very clear, however, that what has changed here is that Drexler and others are now advocating a different, and much safer, manufacturing approach: if the self-replicating assembler route were to be pursued the danger would be present. Phoenix and Drexler do also put forward a number of arguments as to why, risks apart, assembler-based self-replication would be hard, inefficient, and difficult to make robust enough for the "grey goo" scenario to be relised. Again, these arguments seem very plausible, but do not alter the fact that any self-replicating nanotechnology would be pregnant with risks and should probably be avoided.

On the less positive side, Phoenix and Drexler emphasize that the development of nanotechnologies does pose very grave risks for humanity. Their concluding sentence is:

Problems including weapon systems, radical shifts of economic and political power, and aggregate environmental risks from novel products and large-scale production will require close attention and careful policy-making.

This is certainly a judgement I have no difficulty at all agreeing with.

Bibliography

Eric K. Drexler, Engines of Creation. Fourth Estate (London) 1990.

Chris Phoenix and Eric Drexler, Safe Exponential manufacturing. Nanotechnology 15 2004 pp. 869--872. Institute of Physics Publishing. http://stacks.iop.org/Nano/15/869

Bill Joy on Civilization-changing Events

Printable Version (PDF)

I wrote previously about Martin's Rees's excellent book, Our Final Hour, which warns of the dangers of what he calls "bio-error and bio-terror". The New York Times (free registration required) has now carried an interview with Bill Joy (6th June 2004), in which he discusses some aspects of a similar theme. In particular, he's worried about "grey goo", about which he has also written in Wired, and its potential to cause a "civilization changing event".

I first read about "grey goo" in Eric Drexler's Engines of Creation. Drexler is a thinker, practitioner and cheerleader for the nanotechnology movement, and he is working towards the "Assembler Breakthrough". This is the moment when we produce the first nanomachine capable of building other devices (including ones like itself) by assembling them atom-by-atom. The "grey goo" problem arises if (through error or possibly malice) assemblers start voraciously consuming most or all of the matter with which they come into contact. In Drexler's words:

Tough, omniverous "bacteria" could out-compete real bacteria: they could spread like blowing pollen, replicate swiftly, and reduce the biosphere to dust in a matter of days. Dangerous replicators could easily be too tough, small and rapidly spreading to stop—at least if we made no preparation.

Drexler's own conclusion is that: 'we cannot afford certain kinds of accidents with replicating assemblers.'

Bill Joy appears to think that this is a good argument for thinking long and hard before embarking down routes that might lead to the technologies of Drexler's dreams and nightmares. Through the rather different form of his latest novel, Prey, Michael Crichton makes a related point, as he worries about allowing self-replicating, evolving swarms of nanoparticles into the environment. One doesn't have to find every aspect of Crichton's book plausible to feel that he has correctly highlighted some serious dangers in the way that these technologies might develop, especially in a loosely regulated capitalist economy where the pressures on start-ups to generate good returns of capital can be hard to resist.

The perhaps unlikely triumvirate of a leading astronomer, a leading nanotechnologist and a best-selling author all argue that technologies scientists and engineers are working towards today have significant disruptive potential far far the appreciation of almost everyone outside their circles. And in different ways, they all argue that we need to stop and think very carefully how we can manage or control or inhibit the development and application of these technologies. I think we would do well to take their warnings seriously.

Bibliography

Martin Rees, Our Final Hour. Basic Books (New York) 2003.

Eric K. Drexler, Engines of Creation. Fourth Estate (London) 1990.

Michael Crichton, Prey. Harper Collins (London) 2002.