In his classic tome on nanotechnology, "Engines of Creation," K. Eric Drexler, founder of the Foresight Institute in Palo Alto, Calif., envisions how molecular manufacturing could transform the world in ways both beneficial and troubling. As part of that vision, he outlined the "gray goo problem": that self-replicating machines, designed and built to reproduce in nature, would run amok and devour the world's resources.
Ironically, what Drexler says he did not imagine when the book was first published in 1986 was how this notion of gray goo would run amok, growing beyond Drexler's original suggestion and dominating popular perception and policy discussions of molecular manufacturing.
It may be a highly speculative scenario, but out-of-control nanobots have proven to be irresistible plot points for best-selling authors, screenwriters, and video game creators. Even though a gray goo apocalypse is pretty removed from reality, especially when considering current nanotechnology research, Drexler's critics say that the hypothetical scenario could frighten away funding and support for nanotech.
"Not only have people been excessively worrying about gray goo, but people have been worrying about people worrying about gray goo," says Chris Phoenix, director of research at the Center for Responsible Nanotechnology in Brooklyn, N.Y. "Gray goo, almost from the beginning, has been a misunderstanding," he explains. The idea that molecular manufacturing could be "one 'oops' away from a disaster has never been accurate."
To clear up some misconceptions, Phoenix and Drexler recently wrote an opinion piece titled "Safe Exponential Manufacturing," which deemphasizes gray goo and attempts to refocus the dialogue about the future impact of molecular nanotechnologies [Nanotechnology, 15, 869 (2004)].
"Runaway replication is well within the realm of physical law, but building a device able to behave that way would be a deliberate and difficult engineering task, not an accident," Drexler says. "There is no technical or economic reason to build anything remotely resembling a runaway replicator."
From a practical point of view, Phoenix and Drexler contend, molecular manufacturing systems will be no more capable of running wild, replicating, self-organizing into intelligent systems, and eating people than a desktop printer is. To function, molecular nanofactories will need programming and raw materials. Stopping them will be as easy as pulling the plug on the power supply.
A more likely and worrisome scenario, Phoenix and Drexler argue, would be molecular nanotechnologies designed as weapons. Among other things, they say, nanofactories could be used to make vast quantities of lethal or environmentally destructive products.
"In the U.K. in particular, recent fears about runaway replication scenarios have hampered serious policy discussion and diverted attention from more pressing issues," Drexler tells C&EN. He and Phoenix hope that by putting the gray goo myth to rest, they will be able to open a dialogue examining the real technological and strategic prospects that molecular manufacturing has to offer.
The paper, which appears in a journal published by the London-based Institute of Physics, has managed to garner attention from the popular press in Britain. Drexler also wrote "You Have Been 'Gray Gooed,'" a more accessible opinion piece that appeared in the June 18 issue of London's The Times Higher Education Supplement.
But recalling the nanobots--even hypothetical ones--might not be so easy. Gray goo is a sticky metaphor, an image that won't be easy to erase from the public's imagination, according to David Rejeski, director of the Foresight & Governance Project at the Woodrow Wilson International Center for Scholars in Washington, D.C. "You can't treat social and cultural memory as if it were a hard drive that you can just wipe clean," he says.
Anyway, Rejeski isn't certain that the gray goo scenario is actually such a bad thing. "The scientific community has a responsibility to think about things that seem inconceivable," he argues.
He admits that, for nanotechnology, there are more pressing issues to be dealt with in the near term--for example, determining the safety of nanoparticles currently in use. But he adds, "I'd hate to see us take longer term risks off the table, because these are important for us to deal with, too."
Rejeski also notes that trying to play down the gray goo scenario may actually prove counterproductive. Instead, he says, to the public, it could look as if Drexler is trying to hide something. Rather than working to downplay the warning of gray goo, Rejeski suggests that the scientific community develop metaphors for the beneficial aspects of nanotechnology that capture the public's imagination as powerfully as gray goo.
Visionaries seeking to create such a metaphor might even take a cue from Drexler and tap into the allure of alliteration. "Several people have asked me, if I could turn the clock back 20 years, what would I do?" Drexler writes in The Times Higher Education Supplement. "I still think it would have been irresponsible to describe the benefits of nanotechnology without addressing the potential risks. I would, however, have chosen a less appealingly alliterative term than 'gray goo.' If a similar scenario had been named 'deliberately engineered dangerous self-replication,' it would have had a lower profile."