The Foresight Institute’s Founding Vision

page 1 | page 2 | page 3 | page 4

Foresight was founded in 1986 on a vision of the emerging field of nanotechnology based on Engines of Creation: The Coming Era of Nanotechnology, a work of popular science written by K Eric Drexler, published in 1986, and available on his web site as a free HTML version.

The Foundations of Foresight

Drexler structured his vision of nanotechnology as a technological revolution in three parts, the first of which he titled “The Foundations of Foresight”. The chapters below are titled as Drexler titled them.

Engines of Construction

Beginning with the twin observations that “Our ability to arrange atoms lies at the foundation of technology” and “With our present technology, we are still forced to handle atoms in unruly herds,” Drexler builds the case that biological molecular machines can be engineered to build with stronger materials than proteins, ultimately producing a “universal assembler” able to produce any arrangement of atoms allowed by the laws of nature, greatly expanding the capabilities of a range of technologies, including computing, manufacturing, medicine, space, and warfare.

His initial observation is that variations in the arrangement of atoms distinguish raw materials from finished products and disease from health. Although we have come far in learning to arrange atoms for our purposes, “the laws of nature leave plenty of room for progress”. He describes two styles of technology: “bulk technology” that has led us from stone chips to silicon chips for computers, and “molecular technology” that precisely controls individual atoms and molecules. Since microcircuits built with current [That is, 1986. In the 30 years since the publication of EOC the smallest dimension of a computer chip has decreased 100-fold, from one micrometer to about 10 nm] microtechnology are measured in micrometers (millionths of a meter), and molecules are measured in nanometers (billionths of a meter), he proposes the term “nanotechnology” to describe molecular technology that will produce nanocircuits and nanomachines.

In describing molecular technology as it existed in 1986, Drexler begins with the picture of molecules as collections of atomic beads joined by bonds that can be broken and reformed. From observing how simple molecular patterns determine the behavior of simple materials like air and water, he moves on to consider how more complex molecular patterns of proteins, DNA, and RNA make up the molecular machinery of living cells. Current molecular biology and biotechnology provide tools that will enable tomorrow’s protein engineers and other biochemists to assemble complex nanomachines not found in nature.

Following and promoting research and development that will lead to and speed the development of nanotechnology has been a core component of Foresight’s mission. From its inception in May of 2005, Foresight’s blog Nanodot has reported progress in these efforts. Check for example progress in protein engineering, DNA nanotechnology, structural DNA nanotechnology and RNA nanotechnology. Progress toward Drexler’s vision of nanotechnology was also regularly covered by Foresight’s quarterly Update newsletter published from June of 1987 through spring of 2007. Current research and progress toward these goals has been the theme of a series of Foresight Institute Conferences from 1989 through 2014. The contributions of individual researchers towards these goals have been recognized by the Foresight Institute Feynman Prizes from 1993 through 2015.

The next step Drexler proposes is “Second-Generation Nanotechnology” based on his observation that “over the centuries, we have learned to use our hands of flesh and bone to build machines of wood, ceramic, steel, and plastic. We will do likewise in the future. We will use protein machines to build nanomachines of tougher stuff than protein.” Progress over the past 30 years in protein engineering and DNA nanotechnology (see above) has already begun to implement Drexler’s proposal. Nanodot searches for organic chemistry, artificial molecular machines, and nanoscale bulk technologies will reveal progress towards second-generation nanotechnology.

In Drexler’s vision, second-generation nanotechnology leads to what Drexler originally termed “Universal Assemblers” that will be able to “bond atoms together in virtually any stable pattern, adding a few at a time to the surface of a workpiece until a complex structure is complete.” He notes that because assemblers will allow us to build virtually anything that the laws of nature permit, we will be able to build virtually anything that we can design—including more assemblers.

In pointing to biology as an existence proof that complex and highly functional systems can be assembled from parts built to atomic precision, Drexler expands upon ideas he published in 1981 in the Proc. Natl. Acad. Sci. USAProtein design as a pathway to molecular manufacturing“, available on the web site of Foresight’s sister organization, the Institute for Molecular Manufacturing, and including an afterword from Drexler written in 1988. To eliminate confusions that later arose from people conflating the properties of biological systems with the properties of the nanomechanical systems he was proposing, Drexler published in 1989 in the book Artificial Life and later made available on his web site a paper to clarify the differences “Biological and Nanomechanical Systems: Contrasts in Evolutionary Capacity“. The preface he added to this paper makes clear (1) that the proposal he described in Engines of Creation “of using small self-replicating systems as a basis for high-throughput atomically precise manufacturing” is obsolete, and (2) “that systems entirely unlike living cells can, by several engineering metrics, implement better ways to perform atomically precise fabrication”. In his 1992 technical book Nanosystems: Molecular Machinery, Manufacturing, and Computation and in more recent talks (for example), and posts, Drexler places much more emphasis on physical law as a basis for high-throughput atomically precise manufacturing.

In addition, the concept “of using small self-replicating systems as a basis for high-throughput atomically precise manufacturing” led to concerns about runaway replicators devouring the biosphere—the infamous “Gray Goo” scenario (see Engines of Destruction). Concerns about such scenarios were part of the reason that proposals for molecular manufacturing met with such resistance from some very influential members of the mainstream nanotechnology community—a community formed around a broader definition of nanotechnology than the one Drexler had introduced in 1986. Drexler discussed these issues extensively in an article published in 2004: “Nanotechnology: From Feynman to Funding” (Bulletin of Science, Technology & Society February 2004 24: 21-27. doi: 10.1177/0270467604263113). The full text is available on Drexler’s web site. Also during 2004, a paper by Chris Phoenix and Drexler argued that all the benefits of replicating assemblers without the dangers of runaway replicators could be obtained by using nanofactories for exponential manufacturing “Safe exponential manufacturing” (Nanotechnology 15: 869-872 (2004) doi:10.1088/0957-4484/15/8/001). The full text is available here.

After confronting and disposing of suggested reasons why the nanotechnology he described might not be possible—the uncertainty principle of quantum physics, the molecular vibrations of heat, radiation, the fact that evolution has failed to produce assemblers—Drexler wraps up his introduction of nanotechnology with descriptions of nanocomputers, disassemblers, and how the world will be “made new” by nanotechnology.

Among the most important systems that nanotechnology will extend to the limits allowed by physical law are computers. Because today’s computers are electronic, and the quantum behavior of electrons (as opposed to atoms) at nanometer scales is difficult to predict, Drexler points to the benefits of mechanical nanocomputers. They probably will not be as fast as electronic nanocomputers, but they may be smaller, and will certainly be much faster than the best electronic computers available in 1986 (or in 2015, for that matter). [Note: For a very recent study of the advantages of mechanical computers enabled via atomically precise manufacturing, see “Molecular Mechanical Computing Systems” by Ralph C. Merkle et al. on the IMM web site.]

Disassemblers are another very useful device that will be made possible by advanced nanotechnology. (The argument he makes here for disassemblers is not affected by the above arguments for why nanofactories are to be preferred over systems of replicating assemblers.) Disassemblers will be used by scientists and engineers to analyze things by taking them apart a few atoms at a time and recording the positions of all their constituent atoms. A nanocomputer system could then direct the assembly of perfect copies (by systems of assemblers or by nanofactories).

Because the first steps toward assemblers had already been taken by 1986 (and the many steps since then cited by the links above) and because each step will bring immediate rewards (which Foresight has also followed through Nanodot, Updates, and Conferences), Drexler argues that the advance toward assemblers is inevitable and will accelerate. To understand a future of changes he considers as profound as the industrial revolution, he turns to considering the principles of change that have survived past upheavals.

The Principles of Change

In search of tools to help understand the revolutionary changes he anticipates in the coming decades, Drexler looks to the principles of order, chaos, and evolution that have guided phenomena from the crystallization of molecules to the evolution of all life forms. And then he shows that the same principles of change apply to technologies, producing a global technology race in which developments that improve life or enhance wealth or military power will be unstoppable. Consideration of how ideas spread and evolve leads to the search for “mental immune systems” to reject ideas that are worthless or harmful in preparing for revolutionary new technologies. This core insight that both technology and ideas evolve under selective pressure provides the basis for Foresight’s mission. The eventual emergence of advanced nanotechnology seems certain.

Drexler begins by observing that order can emerge from chaos spontaneously. Molecules moving randomly in a liquid can settle into crystalline order as the liquid evaporates. Crystals grow by variation and selection as molecules that bump into the first tiny crystal stick if they bump in the proper position and orientation, and drift away if they do not. “Though the molecules bump at random, they do not stick at random. Order grows from chaos through variation and selection.”

Just as each layer of a crystal acts as a template for the next, a molecule of DNA or RNA can act as a template for a molecular copying machine to make a molecule of identical subunit sequence. Following Richard Dawkins, Drexler calls such molecules or other entities that give rise to copies of themselves, replicators. An occasional miscopy, however, can generate a product strand that differs in length or subunit sequence from the template. In the example he cites, different RNA molecules replicate at different rates depending on length and sequence. The amazing result of this particular experiment with a certain RNA-copying protein was that no matter what RNA sequence they started with, the end result ofter many rounds of copying the template, using the product as template for new rounds of copying, etc., is convergence of the final product to a well-defined sequence of 220 subunits. Prolonged copying, miscopying, and competition for resources to be copied in the next round (access to the copying protein, subunits, etc.) always brought the same result. However, changing the conditions of the experiment, such as adding a trace of an enzyme that destroys RNA molecules, yields a different result as the process adjusts to balancing competing effects, so a new best competitor emerges. “When varying replicators have varying successes, the more successful tend to accumulate. This process, wherever it occurs, is ‘evolution’.” Evolution is essentially the same process, whether it is molecules in a test tube, viruses infecting bacteria, the change of entire organisms over generations, or the biosphere as a whole.

Drexler notes that evolution accumulates patterns of success by eliminating unsuccessful changes. “Evolution proceeds by the variation and selection of replicators.” He describes how this process can be read in fossils found in different geological layers of stone, and in variations in the DNA and proteins of closely related and distantly related organisms. Comparison of genomes reveals evolutionary relationships in the same way that comparison of copying mistakes in ancient manuscripts reveals lineages from the most ancient to most recent copies.

From the evolution of life, Drexler turns to the evolution of technology. The design process for technology involves the generation of alternatives followed by testing them against requirements and constraints. Generation and testing of designs equates to variation and selection of organisms. Competition in markets and battlefields drives the evolution of ever more capable technologies. Because competition is global, local prohibition cannot block advances in commercial and military technology. Thus “… the force of technological evolution makes a mockery of anti-technology movements: democratic movements for local restraint can only restrain the world’s democracies, not the world as a whole.”

As technologies evolve, so too does the design of technologies, Drexler notes. Designs that work are accumulated through a process of variation and selection, increasingly aided by computer simulation using mathematical tools like finite element analysis, and design methods that work are also accumulated. Drexler quotes Alfred North Whitehead, “The greatest invention of the nineteenth century was the invention of the method of invention.”

Delving further into the evolution of technology, Drexler asks what are the replicators, the ‘genes’ of the machines? Designs and ideas behind designs replicate in the minds of engineers, scientists, and anyone who designs. Ideas evolve. Drexler quotes RIchard Dawkins in terming replicating mental patterns “memes”. Through imitation, memes propagate through the meme pool from one brain to another. Drexler observes that people both learn and teach, create the new and misunderstand the old, and don’t believe everything they hear. Memes compete for human attention and effort. They sometimes aid survival and reproductionof the host, but because they evolve only to survive and spread, other times they harm or kill the host, such as the meme for martyrdom-in-a-cause. Thus ideas can evolve to spread without having any connection to truth.

Drexler concludes that just as an immune system evolved to handle parasites, mental immune systems are essential to handle the great diversity of memes in which we live, especially in an era of rapidly evolving technologies in which the new and true and sound bizarre. “The deep-rooted principles of evolutionary change … guide the evolution not only of hardware, but of knowledge itself.

Predicting and Projecting

In explaining how to describe technologies that have not yet been developed, Drexler argues that known physical law sets the limits to what will be possible, and that competitive pressures will push development to those limits, although how, when, and at what cost cannot be predicted due to the existence of myriad uncertainties along the possible paths. Application of known physical law allows separation of wild, implausible-sounding ideas that are possible and ultimately likely, from wild, implausible-sounding ideas that are in fact nonsensical.

To see where the technology race leads, Drexler addresses three questions: what is possible? what is achievable? and what is desirable? His answers: natural law sets limits to what is possible, the principles of evolution limit to what can be achieved, and our differing views argue for a future of diversity and safety.

His discussion of the “Pitfalls of Prophecy” in answer to the question of how anyone can predict the future provides an insightful basis for understanding the conflict that arose in the early years of this century between the advocates of atomically precise manufacturing as the ultimate goal of nanotechnology, on the one hand, and on the other, the many nanotechnology researchers who focused on efforts to achieve greater control over the structure of matter to make many useful devices without having a general purpose high-throughput atomically precise manufacturing technology.

Drexler gives the example of Feynman proposing in 1959 that by building larger machines to build smaller machines, eventually after a number of steps, the ability to do chemical synthesis by putting down atoms where the chemist says to make any substance will be achieved. Such technology would presumably include the ability to make any DNA molecule as needed. Although Feynman could foresee the eventually ability to make DNA by linking atoms together, he could not predict the time or cost to do so. As Drexler notes, biochemists eventually achieved the ability to make DNA “without programmable nanomachines” by combining a variety of specific chemical and other tricks.

Within a decade after the publication of Engines of Creation and the founding of the Foresight Institute, laboratory advances in a number of fields brought increased capability to control the structure of matter at the nanometer scale. Many of these advances clearly led to near-term useful technologies without need for programmable nanomachines to do atomically precise manufacturing. At the same time, other developments caused many researchers doing such work to fear that association with visions of the eventual development of such nanomachines would endanger their funding, at the same time they doubted such visions were achievable. Thus a large part of Foresight’s activities during the early years of this century were devoted to (1) defending the feasibility of what we at that time termed “molecular manufacturing” or “molecular nanotechnology” and (2) advocating for a “balanced” US National Nanotechnology Initiative that would support both near-term incremental progress in nanotechnology and visionary research aimed at molecular manufacturing. For further information see:

Another key insight Drexler employs to discern the shape of emerging technologies is that deep differences exist between science and engineering in terms of basis, methods and aims. Although both science and engineering have evolved through rigorous testing of ideas and elimination of those proved incorrect, scientists are forever searching for more precise and universal theories, while engineers need only know that under particular circumstances particular objects will perform well enough to meet specific objectives. Thus engineering results can achieve a certainty that science can never obtain. Thus while it is impossible to predict future scientific knowledge, it is possible to predict future technology on the basis of known scientific knowledge. This core insight that engineering predictions are feasible provides the basis for Foresight’s mission. It makes sense to discuss and plan for advanced nanotechnology now, years before its emergence.

Because scientists are not interested in propositions they do not have the physical tools to test, and engineers are not interested in objects they do not have the tools to manufacture, no one is interested in future engineering developments that are firmly based on current scientific knowledge, but cannot be manufactured with current fabrication tools. Part of Foresight’s mission is to address this large gap in understanding our future by drawing attention to what advanced nanotechnology that we can begin to design today will enable once we learn how to manufacture with atomic precision. Drexler addressed this topic, which he labeled “Exploratory Engineering“, in a Foresight Background document originally published in 1988.

page 1 | page 2 | page 3 | page 4