• 40s: Cybernetics, the notion the brain did logic in circuits, feedback
• 50s: the computer, stored programs, Logic Theorist
• 60s: LISP, semantic nets, GOFAI
• 70s: SHRDLU, AM
• 80s: AI winter, expert systems, neural nets
• 90s: robots, machine learning
• 00s: DARPA grand challenge level of competence

The main point of this post is to answer any objections of the form: you’ve been working on this so long, why don’t you have it yet?  (Or perhaps, AI is the technology of the future and always will be. )

One key thing to note is that cybernetics was the original line of inquiry that was going to let us understand how the brain worked and allow us to build smart machines.  Many people assume that cybernetics failed since it more or less disappeared as a discipline.  But in fact it learned some very key and useful insights, forming the basis of control theory and neuroscience; but it fell apart due to personalities in its cadre (a veritable soap opera between Wiener and McCulloch and Pitts involving Wiener’s daughter) and political disfavor in the US involving Wiener’s antiauthoritarian stances.

So GOFAI was born with a built-in bias against some of the insights of cybernetics.  That has now been repaired; it was forced by the reintegration of control theory and the growing use of knowledge from neuroscience in the 90s, when AI robotics began to get serious.  There are reasons AI floundered in the 80s, and that’s one — another is a diversion from basic research to applications before it was really ready.

Another point that is rarely made is that AI, the small sub-discipline of CS, isn’t the real major part of the work in the 20th century that will have led to intelligent machines.  It’s the invention of the computer itself and all the work that’s been done to bring us the processing power we need to do the job, and the software to manage it and the complexity of human-comparable systems.  And nobody could reasonably claim that that effort has been standing still, or has come to nothing, or anything even vaguely similar.

An AI will be a hardware/software network and system so complex and powerful that it will make the entire ARPANET of the 70s look like a toy — and it will have to manage its own internals completely automatically.  I personally think that it will need the internal robustness that can only come from incorporating feedback and automatic resource management into the basic fabric of its computing platform.  But that’s the kind of thing that can easily be done in a decade, once someone decides to do it.  And it will be useful for a lot of other applications as well!

I’m about to start a series of posts on the topic of why I think AI is actually possible.  I realize that most of the readers here don’t probably need too much convincing on that subject, but you’d be surprised how many very smart people, many of them professors of computer science, are skeptical to some extent or another on that point.

To start off, though, I’m just soliciting comments on the subject to try and get some feel for where the readership is on the subject, and what are the issues anyone feels are important to the argument.

Start your comment off with an indication when you think we’ll have human-level AI, and go from there:

• in the next decade
• in the 20s
• 2030-2050
• 2050-2100
• thereafter
• never

Feynman anniversary event to be held at University of South Carolina. h/t Nanowerk

In February 1960, the Caltech magazine Engineering & Science published Feynman’s “Plenty of Room”, and it has been re-published ten times since then. This has become one of the best-known papers in the history of nanotechnology. The fiftieth anniversary of the initial publication of “Plenty of Room” presents us with an opportunity to reflect upon Richard Feynman’s legacy in nanotechnology. The University of South Carolina will convene a symposium to consider the talk, the man, and the field of nanotechnology during the past fifty years. The Symposium takes place at the University of South Carolina on Friday and Saturday, 12 and 13 February 2010. All full program (in PDF format) is available. Registration fee: $25; no charge for USC faculty, staff or students.

Note this USC is South Carolina, not Southern California

From Ars Technica: Nobel Intent, a thought-provoking article on what the prevalence of computational science portends for reproducibility in science:

Victoria Stodden is currently at Yale Law School, and she gave a short talk at the recent Science Online meeting in which she discussed the legal aspects of ensuring that the code behind computational tools is accessible enough for reproducibility. The obvious answer is some sort of Creative Commons or open source license, and Stodden is exploring the legal possibilities in that regard. But she makes a forceful argument that some form of code sharing will be essential.

“You need the code to see what was done,” she told Ars. “The myriad computational steps taken to achieve the results are essentially unguessable—parameter settings, function invocation sequences—so the standard for revealing it needs to be raised to that of when the science was, say, lab-based experiment.” This sort of openness is also in keeping with the scientific standards for sharing of more traditional materials and results. “It adheres to the scientific norm of transparency but also to the core practice of building on each other’s work in scientific research,” she said. But the same worries that apply to more traditional data sharing—researchers may have a competitor use that data to publish first—also apply here. In the slides from her talk, she notes that a survey she conducted of computational scientists indicates that many are concerned about attribution and the potential loss of publications in addition to legal issues. (The biggest worry is the effort involved to clean up and document existing code.)

Still, this sort of disclosure, as with other open source software, should provide a key benefit: more interested parties able to evaluate and improve the code. “Not only will we clearly publish better science, but redesigned and updated code bases will be valuable scientific contributions,” Stodden said. “Over time, we won’t stagnate forever on one set of published code.”

via Ars Technica: Nobel Intent: Keeping computers from ending science’s reproducibility.

Roadmaps to Nanotech and AGI

slides are here

Josh

[note -- we know about the permission problem, trying to get it fixed][should be fixed now]

Glenn Reynolds, a past Foresight Director, writes some analysis of the recent special election in Mass.:

Of course, what the GOP apparat does is less important nowadays than it was. As I noted before, there’s a whole lot of disintermediation going on here — Scott Brown got money and volunteers via the Internet and the Tea Party movement, to a much greater degree than he got them from the RNC. Smart candidates will realize that, too.

And lies don’t work as well as they used to. Obama promised transparency and pragmatic good government, but delivered closed-door meetings and outrageous special-interest payoffs. This made people angry. If Republicans promise honesty and less-intrusive government, but go back to their old ways, the likelihood that the Tea Party will become a full-fledged third party is much greater. …

via Instapundit » Blog Archive » SO, BROWN WON. This is big news; while the White House is still in the healthcare bunker, things li….

We don’t deal with politics here but we are concerned with technological developments that improve social decision-making and governance.  The internet has clearly been such a technology.  As one very tiny part of the generation of computer scientists that built it, I will happily accept the plaudits of a grateful world in their behalf …

Of course, the Internet could be improved as a fact-finding device, and ought to, as Eric Drexler notes:

We could benefit immensely from a medium that is as good at representing factual controversies as Wikipedia is at representing factual consensus.

What I mean by this is a social software system and community much like Wikipedia — perhaps an organic offshoot — that would operate to draw forth and present what is, roughly speaking, the best evidence on each side of a factual controversy. To function well would require a core community that shares many of the Wikipedia norms, but would invite advocates to present a far-from-neutral point of view. In an effective system of this sort, competitive pressures would drive competent advocates to participate, and incentives and constraints inherent in the dynamics and structure of the medium would drive advocates to pit their best arguments head-to-head and point-by-point against the other side’s best arguments. Ignoring or caricaturing opposing arguments simply wouldn’t work, and unsupported arguments would become more recognizable.

Success in such an innovation would provide a single place to look for the best arguments that support a point in a debate, and with these, the best counter-arguments — a single place where the absence of a good argument would be good reason to think that none exists.

Today is the last day of the free webcast of the 2010 Foresight Conference being held now in Palo Alto.

The bandwidth coming out of the Sheraton is marginal, so the video may be low-res, but we will be posting high-res videos later, funds permitting (feel free to assist with this goal!).

You can also follow the conference on Twitter at #Foresight2010, and send in your questions in real time to the speakers that way.

Wish you all could be here with us today!  —Chris Peterson

There’s still time to register, but if you just can’t participate in person this year, check out the free webcast of the Foresight Conference being held this weekend in Palo Alto.

The bandwidth coming out of the Sheraton is marginal, so the video will be low-res, but we will be posting high-res videos later, funds permitting (feel free to assist with this goal!).

Unfortunately the Senior Associate Reception debate between Robin Hanson and Mencius Moldbug on futarchy will not be webcast.

You can also follow the conference on Twitter at #Foresight2010, and try sending in your questions in real time to the speakers that way.

Wish you all could be here with us this weekend!  —Chris Peterson

Two nanoparticles connected by a polymer will tend to be drawn together at finite temperatures (though not at absolute zero) because as the polymer chain explores the states available to it, there are many more tangled and balled up ones than stretched-out straight ones — even though there is no overt force pulling the chain to any particular tangled state.  Such a situation is called an entropic spring, and behaviors like this are some of the more interesting aspects of physics at the microscale.

An arXiv paper by physicist Erik P. Verlinde purports to show that gravitational effects have the same mathematical logic behind them (in a very broad analogical sense), arising from the holographic universe description of physics (a far-out variant of string theory).  Now I don’t come close to having the physics to evaluate the theory, but Verlinde appears to be a respectable physicist.  Czech physicist Luboš Motl blogged about it:

So I remain undecided whether or not there is a sharp insight waiting along the lines of Verlinde’s paper.

and then allowed Verlinde to guest-post a long explanatory comment.

The derivation of the Einstein equations (and of Newton’s law in the earlier sections) follows very similar reasonings that exist in the literature, in particular Jacobson’s. The connection with entropy and thermodynamics is made also there. But in those previous works it is not clear WHY gravity has anything to do with entropy. No explanation for this apparent connection between gravity and entropy has been given anywhere in the literature. I mean not the precise details, even the reason why there should be such a connection in the first place was not understood.

My paper is the first that gives a reason why. Inertia, and hence motion, is due to an entropic force when space is emergent. This is new, and the essential point. This means one HAS TO keep track of the amount of information. Differences in this amount of information is precisely what makes one frame an inertial frame, and another a non-inertial frame. Information causes motion.
This can be derived without assuming Newtonian mechanics.

“Space is emergent”???  Yep, in the holographic theory, 3D space is an emergent phenomenon of a 2D information pattern (see the link above).  Weird stuff, but no weirder than other forms of string theory.

As mentioned, I don’t claim to follow this at the technical level, but given how important the math of entropy is at the microscale, it’s fun to speculate about its being important at the most macro of macroscales as well.

A round-up of commentary about the state of nanotech research, given the 50th anniversary of Feynman’s talk:

Adam Keiper in the WSJ

If this dispute over nano-nomenclature only involved some sniping scientists and a few historians watching over a tiny corner of Feynman’s legacy, it would be of little consequence. But hundreds of companies and universities are teeming with nanotech researchers, and the U.S. government has been pouring billions of dollars into its multiagency National Nanotechnology Initiative.
So far, none of that federal R&D funding has gone toward the kind of nanotechnology that Drexler proposed, not even toward the basic exploratory experiments that the National Research Council called for in 2006. If Drexler’s revolutionary vision of nanotechnology is feasible, we should pursue it for its potential for good, while mindful of the dangers it may pose to human nature and society. And if Drexler’s ideas are fundamentally flawed, we should find out—and establish just how much room there is at the bottom after all.

Eric Drexler on Keiper and on the NRC report

The evaluation of the feasibility of molecular manufacturing and recommendations for research form the concluding section of the body of the NRC’s Triennial Review of the National Nanotechnology Initiative. In the three years since the publication of the NRC report, I have seen no document from a Federal-level source that acknowledges these conclusions, and, of course, none that offers a substantive response.
This is of concern, because the NRC report calls for a sharp break with past thinking. To put it bluntly, much of the opinion in general circulation about molecular manufacturing (both pro and con) is rubbish because it is based on mythology, and not on the scientific literature. The NRC report can be criticized on several points, but it isn’t rubbish.

Dexter Johnson on Keiper and Drexler

I am nonplussed. Are we to believe that Prof. Moriarty is one of only a handful of scientists capable of securing funding for his experiments into molecular nanotechnology?

Richard Jones on the same

Of the ideas dealt with in “Plenty of Room”, some have already come to pass and have indeed proved economically and societally transformative. These include the idea of writing on very small scales, which underlies modern IT, and the idea of making layered materials with precisely controlled layer thicknesses on the atomic scale, which was realised in techniques like molecular beam epitaxy and CVD, whose results you see every time you use a white light emitting diode or a solid state laser of the kind your DVD contains. I think there were two ideas in the lecture that did contribute to the vision popularized by Drexler – the idea of “a billion tiny factories, models of each other, which are manufacturing simultaneously, drilling holes, stamping parts, and so on”, and, linked to this, the idea of doing chemical synthesis by physical processes.

Jones ends with a observation about the course of nanotech development:

Perhaps for the first time in recent years a major new technology is largely being developed outside the USA, in Europe to some extent, but with an unprecedented leading role being taken in places like China, Korea and Japan. In these places the “nanotech schism” that seems so important in the USA simply isn’t relevant; people are just pressing on to where the technology leads them.

This is a key observation.  Jones slants it as if to say that therefore, the “schism” wasn’t really important after all.  But to come away with that impression would be to miss a very important point:  The USA is blowing its opportunity to be a leader in one of the most important technologies of the 21st century because of the political shenanigans.

(or a little physics about climate change. Or at least a few clarifications about some of the points being raised.)

In the wake of Climategate, a wide variety of mistakes and misapprehensions are being circulated on the Internet (as if that weren’t happening before).

For example, in this article from the Telegraph:

Phil Jones, the 57-year-old director of the CRU, is the man who has suddenly found himself the number one target of climate change conspiracy theorists the world over after he sent the most damaging of all the emails exposed by the anonymous hacker.
In one message, dated November 1999, he wrote: “I’ve just completed Mike’s trick of adding in the real temps to each series for the last 20 years (ie from 1981 onwards) and from 1961 to hide the decline.”
Gotcha! say the global warming sceptics who have argued for years that average temperatures on Earth are, in reality, either stable or going down. Professor Jones defended himself by claiming the word “trick” was used out of context and simply referred to a legitimate method of handling data.

Well, actually not. “Hide the decline” had nothing whatsoever to do with the flattening of GMST (global mean surface temperature) post-2000, since that was written in 1999. It had to do with what is technically called the “divergence problem”, which is that the tree-ring data used in paleoclimate studies (such as the “hockey stick”) don’t match real, measured, temperatures in recent decades. The reason Jones would want to hide that is that if tree rings don’t register the current hot spell, they might have missed past hot spells as well, and thus couldn’t be used as proof that the current warming was unprecedented.

There are a bunch of other misapprehensions floating around, on both sides of the question. Here’s a grab bag.

• The greenhouse effect doesn’t exist, or is negligible because CO2 is a trace gas, etc. (Found on some of the more argumentative “denier” sites). Try drinking a glass of water with 388ppm botulin toxin and then tell me about trace concentrations. The greenhouse effect is pretty well understood physics.
• There’s some “tipping point” inherent in the greenhouse effect (found on the more argumentative alarmist sites). Nope, the well-understood greenhouse effect gives a logarithmic increase in temperature with added CO2. That is, if increasing CO2 from 194 to 388 increased temperatures one degree, increasing it from 388 to 776 would get us one more degree, and from 776 to 1552 would produce just one more degree after that. This is why the discussion is typically framed in terms of “a doubling of CO2.”
• The earth isn’t really warming, or the warming has stopped. Nope, there really was a “Little Ice Age” and temperatures have been rising for as long as we’ve measured them. If you want to look at data and do a little of your own analysis, Paul Clark’s “Wood for Trees” site is marvelous. It gives access to a number of datasets and an easy-to-use web-based analysis interface. Here it is doing a linear regresion on the longest measured record, HADCRUT. There’s a secular (long-term) rise of about 0.44 degrees per century. On the other hand, it’s clear that there’s a roughly 60-year cycle about that trend and a lot of the rise from the 70’s thru the 90’s was cyclic rather than secular. There are almost certainly lower-frequency cycles as well.
• Data adjustment by the climatologists has hugely skewed the record. Here’s the HADCRUT record plotted against the satellite record, which doesn’t have those problems. You don’t own a thermometer that could tell the difference. Not that there wasn’t some hanky-panky, but the difference is probably due as much to the different proportions of land vs sea measurements than anything else.
• Computer models of climate are worthless. No, they do reasonably well in the regimes they’ve been tuned for. They tracked the rise in GMST up through 1998, which is one reason they developed so much credibility over that period.
• Computer models are valid forecasters on century-long scales. The arguments for this are as follows: Yes, we know that the actual weather is chaotic and the models can’t track it, but climate can be thought of as the basin of attraction that the weather moves in, and if the model captures the basin of attraction properly, the weather in the model will show us what the overall climate will be like. But just one decade after the GMST inflection point of 1998, the models are pushing the envelope of 95% confidence intervals:
  
  (btw, Lucia is oneof the best mathematically sophisticated, balanced bloggers out there (recognized as a “lukewarmer”, i.e. someone between the extremes of the debate, by both sides).
  It’s debatable whether the models fail statistical confidence tests as of even date. But the fact that they are so close to the edge so soon doesn’t inspire confidence that they have a basin of attraction that matches the one of the real Earth for a full century. Specifically, it seems likely that they have too high a parameter for cloud-based feedback and thus overestimate the temperature sensitivity as a function of CO2.
  This is probably the one case where the leaked emails should actually change anyone’s perception of the science; Kevin Trenberth, in 1255523796.txt:

  How come you do not agree with a statement that says we are no where close to knowing where
  energy is going or whether clouds are changing to make the planet brighter. We are not
  close to balancing the energy budget.

  I have some personal experience trying to balance energy budgets in massive semi-empirical simulation codes, and I have lots of sympathy for the modellers. But my experience also tells me that the models are very unlikely to have been validated and tuned well enough to make reliable century-scale predictions.

So what does this all mean, bottom line? Not an awful lot — but just take all those claims you hear, on both sides, with a grain of salt.

Raw data are important in validating scientific work. Even so simple an operation as smoothing by time-averaging can have counter-intuitive effects, such as Simpson’s Paradox:

For a simple and homey example, here are the batting averages of Derek Jeter and David Justice in 1995, 1996, and 1997:

in 1995, Jeter had 12 hits in 48 at-bats, for an average of .250. Justice beat him with 104 hits in 411 times up for .253.

in 1996, Jeter hit 183 times in 582 tries for .314; Justice hit 45 out of 140 for .321, winning again.

in 1997, Jeter was 190 for 654, averaging .291. Justice got 163 hits in 495 tries for an whopping .329 average.

In other words, each year Justice out-hit Jeter according to standard stats.

But: for the three years as a whole, Jeter hit 385 out of 1284 for an overall average of .300, whereas Justice’s numbers were 312 of 1046, for an average of only .298. I.e. for the three years combined, Jeter had a higher average than Justice.

Counter-intuitive? Yes. Should scientists be allowed to get away with releasing only averaged, interpreted, adjusted, or otherwise massaged data? No.

 

Just for fun:

(h/t Roger Pielke, Jr.)

 

(http://www.youtube.com/watch?v=-VRBWLpYCPY)

(h/t Megan McArdle)

 

“Science advances, funeral by funeral.” (often attributed to Timothy Ferris)

The blogosphere has been abuzz over the past week or so with the release of data — emails and program source and documentation — from the Climatic Research Unit at the University of East Anglia, one of the premier climatology research institutions in the world. Those with an interest have doubtless read much more about it than you will read here, but the nutshell version is that skeptics maintain that the emails show that the climatologists have been falsifying data and running a scam, and the mainstream claims that this is just the way science is always done, scientists are after all human. Robin Hanson points out that “If you knew how academia worked, this news would not surprise you nor change your opinions on global warming.”

I think that Robin is right that no one should be surprised. But I have a couple of observations about the nature of science and scientists:

First, scientists don’t really go out into the world with a blank mind and allow the data to suggest new laws of nature to them. Scientists — the ones who make significant breakthroughs anyway — go out with a paradigm firmly fixed in their minds and look for data to prove it. The paradigm is produced by subconscious processes of scientific intuition.

The classic case where this was made public because a top scientist was honest with himself in his notebooks, and these were subsequently published, was Millikan and the oil drop experiments to determine the charge of the electron.

It is plausible to suggest that Ehrenhaft’s methodology approximated the traditional scientific method, which did not allow him to discard anomalous data. Millikan, on the other hand, in his publications espoused the scientific method but in private (handwritten notebooks) was fully aware of the dilemma faced and was forced to select data to uphold his presuppositions.

A model of why this is the way things really work is the satisfiablity problem in computer science. The problem is to find values for variables that give a true value to a logical expression. If you can guess the right values, it’s simple to evaluate the expression and show they’re right. Otherwise, you are left with an intractable search. Similarly in science, guessing the right answer and then proving it is usually the way things really work.

In the Climategate files, it’s clear that the climatologists have a predetermined paradigm and are trying to prove it. The skeptics claim that’s evidence of a scam, but I think that by and large, climatologists really believe the paradigm they’re pushing, and they’re “selecting data to uphold their presuppositions.”

On the other hand, because science really does work this way, it is even more important than would be the case in the naive view, that there be a strong devil’s advocate function and that replication be required before any results are accepted. Here I think the climatologists are on shakier ground.

Science is at its base a way of convincing people that something you believe is true. It says, if you don’t believe me, try it yourself. Do the experiment, do the math. Any reasonable person will have to come to the same conclusion I did. There are other modes of persuasion: the religious (if you don’t believe me, you’ll go to Hell), the political (if you don’t believe me, you’ll be fired/jailed/shot), and the social (everybody else believes this, if you don’t you’re a kook). The other modes are much more deep-seated in the human psyche and held sway for much longer than the relatively few centuries of modern science. Thus when climatologists act all too human, they run the risk of slipping into other modes of persuasion and losing the claim to be doing science.

We didn’t need the Climategate papers to know that the climatologists were refusing to publish their data and their models. This turns the idea of science on its head. Instead of “here’s what I did, try it yourself” we have “not only is my data secret, my theory (the computer model) is secret.” This does not inspire confidence.

Science convinces by letting the critics have all the advantages and take their best shots. If a theory is true, it will still stand. If a theory is protected instead by the other modes of persuasion, it’s suspect.

There is, in the age of the internet, absolutely no reason that science can’t be open source. Let, for example, tree-ring or temperature-station data be placed on the web — the real, raw data, not after someone adjusts and interprets it. Let all the codes representing theories be open source. After all, most of this stuff is paid for with tax money in the first place. Let anyone read it, interpret it, make new or variant theories in the form of statistical or simulation codes. Ideas, like iron, are best formed when they are subjected to lots of heat and beaten on.

Climategate probably won’t have a lot of effect on climate science per se, but if it moves it, and science in general, even a little bit in a more open-source direction, that’ll be something to be thankful for this week.

 

 

From the protein crystallography beamline at Berkeley Labs Advanced Light Source: an action shapshot the Rho transcription factor from E. coli.

The orange spiral in the middle is a strand of RNA, and Rho is everything else.

 

(http://www.youtube.com/watch?v=UPQ0OnlfkkA)

(h/t Technology Review blog)

Peptides control crystal growth with switches, throttles and brakes. From Physorg.com.

 

 

(PhysOrg.com) — By producing some of the highest resolution images of peptides attaching to mineral surfaces, scientists have a deeper understanding how biomolecules manipulate the growth crystals.
The research, which appears in the Nov. 23 online edition of the journal Proceedings of the National Academy of Sciences, explores how peptides interact with mineral surfaces by accelerating, switching and inhibiting their growth.

The direct application of the research is to understand the regulatory capabilities of biomolecules on the growth of minerals (e.g. teeth, kidney stones) in the body.  But by understanding the mechanisms at the atomic scale, we might at some point be able to design proteins that allow us to grow surfaces with designed patterns, or even atomically precise parts.

 

This year is the 20th anniversary of the original Foresight Conference on Nanotechnology.

The neat, clear vision of nanotechnology we had in 1989 rested on two key aspects that would make it a transformative, rather than merely an evolutionary, technology:

• The ability to construct and observe at the atomic scale, and the construction of machines at that scale, taking advantage of various phenomena
• These machines could be production machinery for more machines, shortening capital formation times and increasing economic growth rates

The reality of nanotechnology is shaping up differently from the neat visions of those times, but shaping up it is. There is substantial coverage of the first point today: the techniques for manipulating and observing at the molecular scale are well advanced over 1989. There are things that are arguably machines as well: by some definitions, the last two generations of computer processors have been flat-out nanotechnology. On the atomically precise front, which is closer to what we think really makes a difference as far as nanotechnology is concerned, an increasing proportion of work involves nanostructures with electronic or catalytic properties that perform useful functions.
On the second point there remains an odd dichotomy. Researchers working from the direction of biosystems understand and use the autogenous properties of biomolecular components (e.g. polymerases) and use them as a matter of course. Those coming from the chemistry/surface physics direction, however, don’t seem to have picked up on it, or at least haven’t managed to make the right tools yet.
The bottom line is that 20 years on, the world has picked up strongly on one of the main legs of the nanotech vision, working at atomic scale and precision. The other one, autogenous systems, has been sorely neglected.
In some sense, the two legs of the nanotech vision are the same two properties of life that make living things so different from non-living ones: they have mechanism that is atomically precise and works on that scale; and they reproduce themselves. Besides life, autogenous systems in the real world range from the simple physical models of machine shops that make parts for shop machines, to the memetic ecosystem of ideas that is science itself. Questions that seem like mere technical details, such as growth rates and feedstock closure, turn out to be crucial in understanding major effects ranging from the possibility of gray goo to the prospect of economic displacement. A better understanding of autogeny in software is likely to give us more robust systems and ultimately, true artificial intelligence, since a learning mind is clearly autogenous.
Foresight was a thought leader in 1989 because we had a vision that allowed us to see future possibilities, opportunities and dangers alike, in ways that were not generally apprehended. That is still true. The world at large has picked up on the “atomic scale” leg of the vision, but hasn’t understood the importance of the autogenous systems one.
The first Foresight Conference was notable, among other things, because it was extremely interdisciplinary. Working at the atomic scale involved pulling together knowledge from many branches of physics, chemistry, biology, and other physical sciences. Leading the way in the unfinished business of autogeny will likewise involve pulling together knowledge from a wide variety of fields, ranging from biology and evolution to computer science (consider von Neumann’s classic study of self-reproducing automata) to economics.
For the 20th anniversary of that first groundbreaking conference, Foresight is organizing a new conference to concentrate on the principles, techniques, and impacts (social and economic) of autogenous systems, from nanofactories to self-improving AIs.

Foresight 2010: the Synergy of Molecular Manufacturing and AGI

Join us in for an exciting conference focused on the Synergy of Molecular Manufacturing and general Artificial Intelligence and celebrate the 20th anniversary of the founding of Foresight. The two day conference rate is $175 with discounts for early registration.
Several rapidly-developing technologies have the potential to undergo an exponential takeoff in the next few decades, causing as much of an impact on economy and society as the computer and networking did in the past few. Chief among these are molecular manufacturing and artificial general intelligence (AGI). Key in the takeoff phenomenon will be the establishment of strong positive feedback loops within and between the technologies. Positive feedback loops leading to exponential growth are nothing new to economic systems. At issue is the value of the exponent: since the Industrial Revolution, economies have expanded at rates of up to 7% per year; however, computing capability has been expanding at rates up to 70% per year, in accordance with Moore’s Law. If manufacturing and intellectual work shifted into this mode, the impact on the economy and society would be profound. The purpose of this symposium is to examine the mechanisms by which this might happen, and its likely effects.

We are trying to repeat the format of that first Foresight congerence, with a selection of invited speakers who will weave together an interdisciplinary overview of the state of the art(s) and likely or possible pathways to the future.

Conference page

Registration

[update: fixed conference page link!]

In Popular Mechanics, longtime Foresight friend Prof. Glenn Reynolds looks at the future of nanotech and artificial intelligence, among other things looking at safety issues, including one call that potentially dangerous technologies be relinquished.  He takes a counterintuitive stance, which we’ve discussed here at Foresight over the years:

But I wonder if that’s such a good idea. Destructive technologies generally seem to come along sooner than constructive ones—we got war rockets before missile interceptors, and biological warfare before antibiotics. This suggests that there will be a window of vulnerability between the time when we develop technologies that can do dangerous things, and the time when we can protect against those dangers. The slower we move, the longer that window may remain open, leaving more time for the evil, the unscrupulous or the careless to wreak havoc. My conclusion? Faster, please.

OK, it’s counterintuitive, but it may be right.  —Chris Peterson

Over at Nanoclast, Dexter Johnson writes:

It seems when nanotech is applied to photovoltaics it can either boost their efficiency to new heights or it can cheapen their manufacturing process. But it never seems to provide a solution to both of these. It’s always a tradeoff: increased efficiency but difficult manufacturing processes or a cheaper production process but less efficiency.

The solution to this, of course, is that the efforts in nanotech research should be going toward developing atomically-precise machinery that can do the manufacturing.  Like any form of research and capital formation – vs – consumption question, there is a balance between this and direct application-oriented work, but the more spent on the former, the better in the long run. And the use of the word “nanotechnology” to characterize the latter has confused the issue.

"Nanotechnology" hunting arrows

As I wrote in Nanofuture:

… the stuff that’s going on in most labs today under the name of nanotechnology may make smaller computer chips, or stronger aerospace materials, or whatever, but it’s really more of the same old conventional technology by another name. You don’t need to read a whole new book to learn that people are trying to make more stain-resistant (and expensive) pants, or stronger (and more expensive) tennis racquets, or smaller, faster computers. Nor do you need to worry over the fact that marketing departments will be calling these things, and lots of other things over the coming years, “nanotechnology”–it’s just a word.

… So “nanotechnology” really does have two different meanings. One is the broad, “stretched” version meaning any technology dealing with something less than 100 nanometers in size. The other is the original meaning: designing and building machines in which every atom and chemical bond is specified precisely. I’ll refer to the former as “nanoscale technology” when I need to; but I won’t refer to it much. The capabilities and dangers of nanoscale technology are simple and straightforward extensions of current trends in the capabilities and dangers of chemistry, materials science, and microfabrication. The majority of new techniques being discovered and trumpeted as the latest thing in “nanotechnology” today will be obsolete in ten years.

… Nanotechnology has the potential for increasing our physical capabilities more than did the industrial revolution; expanding our ability to learn and communicate more did than the printing press; accelerating our ability to travel more than did the boat or the wheel; and enlarging the range of places we can live more than clothing did. It could induce greater biological changes in the human organism than the difference between humans and chimpanzees; indeed, greater than the difference between humans and horseshoe crabs. It is coming, possibly in the next decade, probably in the next two-and-a-half, almost certainly in the twenty-first century.

Meanwhile, solar power continues to fall in a Moore’s Law – like fashion; but it won’t really be mature until we get real nanotechnology.

Gallery – A joyride through the nanoscale – Image 1 – New Scientist.

This New Scientist article has some nice images from Whitesides recent book, sort of a retake on the “Secret House” idea.